05 CSSBB Primer

© QUALITY COUNCIL OF INDIANA INTRO-1 (1)
CSSBB 2014
THE
SIX SIGMA
BLACK BELT
PRIMER
© by Quality Council of Indiana - All rights reserved
Fourth Edition - September, 2014
Quality Council of Indiana

602 West Paris Avenue
West Terre Haute, IN 47885
TEL: 800-660-4215
FAX: 812-533-4216
qci@qualitycouncil.com
http://www.qualitycouncil.com
011
CSSBB 2014
CSSBB Primer Contents

I. CERTIFICATION OVERVIEW . . . . . . . . . . . . . . . . . I-1
CSSBB EXAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I-3
CSSBB BODY OF KNOWLEDGE . . . . . . . . . . . . . . I-6
II. ENTERPRISE-WIDE DEPLOYMENT . . . . . . . . . . . II-1

ORGANIZATION-WIDE CONSIDERATIONS . . . . . II-2
SIX SIGMA/LEAN FUNDAMENTALS . . . . . . . . . II-2
IMPROVEMENT METHODOLOGIES . . . . . . . . II-34
SYSTEMS AND PROCESSES . . . . . . . . . . . . . . II-42
STRATEGIC PLANNING . . . . . . . . . . . . . . . . . . II-47
LEADERSHIP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II-57
ROLES AND RESPONSIBILITIES . . . . . . . . . . II-57
ORGANIZATIONAL ROADBLOCKS . . . . . . . . . II-63
III. PROCESS MANAGEMENT . . . . . . . . . . . . . . . . . III-1

OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III-2
STAKEHOLDER IMPACT . . . . . . . . . . . . . . . . . . . III-8
CRITICAL REQUIREMENTS . . . . . . . . . . . . . . . . III-11
BENCHMARKING . . . . . . . . . . . . . . . . . . . . . . . . III-11
BUSINESS MEASURES . . . . . . . . . . . . . . . . . . . III-15
PERFORMANCE MEASURES . . . . . . . . . . . . III-15
FINANCIAL MEASURES . . . . . . . . . . . . . . . . . III-20
CSSBB 2014
IV. TEAM MANAGEMENT . . . . . . . . . . . . . . . . . . . . . IV-1

TEAM FORMATION . . . . . . . . . . . . . . . . . . . . . . . . IV-2
TEAM TYPES . . . . . . . . . . . . . . . . . . . . . . . . . . . IV-2
ROLES & RESPONSIBILITIES . . . . . . . . . . . . . IV-7
SUCCESS FACTORS . . . . . . . . . . . . . . . . . . . IV-14
TEAM FACILITATION . . . . . . . . . . . . . . . . . . . . . IV-19
MOTIVATIONAL TECHNIQUES . . . . . . . . . . . IV-19
STAGES OF DEVELOPMENT . . . . . . . . . . . . . IV-27
LEADERSHIP MODELS . . . . . . . . . . . . . . . . . . IV-36
TEAM DYNAMICS . . . . . . . . . . . . . . . . . . . . . . . . IV-39
MEETING MANAGEMENT . . . . . . . . . . . . . . . . IV-46
DECISION MAKING METHODS . . . . . . . . . . . IV-52
TEAM TRAINING . . . . . . . . . . . . . . . . . . . . . . . . . IV-62
NEEDS ASSESSMENT . . . . . . . . . . . . . . . . . . IV-62
V. DEFINE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V-1
VOICE OF THE CUSTOMER . . . . . . . . . . . . . . . . . . V-2
CUSTOMER IDENTIFICATION . . . . . . . . . . . . . . V-2
CUSTOMER DATA COLLECTION . . . . . . . . . . V-10
CUSTOMER REQUIREMENTS . . . . . . . . . . . . . V-18
PROJECT CHARTER . . . . . . . . . . . . . . . . . . . . . . V-36
BUSINESS CASE . . . . . . . . . . . . . . . . . . . . . . . . V-37
PROBLEM STATEMENT . . . . . . . . . . . . . . . . . . V-38
GOALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V-43
PERFORMANCE MEASURES . . . . . . . . . . . . . V-45
REVIEWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V-48
PROJECT MGMT TOOLS . . . . . . . . . . . . . . . . . . . V-50
WORK BREAKDOWN STRUCTURE . . . . . . . . V-51
GANTT CHARTS . . . . . . . . . . . . . . . . . . . . . . . . V-60
PROJECT DOCUMENTATION . . . . . . . . . . . . . V-62
RACI MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . V-64
ANALYTICAL TOOLS . . . . . . . . . . . . . . . . . . . . . . V-65
CSSBB 2014
VI. MEASURE - DATA . . . . . . . . . . . . . . . . . . . . . . . . VI-1

PROCESS CHARACTERISTICS . . . . . . . . . . . . . . VI-2
FLOW METRICS . . . . . . . . . . . . . . . . . . . . . . . . VI-2
ANALYSIS TOOLS . . . . . . . . . . . . . . . . . . . . . . VI-17
DATA COLLECTION . . . . . . . . . . . . . . . . . . . . . . VI-36
TYPES OF DATA . . . . . . . . . . . . . . . . . . . . . . . VI-36
MEASUREMENT SCALES . . . . . . . . . . . . . . . VI-39
SAMPLING . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI-41
COLLECTION METHODS . . . . . . . . . . . . . . . . VI-43
MEASUREMENT SYSTEMS . . . . . . . . . . . . . . . . VI-48
SYSTEM ANALYSIS . . . . . . . . . . . . . . . . . . . . VI-48
ENTERPRISE MEASUREMENT . . . . . . . . . . . VI-57
METROLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . VI-59
VII. MEASURE - STATISTICS . . . . . . . . . . . . . . . . . VII-1

BASIC STATISTICS . . . . . . . . . . . . . . . . . . . . . . . . VII-2
CENTRAL LIMIT THEOREM . . . . . . . . . . . . . . . VII-3
DESCRIPTIVE STATISTICS . . . . . . . . . . . . . . . VII-5
GRAPHICAL METHODS . . . . . . . . . . . . . . . . . VII-14
STATISTICAL CONCLUSIONS . . . . . . . . . . . . VII-22
PROBABILITY . . . . . . . . . . . . . . . . . . . . . . . . . . . VII-25
COMMON DISTRIBUTIONS . . . . . . . . . . . . . . VII-31
OTHER DISTRIBUTIONS . . . . . . . . . . . . . . . . . VII-42
PROCESS CAPABILITY . . . . . . . . . . . . . . . . . . . VII-52
CAPABILITY STUDIES . . . . . . . . . . . . . . . . . . VII-52
CAPABILITY INDICES . . . . . . . . . . . . . . . . . . . VII-62
NON-NORMAL DATA . . . . . . . . . . . . . . . . . . . VII-67
PERFORMANCE METRICS . . . . . . . . . . . . . . VIII-73
CSSBB 2014
VIII. ANALYZE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII-1

MEASURING/MODELING RELATIONSHIPS . . . VIII-2
REGRESSION . . . . . . . . . . . . . . . . . . . . . . . . . VIII-2
MULTIVARIATE TOOLS . . . . . . . . . . . . . . . . VIII-16
HYPOTHESIS TESTING . . . . . . . . . . . . . . . . . . VIII-28
POINT & INTERVAL ESTIMATES . . . . . . . . . VIII-36
MAJOR TESTS . . . . . . . . . . . . . . . . . . . . . . . . VIII-41
ANOVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII-61
NONPARAMETRIC TESTS . . . . . . . . . . . . . . VIII-81
FMEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII-92
ADDITIONAL ANALYSIS METHODS . . . . . . . . VIII-99
GAP ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . VIII-99
ROOT CAUSE ANALYSIS . . . . . . . . . . . . . . VIII-107
WASTE ANALYSIS . . . . . . . . . . . . . . . . . . . VIII-119
IX. IMPROVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX-1

DESIGN OF EXPERIMENTS . . . . . . . . . . . . . . . . . IX-2
TERMINOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . IX-4
DESIGN PRINCIPLES . . . . . . . . . . . . . . . . . . . IX-11
ONE-FACTOR EXPERIMENTS . . . . . . . . . . . . IX-18
FRACTIONAL FACTORIAL EXPERIMENTS . IX-21
FULL FACTORIAL EXPERIMENTS . . . . . . . . IX-31
LEAN METHODS . . . . . . . . . . . . . . . . . . . . . . . . . IX-38
WASTE ELIMINATION . . . . . . . . . . . . . . . . . . . IX-38
CYCLE TIME REDUCTION . . . . . . . . . . . . . . . IX-50
KAIZEN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX-63
THEORY OF CONSTRAINTS . . . . . . . . . . . . . IX-65
IMPLEMENTATION . . . . . . . . . . . . . . . . . . . . . . . IX-70
CSSBB 2014
X. CONTROL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X-1
STATISTICAL PROCESS CONTROL . . . . . . . . . . . X-2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . X-2
SELECTION OF VARIABLES . . . . . . . . . . . . . . . X-4
RATIONAL SUBGROUPING . . . . . . . . . . . . . . . . X-5
CONTROL CHART SELECTION . . . . . . . . . . . . . X-8
CONTROL CHART ANALYSIS . . . . . . . . . . . . . X-32
OTHER CONTROLS . . . . . . . . . . . . . . . . . . . . . . . X-41
TPM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . X-41
VISUAL CONTROLS . . . . . . . . . . . . . . . . . . . . . X-48
MAINTAIN CONTROLS . . . . . . . . . . . . . . . . . . . . . X-50
MEASUREMENT SYSTEMS RE-ANALYSIS . . X-50
CONTROL PLAN . . . . . . . . . . . . . . . . . . . . . . . . X-52
SUSTAIN IMPROVEMENTS . . . . . . . . . . . . . . . . . X-58
LESSONS LEARNED . . . . . . . . . . . . . . . . . . . . . X-58
TRAINING PLAN DEPLOYMENT . . . . . . . . . . . X-60
DOCUMENTATION . . . . . . . . . . . . . . . . . . . . . . X-62
ONGOING EVALUATION . . . . . . . . . . . . . . . . . X-64
XI. DESIGN FOR SIX SIGMA (DFSS) . . . . . . . . . . . . XI-1

DFSS METHODOLOGIES . . . . . . . . . . . . . . . . . . . XI-2
DESIGN FOR X (DFX) . . . . . . . . . . . . . . . . . . . . . XI-10
ROBUST DESIGN . . . . . . . . . . . . . . . . . . . . . . . . XI-15
XII. APPENDIX/TABLES . . . . . . . . . . . . . . . . . . . . . XII-1

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XII-13
ANSWERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XII-29
CSSBB 2014
CSSBB Primer Question Contents

%
Primer Section CSSBB Exam Primer CD
II. Organization-wide 8.0% 12 32 80

Deployment
III. Process Management 6.7% 10 27 67
IV. Team Management 12.0% 18 48 120
V. Define 13.3% 20 53 133
VI. Measure - Data 7.3% 11 29 73
VII. Measure - Statistics 9.3% 14 37 93
VIII. Analyze 14.7% 22 59 147
IX. Improve 14.0% 21 56 140
X. Control 10.0% 15 40 100
XI. Design for Six Sigma 4.7% 7 19 47
Totals 100% 150 400 1000

© QUALITY COUNCIL OF INDIANA I-1 (8)
CSSBB 2014
I. CERTIFICATION OVERVIEW
Professionalizing Quality Education
I KNOW OF NO MORE
ENCOURAGING FACT THAN THE
UNQUESTIONABLE ABILITY OF
MAN TO ELEVATE HIS LIFE BY A
CONSCIOUS ENDEAVOR.
HENRY DAVID THOREAU
CSSBB 2014
Preface
All test questions and answers (on blue paper) must be
removed when using this text as a reference during a
certification examination.
CSSBB 2014
CERTIFIED SIX SIGMA BLACK BELT
CSSBB Exam
Objective
To provide recognized six sigma fundamental training

and to prepare persons interested in taking the
CSSBB examination.
The CSSBB
Is a professional who can carry out, in a responsible

manner, proven techniques which make up the body
of knowledge recognized by those who are experts in
six sigma.
Eligibility
CSSBB participants must register with ASQ

headquarters. Eligibility requires two completed
projects with signed affidavits or one completed
project with a signed affidavit and three years work
experience within the six sigma body of knowledge.
No education waiver is given.
CSSBB 2014
CERTIFIED SIX SIGMA BLACK BELT
Duration
A written test lasts 4 hours. Prometric exams last 4.5

hours. The exam is open book and consists of
multiple choice questions.
Other Details
Additional information can be obtained by calling ASQ

headquarters at (800) 248-1946 or http://www.asq.org
Bibliography Sources
The CSSBB student should obtain the bibliography

furnished by ASQ. Examples:
Breyfogle, III, F.W. (2003). Implementing Six Sigma:

Smarter Solutions Using Statistical Methods.
Harry, Mikel & Schroeder, R. (2000). Six Sigma: The

Breakthrough Management Strategy Revolutionizing
the World's Top Corporations.
Pande, P.S., Neuman, R.P., & Cavanagh, R.R. (2000).

The Six Sigma Way.
CSSBB 2014
CSSBB BODY OF KNOWLEDGE
ASQ Certified Six Sigma Black Belt

Body of Knowledge
The detailed Body of Knowledge is given in the CSSBB
Primer and on the ASQ web site http://www.asq.org
I. Organization-wide Deployment [12 Q]
II. Process Management and Measures [10 Q]
III. Team Management [18 Q]
IV. Define [20 Q]
V. Measure [25 Q]
VI. Analyze [22 Q]
VII. Improve [21 Q]
VIII. Control [15 Q]
IX. Design for Six Sigma (DFSS) [7 Q]

CSSBB 2014
CSSBB BODY OF KNOWLEDGE
Levels of Cognition
In addition to content specifics, the subtext for each
topic in this BOK also indicates the intended complexity
level of the test questions for that topic. These levels are
based on “Levels of Cognition” (from Bloom’s
Taxonomy, Revised 2001) and are presented below in
rank order, from least complex to most complex.
C Remember
C Understand
C Apply
C Analyze
C Evaluate
C Create
© QUALITY COUNCIL OF INDIANA II-1 (14)
CSSBB 2014
II. ENTERPRISE-WIDE DEPLOYMENT I.A.1
SIX SIGMA HAS FOREVER CHANGED

GE®. EVERYONE... IS A TRUE
BELIEVER IN SIX SIGMA, THE WAY
THIS COMPANY NOW WORKS.
JOHN F. WELCH
FORMER GE CHAIRMAN
CSSBB 2014

ORGANIZATION-WIDE CONSIDERATIONS/FUNDAMENTALS
Enterprise-Wide Deployment
Enterprise-wide Deployment is reviewed in the following
topic areas:
C Organization-wide view
C Leadership
Organization-wide considerations are presented in the

following topic areas:
C Six sigma and lean fundamentals

C Continuous improvement methodologies
C Business systems and processes
C Strategic planning and deployment
CSSBB 2014

Value of Six Sigma

Six sigma is a highly disciplined process that focuses
on developing and delivering near-perfect products and
services consistently. It is also a management strategy
to use statistical tools and project work to achieve
breakthrough profitability and quantum gains in quality.
Motorola®, under the direction of Chairman Bob Galvin,

used statistical tools to identify and eliminate variation.
From Bill Smith’s yield theory in 1984, Motorola®
developed six sigma as a key business initiative in 1987.
CSSBB 2014

Value of Six Sigma (Continued)

Motorola® noted that many operations, such as complex
assemblies, tended to shift 1.5 sigma over time. So a
process, with a normal distribution and normal variation
of the mean, would need to have specification limits of
± 6 sigma in order to produce less than 3.4 defects per
million opportunities.
-6 -4 -2 0 2 4 6
The ± 1.5 Sigma shift

CSSBB 2014


Sigma Level ppm
6 sigma 3.4 ppm
5 sigma 233 ppm
4 sigma 6,210 ppm
3 sigma 66,810 ppm
2 sigma 308,770 ppm
1 sigma 697,672 ppm
Defect Levels
Note that Table II in the Appendix provides defect levels

at other sigma values. Various authors report slightly
different failure rates based upon rounding effects and
slight miscalculations.
It should be noted that the term “six sigma” has been

applied to many operations including those with non-
normal distributions, for which a calculation of sigma
would be inappropriate. The principle remains the
same, deliver near perfect products and services by
improving the process and eliminating defects. The end
objective is to delight customers.
CSSBB 2014


The six sigma steps for many organizations are
described as DMAIC:
Define: Select the appropriate responses (the “Ys”)

to be improved.
Measure: Data must be gathered to measure the

response variable.
Analyze: Identify the root causes of defects,

defectives, or significant measurement
deviations whether in or out of specifications.
(The “Xs”, independent variables).
Improve: Reduce variability or eliminate the cause.
Control: With the desired improvements in place,

monitor the process to sustain the
improvements.
CSSBB 2014


Because of the integration of a number of tools, such as
lean manufacturing, DOE (design of experiments), and
DFSS (design for six sigma), six sigma has been
referred to as TQM (total quality management) on
steroids.
The business successes that result from a six sigma

initiative include:
C Cost reductions
C Market - share growth
C Defect reductions
C Culture changes
C Productivity improvements
C Customer relations improvements
C Product and service improvements
C Cycle - time reductions
CSSBB 2014


Motorola® credits the six sigma initiative for savings of
$940 million over three years. AlliedSignal® (now
Honeywell®) reported an estimated $1.5 billion in
savings in 1997.
Harry reports that the average black belt project will

save about $175,000. There should be about 5 to 6
projects per year, per black belt. The ratio of one black
belt per 100 employees can provide a 6% cost reduction
per year. For larger companies, there is usually one
master black belt for every 100 black belts.
CSSBB 2014


Snee provides some reasons why six sigma works:
C Bottom line results

C Senior management is involved
C A disciplined approach is used (DMAIC)
C Short project completion times (3 to 6 months)
C Clearly defined measures of success
C Infrastructure of trained individuals
C Customers and processes are the focus
C A sound statistical approach is used
When operations reach six sigma quality, defects

become so rare that when they do occur, they receive
the full attention necessary to determine and correct the
root cause. As a result, key operations frequently end
up realizing better than six sigma quality.
CSSBB 2014

Six Sigma Foundations

Listed below are some well-known gurus and what they
have contributed to the business and technical
foundations of six sigma. This list is far from inclusive.
Guru Contribution
Philip B. Senior management involvement
Crosby 4 absolutes of quality management
Quality cost measurements
W. Edwards Plan-do-study-act (wide usage)
Deming Top management involvement
System improvements
Constancy of purpose
Armand V. Total quality control/management
Feigenbaum Top management involvement
Kaoru Ishikawa 4M (cause-and-effect) diagrams
Companywide quality control
Next operation as customer
CSSBB 2014

Six Sigma Foundations (Continued)
Guru Contribution
Joseph M. Top management involvement
Juran Quality trilogy
Quality cost measurement
Pareto analysis
Walter A. Assignable cause vs. chance cause
Shewhart Control charts
Plan-do-check-act (in design)
Use of statistics for improvement
Genichi Loss function concepts
Taguchi Signal to noise ratio
Experimental design methods
Concept of design robustness
Bill Smith First introduced six sigma
Mikel Harry The main architect of six sigma
Forrest Author of Implementing Six Sigma
Breyfogle III
CSSBB 2014

Philip B. Crosby (1928 - 2001)

Philip B. Crosby was vice-president of ITT for 14 years.
In 1979, he founded Philip Crosby Associates, Inc. in
Winter Park, Florida. Mr. Crosby consulted, spoke, and
wrote about strategic quality issues throughout his
professional life.
The other quality deep thinkers could be viewed as

academicians, but Crosby was considered a
businessman. This explained the numbers of top
management that flocked to his quality college.
Crosby believed that quality was a significant part of the

company and senior managers must take charge of it.
He believed the quality professional must become more
knowledgeable and communicative about the business.
Crosby stated that corporate management must make
the cost of quality a part of the financial system of their
company.
CSSBB 2014

Philip B. Crosby (Continued)

One of Crosby’s most popular statements on quality
was,
“Quality is conformance to requirements.”
Philip Crosby preached four absolutes of quality

management:
1. Quality means conformance to requirements

2. Quality comes from prevention
3. The quality performance standard is zero defects
4. Quality measurement is the price of nonconformance
The four absolutes of quality management are basic

requirements for understanding the purpose of a quality
system.
CSSBB 2014

Dr. W. Edwards Deming (1900 - 1993)

Dr. Deming was an honorary member of ASQ. He was
awarded the ASQ Shewhart Medal in 1955. During his
life Dr. Deming published over 200 papers, articles, and
books.
W. Edwards Deming was the one individual who stood

for quality and for what it means. He is a national folk
hero in Japan and was perhaps the leading speaker for
the quality revolution in the world. At the Hawthorne
plant he became acquainted with W. Shewhart and
studied Shewhart’s statistical methods.
Deming made several visits to Japan between 1946 and

1948 for the purpose of census taking. JUSE invited
Deming back in 1950 for executive courses in statistical
methods. He refused royalties on his seminar materials
and insisted that the proceeds be used to help the
Japanese people. JUSE named their ultimate quality
prize after him.
His message to America is listed in his famous 14 points

and 7 deadly diseases.
CSSBB 2014

Dr. W. Edwards Deming (Continued)

The Fourteen Obligations of Top Management:
1. Create constancy of purpose for improvement of

products and service
2. Adopt a new philosophy; we are in a new economic
age
3. Cease dependence upon inspection as a way to
achieve quality
4. End the practice of awarding business based on
price tag
5. Constantly improve the process of planning,
production, and service, including people
6. Institute training on the job
7. Institute improved supervision (leadership)
8. Drive out fear
9. Break down barriers between departments
10. Eliminate slogans/targets asking for increased
productivity without providing methods
11. Eliminate numerical quotas
12. Remove barriers that stand between all workers and
their pride of workmanship
13. Institute programs for education and retraining
14. Put all emphasis in the company to work to
accomplish the transformation
CSSBB 2014

Deming’s Seven Deadly Diseases
1. Lack of constancy of purpose to plan a marketable

product and service to keep the company in
business and provide jobs
2. Emphasis on short-term profits
3. Personal evaluation appraisal, by whatever name,

for people in management, the effects of which are
devastating
4. Mobility of management; job hopping
5. Use of visible figures for management, with little or

no consideration of figures that are unknown or
unknowable
6. Excessive medical costs
7. Excessive costs of warranty, fueled by lawyers that

work on contingency fees
Among other educational techniques, Deming promoted

the parable of the red beads, the PDSA cycle, and the
concept of 94% management (system) causes versus
6% special causes.
CSSBB 2014

Dr. W. Edwards Deming (Continued)

Deming’s Chain Reaction
Deming shared the following chain reaction with Japan
in the summer of 1950:
Improve quality º Decrease costs (less rework, fewer

delays) º Productivity improves º Capture the market
with better quality and price º Stay in business º
Provide jobs.
CSSBB 2014

Dr. Armand V. Feigenbaum (1920 - 2014)

Dr. Feigenbaum is generally given credit for establishing
the concept of “total quality control” in the late 1940s
while he was at General Electric. His TQC statement
was first published in 1961, but, at that time, the concept
was so new no one listened.
The TQC philosophy maintains that all areas of the

company must be involved in the quality effort. The
quality effort has generally only affected the shop floor
people, but must extend to all sections of the company.
Products must not only be made quicker and faster, but
also sold faster. Feigenbaum noted that the quality
professional has an opportunity to become more than a
functional specialist.
CSSBB 2014

Dr. Armand V. Feigenbaum (Continued)

The success of TQC includes these principles:
C TQC is a company wide process

C Quality is what the customer says it is
C Quality and production costs are in partnership
C Higher quality will equate to lower costs
C Both individual and team zeal are required
C Management must relentlessly emphasize quality
C Quality and innovation must work together
C All of management must be involved in quality
C Use new and existing technologies
C Quality is the cost-effective route to productivity
C Quality involves both customers and suppliers
CSSBB 2014

Dr. Kaoru Ishikawa (1915 - 1989)

Kaoru Ishikawa was involved with the quality movement
in its earliest beginnings and remained so until his death
in 1989. A review of Ishikawa’s training tapes, produced
in 1981, contain many of the statements of quality that
are in vogue today. Subjects such as total quality
control, next operation as customer, training of workers,
empowerment, customer satisfaction, elimination of
sectionalism, and humanistic management of workers,
are examples.
To reduce confusion between Japanese-style total

quality control and western-style total quality control, he
called the Japanese method the companywide quality
control (CWQC).
CSSBB 2014

Dr. Kaoru Ishikawa (Continued)

CWQC involves the participation of workers from the top
to the bottom of an organization and from the start to
the finish of the product life cycle. CWQC requires a
management philosophy that has respect for humanity.
One of the first concepts that western management took

back to their own shores was the quality circle. The
quality circle represents the bottom up approach.
Ishikawa also wrote that he originated the concept “next

operation as customer” in 1950. Operators concerned
about their own defects were considered spies
whenever they traveled to the next department to view
their original work. The separation of departments was
referred to as sectionalism.
Kaoru Ishikawa was known for his lifelong efforts as the

father of Japanese quality control efforts. The fishbone
diagram is also called the Ishikawa diagram in his
honor.
CSSBB 2014

Dr. Joseph M. Juran (1904 - 2008)

Dr. Juran was the Founder and Chairman Emeritus of
The Juran Institute. He held a B.S. from the University
of Minnesota, a J.D. from Loyola University and
numerous honorary doctorates degrees.
Notable, among Dr. Juran’s 15 books, are:
Juran on Planning for Quality

Juran on Leadership for Quality
Juran on Quality by Design
Quality Planning & Analysis
Juran’s Control Handbook
The publication of his book...Quality Control Handbook

and his work in quality management led to an invitation
from JUSE in 1954. Juran’s first lectures in Japan were
to the 140 largest company CEOs, and later to 150
senior managers. Juran commented that no one was
more surprised than he to see CEOs at the seminars.
CSSBB 2014

Dr. Joseph M. Juran (Continued)

Dr. Juran had a prime basic belief that quality must be
improved at a revolutionary rate.
Juran’s basics for success can be described as follows:
C Top management must commit the time and

resources for success
C Specific quality improvement goals must be in the

business plan and include:
C The means to measure results against goals

C A review of results against goals
C A reward for superior quality performance
C The responsibility for improvements must be

assigned to individuals
C People must be trained for quality management and

improvement
C The workforce must be empowered to participate in

the improvement process
CSSBB 2014

Juran Trilogy
Juran felt that managing for quality requires the same
attention that other functions obtain. Thus, he
developed the Juran or quality trilogy which involves:
C Quality planning
C Quality control
C Quality improvement
Juran sees these items as the keys to success. Top

management can follow this sequence just as they
would use one for financial budgeting, cost control, and
profit improvement.
For any project, quality planning is used to create the

process that will enable one to meet the desired goals.
The concept of quality control is used to monitor and
adjust the process. Only quality improvement activities
will reduce the chronic losses and move the process to
a better and improved state of control.
CSSBB 2014

Dr. Walter A. Shewhart (1891 - 1967)

Shewhart’s most notable book was: Economic Control
of Quality of Manufactured Product (1931)
Dr. Shewhart worked for the Western Electric Company,

a manufacturer of telephone hardware for Bell
Telephone, from 1918 until 1924. Bell Telephone’s
engineers had a need to reduce the frequency of failures
and repairs. In 1924, Shewhart framed the problem in
terms of “assignable cause” and “chance cause”
variation and introduced the control chart as a tool for
distinguishing between the two.
Walter Shewhart’s statistical process control charts

have become a quality legacy that continues today.
Control charts are widely used to monitor processes
and to determine when a process changes. Dr. Deming
stated that Shewhart’s genius was in recognizing when
to act, and when to leave a process alone.
CSSBB 2014

The Shewhart Cycle

The historical evolution of the PDCA problem solving
cycle is interesting. Deming presented the following
product design cycle (which he attributed to Shewhart)
to the Japanese in 1951:
1. Design the product (with appropriate tests)
2. Make the product (tested both in laboratory and

production)
3. Put the product on the market
4. Test the product in service through market research
5. Redesign the product, in light of consumer reaction,

and continue the cycle
Perhaps from this concept, the Japanese evolved a

general problem solving process called PDCA.
CSSBB 2014

Dr. Genichi Taguchi (1924 - 2012)

Dr. Taguchi is called the “Father of Quality
Engineering.” One of his profound statements on
quality was, “Quality is related to the financial loss to
society caused by a product during its life cycle.”
Taguchi’s techniques enabled engineers to develop

products and processes in a fraction of the time as
required by conventional engineering practices.
CSSBB 2014

Dr. Genichi Taguchi (Continued)

Taguchi’s plan takes a different view of product quality:
1. The evaluation of quality
Use the loss function and signal-to-noise ratio as

ways to evaluate the cost of not meeting the
target value.
2. Improvement of quality and cost factors
Use statistical methods for system parameter and

tolerance design of the product.
3. Monitoring and maintaining quality
Reduce the variability of the production line.
Taguchi methods, and other design of experiment

techniques, have been described as tools that tell us
how to make something happen, whereas most
statistical methods tell us what has happened.
CSSBB 2014

Bill Smith (1929 - 1993)

Bill Smith was Vice President and Senior Quality
Assurance Manager for Land Mobile Products Sector,
Motorola, when he introduced six sigma as a measure of
quality in 1986. Mr. Smith determined that quality levels
beyond three sigma were necessary. He suggested a ±
six sigma level allowing for a ± 1.5 sigma shift.
Mr. Smith helped Robert W. Galvin, Chairman and CEO

of Motorola, recognize the need to control variation and
to work toward 3.4 defects per million or for six sigma
levels of quality. Later with Mikel Harry, Smith
developed the initial four-step six sigma stages:
measure, analyze, improve, and control, to reduce the
defect levels.
CSSBB 2014

Mikel Harry
Mikel Harry and Richard Schroeder founded Six Sigma
Academy in 1994 as a consulting firm specializing in the
six sigma methodology. Mikel Harry has called Bill
Smith “The father of six sigma” and gave himself the
title “The godfather of six sigma.” Many industry people
have called Mikel Harry the main “architect” of the six
sigma movement, as he has been the most widely
known driver in the industry.
Forrest Breyfogle III

Forrest Breyfogle III is the Founder and CEO of Smarter
Solutions, Inc., based in Austin, Texas. The company
was formed in 1992 to provide training and consulting
services for the six sigma methodology. Mr. Breyfogle’s
book Implementing Six Sigma, 2nd ed. has made a
significant contribution to the understanding of six
sigma.
CSSBB 2014

Lean Enterprise
The lean enterprise encompasses the entire production
system, beginning with the customer. It includes sales
outlets, the final assembler, product or process design,
and all tiers of the supply chain. No implementation of
lean manufacturing can reach its full potential without
including the entire enterprise in its planning.
Lean Manufacturing
Lean techniques are, in their most basic form, the
systematic identification and elimination of waste, the
implementation of the concepts of continuous flow, and
customer pull. The touted benefits of lean production
systems include lower production costs, fewer
personnel, quicker product development, higher quality,
higher profitability, and greater system flexibility. By
continually focusing on waste reduction, there is truly
no end to the benefits that can be achieved.
Generally, five areas drive the lean producer: cost,

quality, delivery, safety, and morale.
CSSBB 2014

Lean Techniques in Service

Every system contains waste. Whether one is producing
a product, processing a material, or providing a service,
there are elements which are considered waste. The
techniques for analyzing systems, identifying and
reducing waste, and focusing on the customer are
applicable in any system, and in any industry.
Any implementation of lean techniques will be different,

depending on various factors such as industry, internal
culture, and internal business considerations. The tools
used to implement lean operations, and the order in
which one combines them, are highly dependent on
whether a company is a discrete manufacturer,
continuous producer, or provider of a service.
CSSBB 2014

Lean Pioneers
The following is a list of major contributors to the
concept of lean enterprise.
Pioneer Contribution
Frederick Wrote Principles of Scientific Management
W. Taylor Divided work into component parts
Was a renowned efficiency expert
Maximized output using scientific methods
Henry Called the father of mass production
Ford Advocated waste reduction
Founded Ford Motor Company
Made transportation affordable
Sakichi Known as a hands-on inventor
Toyoda Developed the jidoka concept
Initiated the Toyota Motor Company
Kiichiro Continued the work of Sakichi
Toyoda Promoted mistake proofing concepts
President of Toyota Motor Company
CSSBB 2014

Lean Pioneers (Continued)
Pioneer Contribution
Eiji Was the cousin of Kiichiro Toyoda
Toyoda Developed an automotive research lab
Hired outstanding people within TMC
Became the Chairman of TMC
Taiichi Created the Toyota production system
Ohno Integrated the TPS into the supply chain
Had the vision to eliminate waste
Shigeo Developed the SMED system
Shingo Help develop other TPS elements
Womack Promoters of lean enterprise
& Jones Co-authors of lean thinking books
Anand CEO of TBM Consulting Group
Sharma Prominent lean enterprise author
Michael Known for lean six sigma books
George Founder of The George Group
CSSBB 2014

Frederick Winslow Taylor (1856 - 1915)

Frederick Taylor was the first efficiency expert; the
original time and motion study specialist. He applied
scientific methods to obtain maximum output. This was
accomplished by having management in control of the
workplace and by detailing the minute routine of the
worker. Through operations analysis, Taylor took away
job complexity.
Some key Taylor concepts are:
C Understand each element of the task

C Select, train, and develop the worker
C Divide work between management and worker
C Cooperate with the worker to follow the procedures
CSSBB 2014

Henry Ford (1863 - 1947)

The Ford Motor Company was founded in 1903 with the
introduction of the Model A. By 1908, after 20 design
changes, the Model T was created. Mr. Ford had a
vehicle that was designed for both the ease of
manufacture and use. The vehicle had parts with
interchangeability and simplicity. The common man
was able to drive and repair his own car. In 1927 a
second Model A was launched to meet the features
offered by other U.S. competitors.
Henry Ford was the master of “mass production.” The

successful implementation of the assembly line at the
Highland Park Plant in Detroit, in 1913 reduced costs
and increased productivity for Ford Motor Company.
The reduced manufacturing costs made cars more
affordable for Americans.
Mr. Ford went beyond just managing the internal

resources of the plant. He sought to reduce costs and
increase productivity by controlling the costs of raw
materials.
CSSBB 2014

Henry Ford (Continued)

Mr. Ford was an advocate of reducing waste in every
operational area. Some examples include:
C Using straw to make steering wheels

C Reworking and reusing worn steel rails
C Remelting scrap steel at the River Rouge plant
C Reworking broken tools and equipment
C Converting paper and rags into binder board
Mass production involved the interchangeability of parts

and workers, simpler tasks, and better organization.
These techniques were widely used for 60-70 years and
were adopted by companies in North America and
Europe.
CSSBB 2014

Sakichi Toyoda (1867 - 1930)

Sakichi Toyoda was a businessman and was called the
“King of Inventors.” He was a carpenter by trade and,
thus, able to work with his hands. In 1897 he invented
the first Japanese power loom. Because other members
of the Toyoda family and friends were in the cottage
industry of weaving, this led him to try to reduce the
amount of manual labor and effort required for weaving.
For his efforts he used a steam engine as the source of
power for the looms.
A prime concept used at Loom Works was jidoka

(automation with a human touch). This invention was
designed to stop the loom whenever a thread broke. A
human did not always have to be present to oversee
each loom. This enabled workers to handle more than
one loom and provide more value-added work.
CSSBB 2014

Kiichiro Toyoda (1895 - 1952)

Kiichiro Toyoda was the son of Sakichi Toyoda and
Second President of Toyota Motor Company. In 1929,
Kiichiro Toyoda went to England and negotiated the
patent rights to the “mistake proof” loom. The funds
from the sale helped to finance the automotive efforts of
Toyoda Automatic Loom Works. Kiichiro Toyoda made
a tour of US auto plants in 1929, followed by his own
research efforts on motor vehicles in 1930.
The Toyota Motor Company (TMC) was spun-off as a

separate company in 1937. From the beginning, the
concept of just-in-time production was used. Due to a
lack of materials, this concept had to be used for
economics and to increase cash flow. Mr. K. Toyoda
was very much influenced by his trips to Ford plants
and by seeing the supermarket process of restocking
goods on the shelves.
CSSBB 2014

Eiji Toyoda (1913 - 1999)

Eiji Toyoda was a younger cousin to Kiichiro Toyoda.
He attended Tokyo Imperial University studying
Mechanical Engineering (1933 - 1936). Upon graduation,
he was persuaded to join his cousin’s business and
started a research lab called the “car hotel.” This
garage housed Eiji and his staff as they conducted
research on engines, repaired cars, and worked on other
special projects. He worked in the auto business during
the war effort making trucks. He became a Director of
Toyota Motor Company in 1945, and Managing Director
in 1950.
During 1950, he traveled to the United States for a 3-

month tour of the auto plants and their suppliers. This
trip provided evidence to Eiji Toyoda that little Toyota
Motor Company could compete in the automotive arena,
but not using the same “mass production” techniques.
There was waste in the system and TMC could build a
new system from that.
CSSBB 2014

Taiichi Ohno (1912 - 1990)

Taiichi Ohno was the creator of the Toyota Production
System. He joined Toyoda Spinning and Weaving in
1932. In 1943, he transferred to Toyota Motor Company.
By 1947 he managed the machine shop where he
experimented with parallel lines and/or L-shaped
processes. There was much resistance from the
machine operators. Since he was from the weaving
company, he was aware of jidoka (automation with a
human touch) and used it productively in the auto
company.
In the 1950s, he also toured the United States auto

plants to view and evaluate the “mass production”
process. From the tour, Ohno learned that the mass
production system could achieve economies of scale
and reduced costs, but the system was still full of waste.
The waste was present in the forms of over production,
excess inventory, long setup times, rework, etc.
CSSBB 2014

Taiichi Ohno (Continued)

Some sample techniques of the Toyota Production
System that Ohno and his team developed:
1. Pull system (supermarket, 7-Eleven idea)

2. Muda (7 types of waste)
3. Quick die changes (from days, to minutes)
4. Flexible job assignments
5. Removing non-value added work
6. Kanban methods
7. U-shaped cells
8. One-piece flow
9. Production leveling
Mr. Ohno had the vision and focus to uncover and

eliminate waste within Toyota and their suppliers.
From 1950 on, as a manager and executive, and with the

backing of President Eiji Toyoda, he pushed and fought
to install the concepts of lean throughout Toyota and
into the supply base.
CSSBB 2014

Shigeo Shingo (1909 - 1990)

Shingo was one of Japan’s foremost consultants on
manufacturing operations improvement. He has written
many books on improvement. In 1945, he became a
consultant to industry through the Japan Management
Association (JMA). He started performing quick die
change work in 1950 at Toyo Industries.
By 1959, Shigeo Shingo formed his own consulting firm,

and provided consulting throughout the Far East. Much
of his work centered on mistake proofing, zero quality
control, and supplier sourcing. He worked closely on
the SMED concept.
Shigeo Shingo trained and consulted for TMC from 1954

until 1982. While he was not a Toyota employee, he was
a consultant that assisted in the development of the
Toyota Production System.
CSSBB 2014

James Womack and Daniel Jones

James Womack and Daniel Jones have been linked
together as researchers on the capabilities of the
automotive industry since 1979.
They conducted a 5-year study of the Toyota system and

the rest of the automotive industry. This study led to the
monumental book The Machine that Changed the World.
This book changed the world of mass production by
detailing how lean manufacturing produces products
with perhaps one-half the resources as before (human
effort, space, investment, engineering, and time).
CSSBB 2014

Anand Sharma
Anand Sharma is President and CEO of TBM Consulting
Group. His supporters state that he is an expert who
can figure out what is wrong with an organization by
walking the shop floor. He proclaims, “Where other
people see complexity, I look at how simple things can
be.”
TBM Consulting Group, employing over 70 employees,

has worked with over 500 enterprises on improving
manufacturing productivity and profits. Mr. Sharma
prides himself on refusing to work with firms that will lay
off workers due to use of his system.
CSSBB 2014

Michael George
Michael George is Chairman and CEO of The George
Group based in Dallas, Texas. His company focuses on
operational performance and shareholder value through
six sigma, lean six sigma, management of complexity,
and innovation efforts.
In 1969 he traveled to Japan to study the Toyota

Production System. The George Group was formed in
1986. Mr. George is the holder of several patents on the
reduction of process cycle time and complexity. He has
authored or co-authored a multitude of lean six sigma
books including: Fast Innovation, Lean Six Sigma, Lean
Six Sigma for Service, and Conquering Complexity in
Your Business.
CSSBB 2014

ORGANIZATION-WIDE CONSIDERATIONS/METHODOLOGIES
Integration of Lean and Six Sigma

There is an ongoing debate in some organizations
regarding the difference between lean and six sigma,
and whether they are mutually exclusive. Lean is about
eliminating wastes, taking time out of processes, and
creating better flow.
Six sigma has been defined in a variety of ways. One

definition states, “Six sigma is ... a business strategy
and philosophy built around the concept that companies
can gain a competitive edge by reducing defects in their
industrial and commercial processes.”
CSSBB 2014

Integration of Lean and Six Sigma (Cont’d)

A few key characteristics of lean and six sigma are
discussed and compared below.
Topic Six Sigma Lean

Improvement Reduce variation Reduce waste
Justification Six sigma (3.4 DPMO) Speed (velocity)
Main Savings Cost of poor quality Operating costs
Learning Long Short
Curve
Project Various approaches Value stream
Selection mapping
Project Length 2 - 6 months 1 Week - 3
months
Driver Data Demand
Complexity High Moderate
CSSBB 2014


Both six sigma and lean focus heavily on satisfying
customers. Six sigma makes customers the primary
driver for action in a “war on variation” and identifies
opportunities that promise a large, fairly immediate,
financial reward. Lean considers customer inputs and
conducts a “war on waste.”
Both six sigma and lean empower people to create

process stability and a culture of continuous
improvement. The cornerstones of a lean strategy are
tools such as value stream mapping (VSM), workplace
organization (5S), total productive maintenance (TPM),
kanban/pull systems, kaizen, setup reduction, teamwork,
error- proofing, problem solving, cellular manufacturing,
and one-piece flow.
CSSBB 2014


Many problem identification and problem solving
techniques are commonly used with both lean and six
sigma methodologies. These include brainstorming,
cause-and-effect diagrams, 5 “whys”, Pareto analysis,
8-Ds, FMEAs, and others. Both six sigma and lean
methodologies have a heavy emphasis on careful
problem definition. Six sigma better promotes a
rigorous, systematic process to find the true root
cause(s) of the problem.
Value stream mapping (VSM) is the principal lean

diagnostic tool. It is credited to Toyota, who called it
material and information flow mapping. VSM creates a
visual representation of what is happening in a process
to improve system performance.
CSSBB 2014


If major business problems fall into the following
categories:
C There seems to be a lot of waste

C There is a need to minimize inventories
C There is a need to minimize redundancies
C There is a need to improve work flows
C There is a need to speed up processes
C There are human mistakes
If so, then lean tools should be utilized to:
C Eliminate wastes
C Increase speeds
C Minimize inventories
C Simplify processes
C Improve flows
C Mistake proof processes
CSSBB 2014


If organizational challenges exhibit the following
attributes:
C There are quality issues

C There is excessive variation
C There are complex problems
C There are challenging root cause identifications
C There are numerous technical considerations
In these cases, six sigma tools should be utilized to:
C Minimize variation
C Apply scientific problem solving
C Utilize robust project chartering
C Focus on quality issues
C Employ technical methodologies
Most executives recognize that they have a combination

of both sets of issues.
CSSBB 2014


What has been occurring for some time (at least the past
several years) is a marriage of lean and six sigma
initiatives into a unified approach. Presented
graphically, if lean specific projects represent a 6%
corporate improvement over time, and six sigma
initiatives represent another 6% improvement, then a
combination could potentially represent an improvement
of 12% (or more). Refer to the Figure below.
12 LEAN SIX SIGMA LEAN SIX SIGMA

% IMPROVEMENT
10
8
6
4
2
TIME º TIME º TIME º
An increasing number of organizations (manufacturing,

service, hospitals, municipalities, military, insurance,
etc.) have been unifying their efforts into a lean six
sigma approach.
CSSBB 2014

Integration of Lean and Six Sigma (Cont’)

Listed below are a variety lean six sigma tools in a
DMAIC matrix.
Define Measure Analyze Improve Control

Value Stream Prioritization Regression DOE SPC
Mapping Matrices Analysis
Problem MSA Studies 5 - Whys Kaizen Visual
Statement Events Controls
Voice of the Capability Cause - Effect TOC Control Plans

Customer Studies Diagrams
Communi- Videotaping Root Cause Pull Systems TPM
cation Plans Analysis
CTQ Issues Time Studies ANOVA SMED/SUR Standard Work
Business SIPOC Multi-Vari 5S or 6S Procedures and
Results Analysis Instructions
Benchmarking Collecting Hypothesis Work Flow Training
Data Testing Improvement Requirements
CSSBB 2014

Linking Projects to Organizational Goals

A readiness assessment includes a review of the
following areas:
Assess the outlook and future path of the business:
C Is the strategy course clear for the company?

C Can we meet our financial and growth goals?
C Do we respond effectively to new circumstances?
Evaluate the current organizational performance:
C What are our current overall business results?

C How effectively customer requirements are met?
C How effectively are we operating?
Review the capacity for change and improvement:
C How effective are we in managing system changes?

C Do we manage cross functional processes well?
C Are our current efforts in conflict with six sigma?
CSSBB 2014

Linking Projects to Goals (Cont’d)

Pande states that a decision on six sigma might be
negative if the following conditions exist:
C The company already has an effective improvement

effort in place
C Current changes are already overwhelming the

company's resources
C The potential gains aren’t sufficient to finance the

necessary investments
CSSBB 2014

Linking Projects to Goals (Cont’d)

Harry details a methodology to focus the deployment of
six sigma projects. There are a considerable number of
options, dependent upon the goals and objectives of the
organization. Considerations include:
C Focus on project cost savings

C Focus on customer satisfaction deliverables
C Focus on processes
C Focus on problems
C Focus on a targeted location
C Focus on design
C Focus on supplier processes
CSSBB 2014

Kaizen Events
A kaizen event consists of a cross functional team
(usually consisting of 6 to 18 people) assigned to
improve an existing process in a limited amount of time
(typically 3 to 5 days).
These events gather managers, process owners,

operators, black belts and/or green belts together to
map and improve an activity. It is important that the
project be within the scope of the participants. Equally
important to the improvement is the solicitation of input
and buy-in from other affected parties.
Kaizen events are also discussed in Section IX.

CSSBB 2014

ORGANIZATION-WIDE CONSIDERATIONS/BUSINESS SYSTEMS
Systems
From an organizational standpoint, a system is defined
as a series of actions, activities, elements, components,
departments, or processes that work together for a
definite purpose. System effectiveness is a measure of
the degree to which a system can be expected to
achieve a set of specific requirements.
CSSBB 2014

Business Systems
Business systems are made up of a variety of
processes. These processes include, marketing, sales,
strategic planning, production, delivery, human
resources, accounting, maintenance, development and
engineering. All of these functions must work together
to achieve customer satisfaction.
Management leadership is a measure of how senior

executives guide the organization and how the
organization addresses its responsibilities to the public
and practices good citizenship. Some key activities:
C Strategic planning
C Customer and market focus
C Information and analysis
C Human resource focus
C Process management
CSSBB 2014

Processes
Processes are definable portions of a system or
subsystem that consist of a number of individual
elements, actions, or steps.
Systems, subsystems, and processes exist for all

human activities. They are utilized in both the
manufacture of a product or the delivery of a service.
CSSBB 2014

Business Processes
Process interactions must also be evaluated to ensure
that changes that improve one operation do not create
more defects in another operation.
The following figure illustrates how work and ideas flow

within systems.
Inputs Process 1 Outputs

Data Products
Process 2 Services
Options & Ideas
Customer
Orders Remedies
End
Process 3
Specifications Designs
Money Root Causes
Customer Needs Process 4 Training
Suppliers Others
Process 5
Feedback
CSSBB 2014

Business Responsibilities
In today’s world, organizations must keep pace with
ever increasing changes. The complexity of the
business requires numerous functions in order to be
competitive. A brief description of common business
functional responsibilities include the following:
C Human Resources
C Engineering
C Sales and Marketing
C Finance
C Product Liability
C Manufacturing
C Safety and Health
C Legal and Regulatory
CSSBB 2014

Business Responsibilities (Continued)

Other common business responsibilities:
C Research and Development (R&D)

C Purchasing
C IT or MIS
C Production Planning and Scheduling
C Quality
C Environmental
C Technology
C Servicing
CSSBB 2014

ORGANIZATION-WIDE CONSIDERATIONS/STRATEGIC PLANNING
SWOT Analysis
In the strategic planning process, a firm’s internal
strengths and weaknesses, as well as their external
opportunities and threats, have to be determined.
SWOT is an acronym meaning strengths, weaknesses,
opportunities, and threats.
SWOT analysis requires that a comprehensive appraisal

of internal and external situations be undertaken before
suitable strategic options can be determined. Good
strategies are built on the strengths of a company and
on exploiting opportunities. Each company has a
different set of opportunities and threats with differing
strengths and weaknesses. Thus, strategies will be
different for each firm.
CSSBB 2014

SWOT Analysis (Continued)
The table below shows some SWOT analysis ideas.
Internal Strengths: Internal Weaknesses:

core competencies in critical areas too many goals
solid finances lack of strategic focus
market leader obsolete facilities
proprietary technology outdated technology
cost advantages inexperienced management
good marketing skills manufacturing problems
management excellence weak marketing skills
world class manufacturing lack of growth capital
good technical and workforce skills weak cash flow
superior brand names inadequate R&D
web skills can not implement plans
External Opportunities: External Threats:
expansion to new markets global competition
product lines can be broadened substitute products are available
transfer technical skills to new products slow market growth
low industry rivalry legal and regulatory requirements
minimal regulatory requirements recessionary cycle
new emerging technologies strong customers or suppliers
positive growth cycle new competitors
business-to-business on the internet business-to-business on the internet
e-commerce e-commerce
CSSBB 2014

Strengths and Weaknesses

A strength is something that the company is good at
doing. The strength can be a skill, expertise, a patent,
key resource, technology, market position, or anything
that provides an advantage.
It is desirable to compete on the basis of strengths. If a

firm has no outstanding strength or competency, their
ability to successfully compete can be difficult. Some
strengths that a company may enjoy include:
C Engineering expertise
C Technical patents
C Skilled workforce
C Solid financial position
C Reputation for quality
A weakness is something that the firm lacks or is a

condition that puts it at a disadvantage. Management
may not be effective in evaluating their internal
strengths and weaknesses objectively. Either the
strengths can be overly magnified or the weaknesses
minimized, such that the process is negated.
CSSBB 2014

Strengths and Weaknesses (Continued)

To be adequate, SWOT analysis should cover the
following key areas:
C An evaluation of each sub-unit of business (or line

of business) and critical success indicators. Are
there winners and losers? Is money being made?
C The status of tracking or control systems for the

critical success indicators.
C An indication of the firm’s level of creativity, risk

taking, and competitive approach.
C An assessment of the resources available to

implement plans. What are the current allocated
resources, resource plans, etc.?
C An analysis of the current organizational culture

and of the firm’s way of doing business. Does the
culture reward teamwork, success, and innovation?
CSSBB 2014

Strengths and Weaknesses (Continued)

Additional weaknesses include:
C Poor cash flow

C Outdated technology
C High overhead expenses
C Lack of access to skilled labor
C Poor quality perception by customers
Opportunities and Threats

The firm must be able to assess the external
environment in preparation for challenges. SWOT
analysis is designed to make management go beyond
their plant boundaries in the strategic planning effort.
The external environment can include assessment of the
following:
C Economic environment
C Socio-political environment
C Social environment
C Technological environment
C Competitive environment
CSSBB 2014

Opportunities and Threats (Continued)

A firm’s external world will provide opportunities and
threats. The strategy must match up with:
C Opportunities suited to the firm’s capabilities

C Defenses against external threats
C Changes to the external environment
CSSBB 2014

SWOT Analysis Deployment

SWOT analysis requires management to develop an
objective view of the firm.
Hamel rightfully argues that diversity in the gene pool

must exist. This can be accomplished by including all
levels of people in the strategic meetings; and even
more so by inviting the newest employees to participate.
Upon the completion of the SWOT analysis, the

implementation phase begins. If the firm has
deficiencies (weaknesses), then efforts must be made to
shore them up, or to acquire resources to build them up.
Action plans have to be developed to become
competent or world class in any desired areas.
CSSBB 2014

PEST Analysis
The environment that organizations face can be
analyzed through a combination of four factors:
political, economic, social, and technological. The
acronym for this approach is called PEST. The concept
is to constantly scan these four areas in order to detect
changes in the external environment. Subtle events that
appear quite small and insignificant can eventually have
huge consequences on business plans.
Frequent monitoring of the environment can aid in the

discovery and understanding of corporate risks.
CSSBB 2014

PEST Analysis (Continued)

The table below illustrates some items to consider in a
PEST analysis.
Political Economic Social Technological
Income taxes Recession Birth rates Production

technology
Corporate tax rates Employment Death rates New materials
levels
National elections Inflation Levels of R&D investment
education by industry
Monetary policies Interest rates Population New processes
diversity
Exchange rates Stock market Cultural diversity Information
trends
Intellectual Business cycles Age of Diffusion rate of
property protection population technology
Business Salary growth Social classes Knowledge
regulations management
Education and Material prices Creative class Emerging
training growth technologies
Industrial policies Subsidies Life styles Energy sources
PEST analysis is a method to uncover hidden factors in

order to assess and counter corporate risks.
CSSBB 2014

Portfolio Analysis
Technical processes include technology portfolio
analysis, research and technology development (R&TD),
product commercialization, and post-launch engineering
work. The approach involves enabling and enhancing
technical processes to prevent problems before they
become an issue.
Inbound R&TD is focused on strategic technology

portfolio definition, development, optimization, and
transfer.
Outbound post-launch engineering is focused on

operations in post-launch production and service
engineering support.
The product and technology portfolio renewal process

is the first of two strategic processes in which research
and development (R&D) professionals can use six sigma
methods.
The second process is the formal development of new

technologies that the product and technology portfolio
process requires.
CSSBB 2014

Portfolio Analysis (Continued)

Product portfolio Technology Specify R&TD Specify product
requirements portfolio to fulfill portfolio
and candidate requirements technology to fulfill product
analysis and architectures requirements requirements
The strategic component consists of the inbound

technical processes, research and technology
development; and the tactical component is product
design engineering done during commercialization.
Strategic Processes - Inbound

C Product portfolio renewal R&D develops platforms and modular
B Multigenerational product lines designs that help enable the product line
C Research and technology development strategy
B Platforms
B Modular designs
Inbound R&D and product engineering
Tactical Processes - Inbound Product design engineering

C Product commercialization commercializes individual elements of
B Specific product inbound marketing the new product portfolio
B Specific product design engineering
Operational Processes - Outbound

C Post-launch product line management
B Product line outbound marketing
B Sales and product line support
B Production and service engineering
Process Linkage Diagram

CSSBB 2014

Portfolio Analysis (Continued)

The IDEA process for product portfolio definition and
development consists of the following phases:
C Identify markets, their segments, and opportunities

using technology benchmarking and road mapping
C Define portfolio requirements and product

architectural alternatives
C Evaluate product alternatives against competitive

portfolios, then select
C Activate ranked and resourced individual product

commercialization projects
The define phase is the key transfer point for delivering

product portfolio requirements to the R&TD
organization. R&TD receives these diverse
requirements and translates them into technology
requirements.
CSSBB 2014

Hoshin Kanri X-Matrix Planning

A simplified hoshin kanri X matrix is shown below:
PROJECTS
Strategies
AIMS 3 DELIVERY 5 WHO

Policies 1 4
Goals Responsibility
2
RESULTS
Effects
A Simplified Hoshin X-Matrix Diagram

CSSBB 2014

Hoshin Kanri X-Matrix Planning (Cont’d)

1. Aims. What are the targeted improvements?
2. Results. How much improvement is needed?
3. Projects. How are the improvements to be made?
4. Delivery. What is the implementation timeline?
5. Responsibility. Who is responsible?
The above categories can also be identified as:
1. What
2. How much of what
3. How - a step by step approach
4. How much of how and when - a timeline
5. Who - the department or responsible individual
The hoshin kanri X-matrix is typically reviewed monthly
with an expanded annual review. This annual review will
assist with the next planning cycle.
An expanded X-matrix is shown in the Primer.

CSSBB 2014
II. ENTERPRISE-WIDE DEPLOYMENT I.B.1

LEADERSHIP/ROLES & RESPONSIBILITIES
Enterprise Leadership is presented in the following topic

areas:
C Roles and responsibilities

C Organizational roadblocks and change management
Leadership Responsibilities
There are numerous ways to structure a six sigma
implementation strategy. Successful applications,
however, share a common core of management support,
training, recognition, and reinforcement.
CSSBB 2014

Management Support
Effective six sigma programs do not happen
accidentally. Careful planning and implementation are
required to ensure that the proper resources are
available and applied to the right problems. Key
resources may include people trained in problem
solving tools, measurement equipment, analysis tools,
and capital resources.
CSSBB 2014

Training
The role of training, in the successful implementation of
six sigma, is fundamental. Needed skills for
breakthrough improvements cannot be developed
without proper training. All companies that
implemented successful six sigma programs have found
that training investments pay back significant benefits.
The diagram below outlines a training plan. The relative

volume of each diagram level represents the relative
number of people receiving training.
Senior Management - Sponsorship Training
Master Black Belt Candidates - Master Black Belt Training
Management - Executive Training
Black Belt Candidates - Black Belt Training
Supervisors - Overview Training
Green Belt Candidates - Green Belt Training
Everyone - Six Sigma Orientation Training

CSSBB 2014

Training (Continued)
In some organizations, black belts are full time positions
that report directly to management sponsors, who, in
turn, assign specific projects to them. Green belts are
typically within the normal organizational structure and
are assigned to process improvement teams as needed.
Black belts have specific mentoring responsibilities,
including the development of individuals assigned to
them.
Black belts are typically responsible for mentoring 1 to

3 green belts or black belt candidates.
Master black belts are responsible for coaching and

training black belts in order to make the best use of their
skills. Master black belts also train and coach
management in order to help them support the black
belt program.
CSSBB 2014

Reward and Reinforcement

Reward and reinforcement may be one of the hardest
parts of successfully institutionalizing a six sigma
program. Black belts and green belts must have
positive career paths in order to encourage the best
candidates to commit to the extensive training and
development required.
It is also important that green and black belts experience

the rewards of achieving significant savings for the
company. At the same time, other team members must
be recognized for their contribution to performance
improvements.
CSSBB 2014

Six Sigma Roles

Many organizations have implemented the following
roles in their six sigma programs.
C Black Belts C Executive Sponsors

C Master Black Belts C Champions
C Green Belts C Process Owners
Black Belts
Six sigma black belts are most effective in full-time
process improvement positions. Six sigma black belts
are individuals who have studied and demonstrated skill
in implementation of the principles, practices, and
techniques of six sigma for maximum cost reduction
and profit improvement.
CSSBB 2014

Black Belts (Continued)

Black belts may be utilized as team leaders responsible
for measuring, analyzing, improving, and controlling key
processes that influence customer satisfaction and/or
productivity growth. Black belts may also operate as
internal consultants, working with a number of teams at
once.
Black belts often receive coaching from a master black

belt to guide them through projects. Black belts have
the following duties in their company:
Mentor: Use a network of trained individuals

Teacher: Train local personnel
Coach: Provide support to project personnel
Identifier: Discover improvement opportunities
Influencer: Be an advocate of six sigma tools
CSSBB 2014

Master Black Belts

Master black belts are typically in full-time process
improvement positions. They are teachers who mentor
black belts and review their projects. Selection criteria
for master black belts includes both quantitative skills
and the ability to teach and mentor. For master black
belt recognition, an individual must be an active black
belt who continues to demonstrate skill through
significant positive financial impact and customer
benefits on projects.
CSSBB 2014

Green Belts
Six sigma green belts are not usually in full-time
process improvement positions. Green belts must
demonstrate proficiency with statistical tools by using
them for positive financial impact and customer
benefits. Individuals may remain green belts or, with
experience, they may become black belts. Green belts
operate under the supervision and guidance of a black
belt or master black belt.
CSSBB 2014

Executive Sponsors
Executive leadership sets the direction and priorities for
the organization. The executive team is comprised of
the leaders that will communicate, lead, and direct the
company’s overall objectives towards successful and
profitable six sigma deployment. Executives typically
receive training that includes a six sigma program
overview, examples of successful deployment and
strategies, and tools and methods for definition,
measurement, analysis, improvement, and control.
CSSBB 2014

Champions
Six sigma champions are typically upper level managers
that control and allocate resources to promote process
improvements and black belt development. Champions
are trained in the core concepts of six sigma and
deployment strategies used by their organization.
Champions work with black belts to ensure that senior
management is aware of the status of six sigma
deployment. Champions ensure that resources are
available for training and project completion.
CSSBB 2014

Process Owners
Key processes should have a process owner. A process
owner coordinates process improvement activities and
monitors progress on a regular basis. Process owners
work with black belts to improve the processes for
which they are responsible. Process owners should
have basic training in the core statistical tools but will
typically only gain proficiency with those techniques
used to improve their individual processes. In some
organizations, process owners may be six sigma
champions.
CSSBB 2014

Six Sigma Structure

Companies have differing duties and terminologies for
organizational roles that support six sigma
improvement. Some of the common functions and
optional structures are below.
Functions Structure Options

Executive C Six sigma steering committee
direction C Quality council
C Executive steering council
Six sigma C Six sigma manager
management C Six sigma director
C Master black belt
Process C Champion
owner C Sponsor
Sponsor C Process owner
C Champion
Coach C Master black belt
C Black belt
Team leader C Trained facilitator
C Black belt
C Green belt
Team member C Associate with team training
C Associate with process knowledge
C Green belt
CSSBB 2014

LEADERSHIP/ORGANIZATIONAL ROADBLOCKS
Organizational Roadblocks
An in-depth description of organization structures is not
provided here. The purpose of this discussion is to
focus on some potential problems and/or roadblocks
related to existing structures and cultures.
Flat Organizations
In flat organizations decision making is forced to lower
organizational levels. Therefore, the decisions can be
erratic and inconsistent. There are fewer decision
levels; therefore, the loss of key people can have a
negative impact on these organizations.
CSSBB 2014

Tall (Vertical) Organizations

In a vertical organization, the lower levels have their
decisions reviewed by the individuals above them. This
also means that communications are not as rapid. As
jobs are added to handle the details of decision making,
more layers of management are created. Sometimes,
these levels become impediments. Upper management
is often out of the loop. The vertical organizational
culture can become too bureaucratic.
Functional Organizations
The people in functional organizations can become very
specialized in their field of expertise. Well rounded
individuals may be difficult to find. Coordination of
projects or problems can be more difficult than in a flat
organization. There is a danger that sub-organizational
values and shared assumptions may become too inbred.
CSSBB 2014

Product Organizations
In a product organization, there can be a duplication of
selected services. Individuals may lack overall
corporate focus since they are concerned with a smaller
piece of the pie. Although segments of the company are
better directed at meeting competitor challenges, they
can also compete against each other for company
resources and consumer markets.
Geographic Organizations
This arrangement presents similar problems as
encountered with functional organizations. Additionally,
many undesirable nationalistic or regional cultural
features may arise.
CSSBB 2014

Matrix Organizations
In matrix organizations, a specialist can report to two or
more people, violating the one boss rule. It is often
difficult for the specialist to decide which superior to
respond to first.
Team Based Organizations

The entire employee selection process is much more
stringent for a team based structure. Management time
is directed at employee training and support. An
organization that is experiencing severe short-term
threats should not undertake this organizational.
There are also threats to all levels of management,

particularly middle management. The loss of some of
these individuals would mean a tremendous loss in job
knowledge and expertise.
CSSBB 2014

Cross Functional Collaboration

In traditional, functionally designed organizations,
segments of vital activities are captive within and across
many departments. Various department heads are
responsible for the activities within their department,
which allows for good management controls, but no one
owns the overall process and the results.
Functional departments develop strong functional

mindsets and will approach problems differently than
other functional units. It is difficult when departments
speak different “languages,” or have conflicting goals.
A multitude of six sigma writers and trainers have

stressed the importance of cross functional
collaboration in the achievement of six sigma
objectives.
CSSBB 2014

Organizational Change
Business and technological changes have accelerated
at an alarming rate. Customer expectations of products
and services are also increasing. What was delightful to
the customer yesterday, is now expected. A company
must improve its products to meet the customers’ new
expectations and must perpetually improve operations,
processes, costs, cycle time, technology, productivity,
etc. In times of change, everyone is involved in the
change effort. The change agent role is not limited to
the top leaders of the corporation.
The Change Process

The classical model for a change process consists of
three phases: unfreezing, movement, and refreezing.
CSSBB 2014

Change Agents
The change agent is the person or group that acts as the
catalyst and assumes the responsibility for managing
the change process. A manager may act as a sponsor
or patron of the change process. The sponsor is a key
political supporter and may provide the change agent
with funds, staff, and resources. The process to be
changed is defined as the target.
Change agents can be managers or other employees of

the organization. They would be termed internal change
agents. External change agents are outside individuals
who are free from the political restraints of the
organization. They can offer more objective viewpoints
and analysis of the situation at hand. However, outside
change agents will not have the insider’s knowledge of
the organization’s culture, history, procedures, and
personnel.
CSSBB 2014

Internal Change Agents

Advantages
C More knowledge company

C May be more available
C Lower cost
C A known quantity
C More local authority
Disadvantages
C Too close to the problem

C May be part of the problem
C May be biased
C May be unwillingly
CSSBB 2014

External Change Agents

Advantages
C More objective
C More diverse experience
C Has a broader network
C Technically prepared
Disadvantages
C Less company knowledge

C Higher cost
C An unknown quantity
C Longer startup time
C Bad management image
CSSBB 2014

Types of Organizational Change

Organizations generally undergo change in four major
areas: strategy, technology, structure, and personnel.
C Strategic changes occur when the company shifts

its direction and resources toward new businesses
or markets
C Technological changes occur when the company

decides that automation or modernization of key
processes are essential for overall competitiveness
C Structural changes occur when the company

undergoes a management delayering process, or
goes from a functional structure to a product
structure
C Changing the attitudes and behaviors of company

personnel is often undertaken through
organizational development techniques
CSSBB 2014

Large Scale Change

Most change agents, enacting revolutionary
organizational changes, will advise management to
allow 3 to 5 years for the change to take effect. The
chief executive officer is often anxious for results to
appear much quicker than that.
The efforts to remodel an organization and to align

personnel takes years due to the current methods of
imparting the philosophy of change.
A new movement coming from the organizational

development area is the concept of very large groups
coming together to work on a change.
A key element is to make sure that everyone who can

make a decision is in the room.
CSSBB 2014

Resistance to Change
People resist change because they will be asked to do
something that they may be unfamiliar with. They could
also be asked to accept a change which could cause
them a personal loss. The change agent must anticipate
resistance to change and find ways to overcome it.
CSSBB 2014
II. ENTERPRISE-WIDE DEPLOYMENT

QUESTIONS
2.1. Lean and six sigma share in common which of the following
attributes:
a. They both focus on continuous improvement

b. Neither requires top management commitment
c. They both focus on internal customers
d. They both require long learning curves
2.4. A SWOT analysis is an easy way for a company to evaluate itself and
plan a strategy. A key concern is:
a. Lack of resources to do an analysis

b. Situations changing over time
c. Lack of objectivity in the analysis
d. Lack of planned change
2.7. In the hoshin kanri X-matrix what are the targeted organizational
improvements called?
a. Aims
b. Projects
c. Hows
d. Goals
Answers: 2.1. a, 2.4. c, 2.7. a

CSSBB 2014

QUESTIONS
2.15. Which two of the SWOT analysis elements clearly require a firm
to assess the external business environment?
a. Strengths and weaknesses

b. Weaknesses and opportunities
c. Opportunities and threats
d. Threats and strengths
2.17. The concept most closely associated with lean production is:
a. Better quality
b. Faster production
c. Flexible production
d. Elimination of waste
2.18. The most negative management approach to overcome employee

resistance to change is:
a. Mandating the change

b. Negotiating the conditions of the change
c. Providing training on the change
d. Involvement with the change project
Answers: 2.15. c, 2.17. c, 2.18. a

CSSBB 2014

QUESTIONS
2.21. Which luminary is generally recognized as being the creator of

the control chart?
a. Deming
b. Shewhart
c. Harry
d. Ishikawa
2.23. The dissemination of knowledge within a modern organization is

normally coordinated by:
a. IT or MIS department
b. Quality department
c. Engineering department
d. Technology department
2.26. Many tools can be used in either lean or six sigma projects. A
problem solving approach that unifies project follow-up is:
a. SIPOC
b. DOE
c. DMAIC
d. TPM
Answers: 2.21. b, 2.23. a, 2.26. c

CSSBB 2014

QUESTIONS
2.31. Review the following assessment statements:
There is low industry rivalry

There are few regulatory requirements
There is currently a positive growth cycle
What SWOT analysis area is most clearly being portrayed?
a. Strengths
b. Weaknesses
c. Opportunities
d. Threats
2.32. The function of a coach in a six sigma organization is most likely

to be filled by which of the following:
a. Steering committee members

b. Process owners
c. Black belts
d. Green belts
Answers: 2.31. c, 2.32. c

© QUALITY COUNCIL OF INDIANA III-1 (121)
CSSBB 2014
III. PROCESS MANAGEMENT
IT WOULD BE EASY TO DISMISS SIX

SIGMA AS A FAD IF IT WEREN'T FOR
THE CALIBER OF THE RESULTS IT’S
PRODUCING AND THE COMPANIES
ADOPTING IT.
PANDE, NEUMAN, AND CAVANAGH
CSSBB 2014
III. PROCESS MANAGEMENT II.EXTRA

PROCESS MANAGEMENT OVERVIEW
Process Management Overview

Process Management is reviewed in the following topic
areas:
C Process management overview

C Stakeholder impact
C Benchmarking
C Performance measures
C Financial measures
Business process management (BPM) is a fundamental

concept of six sigma. Efforts to improve individual
(local) process components are replaced by systematic
methods to understand, control, and improve (even
optimize) overall business results. These methods have
evolved from the basic tenets of quality and continuous
improvement to address specific business objectives.
CSSBB 2014

Process Management Overview (Cont’d)

BPM is focused on understanding, controlling, and
improving business processes to create value for all
stakeholders. Six sigma builds on classic concepts to
ensure desirable results. Juran defines three principal
dimensions for measuring the quality of this process:
C Effectiveness: how well the output meets customer

needs
C Efficiency: the ability to be effective at least cost
C Adaptability: the ability to remain effective and

efficient in the face of change
This clearly addresses the need for business processes

to provide value to both the customer (effectiveness)
and shareholders (efficiency), now and in the future
(adaptability). Six sigma initiatives strive to manage the
entire business process to maximize these goals.
CSSBB 2014

Process Management Overview (Cont’d)

Most businesses are structured as functional
organizations (vertical units or “silos”) based on
functional groupings such as R&D, product
development, engineering, production, distribution,
marketing, sales, finance, administration, information
technology, etc. Each vertical function also has several
vertical levels from the top executive down.
Products (goods or services) are produced across many

functional boundaries and business levels. Business
process management represents a major advance in
quality improvement thinking by managing the entire
process including those areas between functional
responsibilities.
CSSBB 2014

Process Elements
The SIPOC diagram is a foundation technique for six
sigma management and improvement. An example is
shown below:
Suppliers Inputs Processes Outputs Customers
SIPOC is an acronym for the five major elements in the

diagram:
Supplier: The organization providing resources to

the process of concern
Input: The information, materials, or service

provided
Process: The set of action steps that transforms the

inputs into outputs
Output: The final product or service resulting from

the process
Customer: The person, process, or organization that

receives the output
CSSBB 2014

Process Elements (Continued)

Six sigma relies on the SIPOC model to create, monitor,
and improve closed-loop business systems for process
management, process improvement, and process
design/redesign. SIPOC can help everyone “see” the
business from an overall process perspective by:
C Displaying cross functional activities
C Providing a framework applicable to all process
C Helping maintain the big picture perspective
C Providing methods for adding additional detail

CSSBB 2014

Process Elements (Continued)

When process flow charts are used with the SIPOC
model, business process monitoring, control,
understanding and improvement are greatly enhanced.
To complete the picture, however, it is helpful to
consider one additional factor: the levels of the
business process.
Processes can be viewed as being both comprised of

smaller micro- or sub-processes and constituents of
larger macro-processes. It is often convenient to think
of at least three levels of the overall process because six
sigma methods and procedures change somewhat from
level to level. The three main levels may be described
as business, operations, and process.
Using the SIPOC process model, and understanding the

differences in process levels, will make it easier to
manage the process of business improvement.
CSSBB 2014
III. PROCESS MANAGEMENT II.A

STAKEHOLDER IMPACT
Stakeholder Impact
Businesses have many stakeholders including
stockholders, customers, suppliers, company
management, employees and their families, the
community, and society. Some typical business –
stakeholder relationships are shown below.
SOCIETY
STOCKHOLDERS OR OWNERS
CUSTOMERS
SUPPLIERS
INTERNAL COMPANY
PROCESSES
MANAGEMENT AND EMPLOYEES

CSSBB 2014

STAKEHOLDER IMPACT
Stakeholder Impact (Continued)

Each stakeholder is both supplier and customer,
forming many closed-loop processes that must be
managed, controlled, balanced, and optimized if the
business is to thrive. Communication within the entire
stakeholder community is channeled through internal
company processes. See the figure below.
Reinforcing
Feedback
Supplier Inputs Outputs Customer

Stockholder Investment Profit, Growth Stockholder Y
Customer Orders Goods/Services Customer Value?
Supplier Materials, Machines Additional Orders Supplier N
Business
Employees Commitment Process Pay Employees
Management
Managers Leadership Career Growth Managers
Community Tax Incentives Tax Revenues Community
Society Infrastructure Quality of Life Society
Balancing
Feedback
Stakeholder Relationships
CSSBB 2014

STAKEHOLDER IMPACT
Stakeholder Impact (Continued)

Organizational performance and the related strategic
goals and objectives may be determined for:
C Short-term or long-term emphasis

C Profit
C Cycle times
C Resources
C Marketplace response
Goals may be set for either short-term or long-term

results. American managers, educated by the business
schools in financial matters, have stressed ever
increasing quarterly stockholder dividends. Because of
that, American managers have been criticized for their
short-term outlook. Japanese and European managers
have been willing to take smaller short-term profits to
ensure the long-term growth of their companies.
CSSBB 2014
III. PROCESS MANAGEMENT II.B

BENCHMARKING
Benchmarking
Benchmarking is the process of comparing the current
project, methods, or processes with the best practices
and using this information to drive improvement of
overall company performance. The standard for
comparison may be competitors within the industry, but
is often found in unrelated business segments.
Examples of benchmarking types are explained below:
Process Benchmarking
Process benchmarking focuses on discrete work

processes and operating systems, such as the customer
complaint process, the billing process, or the strategic
planning process. This form of benchmarking seeks to
identify the most effective operating practices from
many companies that perform similar work functions.
CSSBB 2014

BENCHMARKING
Benchmarking (Continued)
Performance Benchmarking
Performance benchmarking enables managers to

assess their competitive positions through product and
service comparisons. This form of benchmarking
usually focuses on elements of price, technical quality,
ancillary product or service features, speed, reliability,
and other performance characteristics.
Project Benchmarking
Benchmarking of project management is easier than

many business processes, because of the opportunities
for selection outside of the group of direct competitors.
Areas such as new product introduction, construction,
or new services are activities common to many types of
organizations. Although the project objectives are
different, the projects will share the same constraint
factors of time, costs, resources, and performance.
CSSBB 2014

BENCHMARKING
Benchmarking (Continued)
Strategic Benchmarking
In general terms, strategic benchmarking examines how

companies compete. Strategic benchmarking is seldom
industry-focused. It moves across industries seeking to
identify the winning strategies that have enabled high-
performing companies to be successful in their
marketplaces.
Benchmarking provides measurements of a company’s

performance compared to the competition, and is an
essential part of six sigma projects during the measure
and analyze stages. It is especially helpful for
evaluating higher level processes at the business and
operational levels where both the measures and their
allowable ranges are often in question.
CSSBB 2014

BENCHMARKING
Benchmarking Sequences
C Determine current practices
C Select the problem area

C Identify key performance factors
C Understand your own processes
C Understand the processes of others
C Select performance criteria based on priorities
C Identify best practices
C Measure the performance within the organization

C Determine the leader(s) in the criteria areas
C Find an appropriate benchmark organization
C Analyze best practices
C Visit the organization as a benchmark partner

C Collect information from the benchmark leader
C Compare current practices with the benchmark
C Note potential improvement areas
CSSBB 2014

BENCHMARKING
Benchmarking Sequences (Continued)

C Model Best Practices
C Drive changes to advance performance

C Extend performance breakthroughs
C Incorporate the new information in the business
C Share results with the benchmark partner
C Seek other benchmarks for further improvement
C Repeat the cycle
Juran presents the following examples of benchmarks

(slightly modified) in an advancing order of attainment:
C The customer specification

C The actual customer desire
C The current competition
C The best in related industries
C The best in the world
CSSBB 2014

BENCHMARKING
Benchmarking Sequences (Continued)

Some companies attempt to achieve a higher
performance level than their benchmark partner. Shown
below is a comparison.
Typical Benchmark Breakthrough Benchmark
t t
Bes Bes
Performance
Performance
r r
You You
om pany om pany
C C
Time Time
In some cases, a benchmarking against the best-in-class
is not possible because:
C The best in world is not known (should be rare)

C There is no related process available (rare)
C The best-in-class is not willing to partner
C The best-in-class is inaccessible
It should be noted that organizations often choose

benchmarking partners who are not best-in-class
because they have identified the wrong partner or
simply picked someone who is handy.
CSSBB 2014
III. PROCESS MANAGEMENT II.C.1

BUSINESS MEASURES/PERFORMANCE MEASURES
Performance Measures
Effective business process management (BPM) requires
an integrated system of metrics in order to achieve the
desired six sigma business improvements. Pearson
describes how this system of metrics might link all three
levels of the enterprise, with the KPOVs (key output
variables) of each level of the process becoming the
KPIVs (key input variables) of the next:
Product Environ-
Develop- ment Price
ment
Quality Market
Process Metrics
Cost Cost Value
Engineering (Customer)
People Quality Quality Availability
Methods Products
Operations
Features Features &
Measure- (Procedures) Profit
ments Availability Throughput Services
Market Share Business
Metrics
ROA (Share-
Cost Cost
Time to Market holder)
Quality Quality
Materials
Suppliers Benefits Features Processes
Machines
Availability Availability
Process Level Process Level Operations Level Business Level

KPIVs KPOVs KPOVs KPOV & CSFs
Operations Level Business Level CSF = Critical success factors

KPIVs KPIVs
CSSBB 2014

Business Level Metrics

Business level metrics are typically financial (external)
and operational (internal) summaries for shareholders
and management.
Business (executive) level metrics comprise summaries

of detailed operations and financial results reported
monthly, quarterly, or annually.
CSSBB 2014

Operations Level Metrics

Six sigma provides new metrics for managing complex
operations. Business effectiveness measures track how
well products are meeting customer needs (external
focus). Breyfogle indicates that they should have a
longer-term perspective and reflect the total variation
that the customer sees.
Operational efficiency measures relate to the cost and

time required to produce the products. They provide
key linkages between detailed process measures and
summary business results, and help identify important
relationships and root causes.
CSSBB 2014

Process Metrics
Detailed process-level metrics include the data from
production people and machinery. This is the
information that operators and supervisors need to run
normal operations. This information is also the subject
of much of the measure, analyze, improve, and control
phases (MAIC) of six sigma, once the improvement
project has been selected and defined.
CSSBB 2014

Other Metrics Considerations

Brown provides helpful recommendations for effective
process performance metrics for the modern enterprise.
C Emphasize the vital few not the trivial many. Overall

business level metrics should be less than 20.
C Metrics should focus on the past, present, and

future.
C Metrics should be linked in a systematic way to

meet the needs of shareholders, customers, and
employees.
C The key to an effective system is to have multiple

metrics, not just one important one. Success is
about balance.
CSSBB 2014

Customer Loyalty Metrics

The specific customer loyalty metrics that should be
considered will vary to a considerable extent on the type
of business that a company is engaged in. Some of the
measures include:
C Are you a repeat customer?

C Are you a returning customer?
C Do you recommend our products?
C How often do you recommend our products?
C How difficult is it to place an order with us?
C How frequently do you buy our products?
C Do you use only our products?
C How likely are you to buy additional materials?
C How satisfied are you with our products?
C How satisfied are you with competing products?
C Are product costs a major decision factor?
C Are shipping costs a major decision factor?
CSSBB 2014

Customer Loyalty Metrics (Continued)

The previous data can be used to consider items like:
C Customer retention rate

C Customer loyalty rate
C User recommendation rate
C User promotional rate
C The attrition (churn) rate*
C The change velocity of customer values
C The direction of customer values
C What customers really think versus what they say
they think
* It should be noted that the customer retention rate

equals 1 - the attrition rate.
CSSBB 2014

The Balanced Scorecard

The balanced scorecard focuses the company’s
attention on four perspectives of the vision and
strategy:
C Financial: How should we appear to our

shareholders? ROI, cash flow, sales
C Internal business process: What must we excel at?

Reduce rework, cycle times, setup times
C Learning and growth: How will we sustain our ability

to change and improve? Surveys, training
C Customers: How should we appear to our

customers? Surveys, market share
Observers and users of the balanced scorecard can look

at the scorecard, see the strategy and goals of the
company, and align themselves accordingly.
CSSBB 2014

BUSINESS MEASURES/FINANCIAL MEASURES
Financial Measures
Harry states simply that six sigma is about making
money. It is about profitability, although improved
quality and efficiency are immediate byproducts. The
financial benefits of six sigma projects are the
measurements that create a link between philosophy
and action. Financial benefits and associated risks are
the factors used to evaluate, prioritize, select, and track
all six sigma projects.
CSSBB 2014

Cost-Benefit Analysis
Project cost-benefit analysis is a comparison to
determine if a project will be (or was) worthwhile. The
analysis is normally performed prior to implementation
of project plans and is based on time-weighted
estimates of costs and predicted value of benefits. The
cost-benefit analysis is used as a management tool to
determine if approval should be given for the project.
CSSBB 2014

Cost-Benefit Analysis (Continued)

The sequence for performing a cost-benefit analysis is:
1. Identify the project benefits
2. Express the benefits in dollar amounts and time
3. Identify the project cost factors
4. Estimate the cost factors in terms of dollar amounts
5. Calculate the net project gain (loss)
6. Decide if the project should be implemented (prior),

or if the project was beneficial (after)
7. If the project is not beneficial using this analysis,

what changes in benefits and costs are possible to
improve the cost-benefit calculation?
CSSBB 2014

Return on Assets (ROA)

Johnson gives an equation for return on assets (ROA)
as:
Net Income
ROA =
Total Assets
Where the net income for a project is the expected

earnings and total assets is the value of the assets
applied to the project.
CSSBB 2014

Return on Investment (ROI)

The following calculation of the return on investment is
widely used:
Net Income
ROI =
Investment
Where net income for a project is the expected earnings

and investment is the value of the project investment.
There are several methods used for evaluating a project

based on the dollar or cash amounts and time periods.
Three common methods are the net present value (NPV),
the internal rate of return (IRR), and the payback period
methods.
CSSBB 2014

Net Present Value (NPV) Method

An equation for net present value (NPV) is:
n
CFt
NPV =  1 + r 
t=0
t
Where n is the number of periods, t is the time period, r

is the per period cost of capital for the organization (also
denoted as i if annual interest rate is used) and CFt is
the cash flow in time period t.
The cash flow for a given period, CFt is calculated as:
CFt = CFB, t - CFC, t
The conversion from an annual percentage rate (APR)

equal to i, to a rate r for a shorter time period, with m
periods per year is:
1
r = 1 + i m -1
If the project NPV is positive, for a given cost of capital,
(r), the project is normally approved.
CSSBB 2014

Internal Rate of Return (IRR) Method

The internal rate of return (IRR) is the interest or
discount rate, i or r, that results in a zero net present
value, NPV = 0, for the project. This is equivalent to
stating that time weighted inflows equal the time
weighted outflows.
n
CFt
NPV = 0 =  1 + r 
t=0
t
The IRR is that value of r which results in NPV being

equal to 0, and is calculated by an iterative process.
Once calculated for a project, the IRR is then compared
with other projects and investment opportunities for the
organization. The projects with the highest IRR are
approved. Most projects have an IRR in the range of 5%
to 25% per year.
CSSBB 2014

Payback Period Method

The payback period is the length of time necessary for
the net cash benefits or inflows to equal the net costs or
outflows. The main advantage of the payback method is
the simplicity of calculation. It is also useful for
comparing projects on the basis of a quick return on
investment. A disadvantage is that cash benefits and
costs beyond the payback period are not included in the
calculations.
Initial  & Incremental  Investment

Payback Period =
Annual (or Monthly) Cash Inflow
A cost-benefit example is provided in the Primer on

pages III - 23/24.
CSSBB 2014

Traditional Cost Concept

Most companies utilize financial reports which compare
actual costs with budgeted costs. The difference is
called a variance and, if significant, may prompt
management action. Departmental budgets may also be
established with the results reported on a monthly or
quarterly basis. These costs are necessary to carry out
the functions of each department, including the control
of product and process quality.
The responsibility for financial control usually rests at

the departmental or plant level. Until the 1950s, few, if
any, companies focused any attention of the costs of
poor quality. These results were hidden among various
labor, material, and miscellaneous expense categories.
CSSBB 2014

Origin of Quality Cost Measurements

In the 1950s and 1960s, some enlightened companies
began to evaluate and report quality costs for the
following reasons:
C Products became increasingly more complex
C The customers’ expectations of products became

more sophisticated
C Customers demanded service after the sale and

expected failure remedy
C Both supplier and customer costs expanded due to

labor and maintenance
C Technical specialists were added to make

improvements
C The management alternatives needed to be in

monetary terms
What resulted was a method of defining and measuring

quality costs and reporting them on a regular basis
(monthly or quarterly).
CSSBB 2014

Origin of Cost Measurements (Cont’d)

The quality cost reports became a vehicle to:
C Determine the status of cost control efforts, and
C Identify opportunities for reducing costs by

systematic improvements
Since the costs of poor quality are high (some

authorities say 15% to 25% of the total cost of sales), the
opportunity for improvement should easily capture the
attention of management and six sigma improvement
teams.
CSSBB 2014

Cost of Poor Quality (COPQ)

The following quality cost definitions are widely used:
Prevention costs: The costs of activities specifically

designed to prevent poor quality in products or
services.
Appraisal costs: The costs associated with measuring,

evaluating, or auditing products or services to ensure
conformance to standards and requirements.
Failure costs: The costs resulting from products or

services not conforming to requirements. Failure costs
are divided into internal and external cost categories:
Internal failure costs: Failure costs which occur

prior to delivery or shipment of the product, or the
furnishing of a service, to the customer.
External failure costs: Failure costs which occur

after shipment of the product, or during or after
furnishing a service, to the customer.
CSSBB 2014

Cost of Poor Quality (Continued)
The relationship of the three levels of product costs are

shown below.
PREVENTION
APPRAISAL
FAILURE
CSSBB 2014

Prevention Costs
The following checklists indicate elements that may be
included in the four major cost categories.
C Applicant screening C Personnel reviews

C Capability studies C Pilot projects
C Controlled storage C Planning
C Design reviews C Procedure reviews
C Education C Procedure writing
C Maintenance C Prototype testing
C Equipment repair C Quality design
C Field testing C Safety reviews
C Fixture design C Surveys
C Forecasting C Time and motion studies
C Housekeeping C Training
C Job descriptions C Vendor evaluation
C Market analysis C Vendor surveys
CSSBB 2014

Appraisal Costs
C Audits C Laboratory testing
C Document checking C Other expense reviews
C Drawing checking C Personnel testing
C Equipment C Procedure checking
calibration C Prototype inspection
C Final inspection C Receiving inspection
C In-process C Shipping inspection
inspection C Testing maintenance
C Inspection and test C Test reporting
Internal Failure Costs

C Accidents C Late time cards
C Accounting errors C Obsolescence
C Design changes C Overpayments
C Employee turnover C Premium freight
C Design changes C Redesign
C Downtime C Reinspection
C Excess interest C Repair
C Excess inventory C Retyping letters
C Excess handling C Rework/sorting
C Excess travel C Scrap
C Failure reviews
CSSBB 2014

External Failure Costs

Note that many of the costs related to internal failure
also appear on this list.
C Bad debts C Liability suits

C Concessions C Obsolescence
C Customer complaint visits C Overpayments
C Customer dissatisfaction C Penalties
C Engineering change notices C Premium freight
C Equipment downtime C Pricing errors
C Excess installation costs C Recalls/repair
C Excess interest expense C Redesign
C Excess inventory C Reinspection
C Excess material handling C Restocking costs
C Excess travel expense C Retesting
C Failure reviews C Returns/rework
C Field service training costs C Scrap/sorting
C Loss of market share C Warranty expenses
CSSBB 2014

Optimum Quality Costs

Some authorities contend that for every dollar spent on
prevention will save approximately seven dollars in
failure costs. Whether this figure can be defended or
not, most companies initially find that they spend an
inadequate amount on prevention activities.
Initially, managers discover that prevention costs are

too low and both internal and external failure costs are
too high. Often, failure costs will exceed the appraisal
costs as well. Even the relationship between internal
and external failure costs may point to needed changes
in planning or product design.
CSSBB 2014

Optimum Quality Costs (Continued)

The interrelationship of quality cost categories varies
widely depending upon the nature of product lines and
processes utilized by a company. In this arena, there
are few absolutes. However, listed below are some
typical ratios for American companies. Note that there
is considerable overlap between the category ranges.
Cost category Percent of total

Prevention 0-5
Appraisal 10 - 50
Internal failure 20 - 40
External failure 20 - 40
The implementation of preventative measures to control

quality often takes a great deal of time. Appraisal
measures are initially undertaken which cause internal
failures to increase but external failures (and total
failures) to decrease.
CSSBB 2014

Quality Cost Improvement Sequence

C Define the company quality goals and objectives:
C The relative position desired among competitors

C The type of long-term quality reputation desired
C Translate the quality goals into requirements:
C Outgoing quality levels required

C Specific types of controls required
C Special tests required
C Estimate capabilities of the current processes

C Develop realistic programs and projects
C Determine the resource requirements
C Set up quality cost categories
C Arrange for accounting to collect the costs
C Ensure accurate figures or reasonable estimates
C Analyze the cost data for improvement candidates
C Use Pareto to isolate vital areas
CSSBB 2014

Quality Cost Comparison Bases

Quality costs should be related to as many different
volume bases as practical. Two or three comparisons
are normal. Some examples are:
Labor bases:
C Total direct labor (worked)
C Standard labor (planned)
Manufacturing cost bases:

C Shop cost of output:
C Direct labor or direct material
C Indirect costs
C Manufacturing cost of output:
C Including the total shop cost of output
Sales bases:
C Net sales billed
C Sales minus direct material
Unit bases:
C Quality costs, dollars per unit of production
C Quality costs related to production
Quality costs are normally summarized monthly

CSSBB 2014

Typical Quality Cost Report

Quality Cost Report for August, 20xx
Dollars ($) Percent of Total
Prevention Costs
Quality Control Administration 5250 2.1
Quality Control Engineering 14600 5.9
Other Quality Planning 1250 0.5
Training 2875 1.2
Total Prevention 23975 9.7
Appraisal Costs
Inspection 55300 22.3
Test 23800 9.6
Vendor Control 1700 0.7
Measurement Control 1950 0.8
Materials Consumed 375 0.2
Product Quality Audits 800 0.3
Total Appraisal 83925 33.8
Internal Failure Costs
Scrap 66500 26.8
Repair, Rework 1900 0.8
Vendor Losses 2500 1.0
Failure Analysis 4000 1.6
Total Internal 74900 30.1
External Failure Costs
Failures - Manufacturing 14500 5.8
Failures - Engineering 7350 3.0
Failures - Sales 4430 1.8
Warranty Charges 31750 12.8
Failure Analysis 7600 3.1
Total External 65630 26.4
Total Quality Costs 248430 100.0
Bases
Direct Labor 94900 8.1
Conversion Cost 476700 40.8
Sales 1169082 100.0
Ratios
Internal Failure Costs to Direct Labor 78.9
Internal Failure Costs to Conversion 15.7
Total Quality Costs to Sales 21.3
CSSBB 2014

Advantages of a Quality Cost System

C Provides a manageable entity and a single overview
C Aligns quality and company goals
C Provides a problem prioritization system
C Distributes costs for maximum profit
C Improves the effective use of resources
C Provides emphasis for doing the job right
C Helps to establish new product processes

CSSBB 2014

Limitations of a Quality Cost System

C Quality cost measurement does not solve quality
problems
C Quality cost reports do not suggest specific actions
C Quality costs are susceptible to short-term

mismanagement
C It is difficult to match effort and accomplishments
C Important costs may be omitted
C Inappropriate costs may be included
C Many costs are susceptible to measurement errors

CSSBB 2014

Quality Cost Pitfalls

C Perfectionism in the numbers
C A number of data validity pitfalls
C Inclusion of non-quality costs
C Implications of reducing quality costs to zero
C Reducing quality costs but increasing total

company costs
C Understatement of quality costs

CSSBB 2014

QUESTIONS
3.2. What key step would follow an evaluation of current practices

against a benchmark?
a. Identifying key performance factors

b. Selecting performance criteria based on priorities
c. Determining a leader in a critical performance area
d. Undertaking significant changes to advance performance
3.6. The percentages of total quality cost are distributed as follows:

prevention 12%; appraisal 28%; internal failure 40%; and external
failure 20%. One would conclude:
a. More money should be invested in prevention

b. Expenditures for failure are excessive
c. The amount spent for appraisal seems about right
d. Nothing
3.11. One may say that business metrics, as opposed to process or

operational metrics, would be most concerned with:
a. Quality and costs

b. Throughput and costs
c. Quality and features
d. Profit and market share
Answers: 3.2. d, 3.6. d, 3.11. d

CSSBB 2014

QUESTIONS
3.15. The best reason for instituting a quality cost system is:
a. To satisfy ISO 9001 requirements

b. To achieve a proper balance of prevention, appraisal, and failure
costs
c. To identify opportunities for improvement
d. To benchmark major competitors
3.18. A calculation for ROI requires what two components?
a. Net income and investment costs

b. Cash flow and interest rates
c. Net income and total assets
d. Investment costs and cash flow
3.23. If the project internal rate of return is estimated to be 8%, the

project will be approved when:
a. The discount rate of 8% yields an NPV of less than 0

b. The company cost of capital is 9% or higher
c. Funds are limited and another project will yield 10%
d. Funds are unlimited and another project will yield 11%
Answers: 3.15. c, 3.18. a, 3.23. d

© QUALITY COUNCIL OF INDIANA IV-1 (171)
CSSBB 2014
IV. TEAM MANAGEMENT III.A.1
IT’S EASY TO GET THE PLAYERS.

GETTIN’EM TO PLAY TOGETHER,
THAT’S THE HARD PART.
CASEY STENGEL
CSSBB 2014
Team Management
Team Management is presented in the following topic
areas:
C Team formation
C Team facilitation
C Team dynamics
C Team training
C Problem solving methodologies
CSSBB 2014

TEAM FORMATION/TEAM TYPES
Team Formation
As a prerequisite to team formation, team members
must:
C Have a reason to work together

C Accept an interdependent relationship
C Commit to team values
Management supports the team process by:
C Ensuring a constancy of purpose

C Reinforcing positive results
C Sharing business results
C Giving people a sense of mission
C Developing a realistic and integrated plan
C Providing direction and support
Management as part of some mechanism such as a

steering committee should provide relief from potential
team constraints.
CSSBB 2014

Types of Teams
The following types of teams are used by industries
throughout the world today:
Six Sigma Teams
The structure and functional roles of six sigma teams

closely follow the description of project and ad hoc
teams that follow, with the addition of black and master
black belt support.
Improvement Teams
A group belonging to any department chooses to solve

a quality/productivity problem. It will continue until a
reasonable solution is found and implemented.
CSSBB 2014

Types of Teams (Continued)

Process Improvement Teams
For a process improvement team, employees may be

drawn from more than one department to look into the
flow of material and semi-finished goods required to
streamline the process.
Project Teams/Task Forces/Ad Hoc Teams
Members are selected based on their experience and

directed by management to look into specific areas such
as the modernization of a piece of equipment or solution
to a customer complaint. These teams are generally ad
hoc and disband upon the completion of their
assignments. Team membership can be all
management, all work area, or a composite of the two.
Cellular Teams
Cellular teams are a variant of natural work teams. The

name derives from the work cell arrangement in which
a number of employees either fabricate or assemble
parts. These teams can be led by a supervisor or may
be self-directed.
CSSBB 2014


Self-Directed Teams
This type of work or project team operates with minimal

day-to-day direction from management. Self-directed
teams are asked to accomplish objectives within time
frames that are truly stretch objectives. Management
must give the team the maximum latitude possible for
achieving their objectives.
Cross Functional Teams
Cross functional teams are made up of individuals who

represent different departments or functional areas.
Individuals who represent a department or functional
area should be subject matter experts. The thoughtful
selection of the members is an important aspect of
building an effective team.
Solutions designed with the active participation of

affected departments tend to be technically superior and
accepted more readily by those who implement them.
CSSBB 2014


Natural Work Team Organization
Leadership in natural work teams is usually given to the

area supervisor. Members come from the supervisor’s
work force. Outside members from specialist
organizations can be included in the membership, either
as active members or as contributing guests. Often, a
facilitator is another important person in this team
organizational structure.
Quality Circles
The concept of circles originated in Japan after WW II.

They were so successful in Japan that many managers
in the United States tried to duplicate them. The circle
is a means of allowing and encouraging people on the
production floor to participate in decisions that will
improve quality and/or reduce manufacturing costs.
CSSBB 2014


Quality Teams
The fundamental purpose of establishing quality teams

is to improve the internal efficiencies of the company
and both internal and external products and service
quality. This is done through the efforts of the team
members to improve quality, methods, and/or
productivity.
Virtual Teams
Since many companies are international, there is a

growing need for a variety of virtual project or ad hoc
teams to address issues such as product quality,
service instructions, delivery issues, and technical
requirements.
CSSBB 2014

Synopsis of Team Applications

Team Type Structure Best Applications
Improvement May be 8 to 10 Can work on quality or productivity
Teams members from a issues. A process improvement team can
single department. consist of multi-department membership
and focus on process flow and product
issues.
Quality May be 8 to 10 May initially work on quality topics or
Teams members from a overall department performance. Can
single department. evolve into self directed teams.
Project Can have broad or Works on specific projects such as the
Teams specific member installation of a conveyor system. Can
selection. May be all also focus on material related items like
or part management. an improved inventory control system.
Usually disbands upon the completion of
a project.
Six Generally 8 to 12 Works on specific process or customer
Sigma members with Black based projects of importance. Usually
Teams Belt or Master Black disbands upon project completion.
Belt support
Cross 8 to 12 members Members are carefully selected.
Functional from different areas, Knowledgeable people are required. Very
Teams departments, or similar to project teams. Tends to deal
disciplines. more with policies, practices and
operations.
Self Directed 6 to 15 members. Requires considerable training and
Teams Generally a natural exposure. Can be given objectives or
work area team. May develop their own. Some companies
need staff support. select people with co-operative skills to
help with success.
CSSBB 2014

TEAM FORMATION/TEAM ROLES
Team Roles
Described below are a variety of major team member
roles and responsibilities as practiced in both six sigma
and other team arrangements.
The Team Member Role

Each team member is responsible for:
C Participating in training to become effective

C Attending team meetings, as required
C Completing assignments between meetings
C Participating actively during meetings
C Encouraging active participation by other members
C Benefitting from the perspectives of others
C Applying the steps of the improvement process
CSSBB 2014

The Leader Role

Some teams have both leaders and facilitators. This is
common for manufacturing line teams. As a general
rule, the team leader focuses on the team product (the
results) and the facilitator is most concerned with the
team process. Within the six sigma framework, the team
leader is normally trained both as a facilitator and black
belt.
CSSBB 2014

The Leader Role (Continued)

The leader will:
C Provide direction and suggest assignments

C Act as a communication hub
C Act as a liaison with management
C Handle administrative details
C Ensure that individual needs are considered
C Conduct meetings
C Assess group progress
C Take the steps necessary to ensure success
C Encourage participation
C Be accepting and tolerant of mistakes
C Work with, not over participants
C Be a good listener
The leader’s role is not to “boss” the team but to ensure

implementation of the team mission and charter.
CSSBB 2014

The Team Facilitator Role

As noted earlier, the team leader in six sigma and other
team arrangements is often the facilitator. However,
many companies find facilitators useful both for team
start-ups and for a variety of other team arrangements.
The team leader and/or facilitator must understand
group dynamics and how a group moves through
developmental stages (forming, storming, norming, and
performing).
CSSBB 2014

The Team Facilitator Role (Continued)

Facilitators are useful in assisting a group in the
following ways:
C Identifying members that need training

C Avoiding team impasses before task completion
C Providing feedback on group effectiveness
C Summarizing points made by the group
C Balancing group member activity
C Helping to secure resources that the team needs
C Providing an outside neutral perspective
C Clarifying points of view on issues
C Keeping the team on track with the process
C Helping with interpersonal difficulties that may arise
C Focusing on progress
C Assessing the change process
C Assessing cultural barriers (attitudes, personalities)
C Assessing group accomplishments
C Asking for feelings on sensitive issues
C Helping the leader to do his/her job more easily
C Coaching the leader and participants
CSSBB 2014

The Team Facilitator Role (Continued)

If there is no facilitator, the team leader, an assigned
black belt, or a coach must assume many of the above
duties.
The facilitator must avoid:
C Being judgmental of team members or their ideas

C Taking sides
C Becoming caught-up in the subject matter
C Dominating group discussions
C Solving a problem or giving an answer
C Making suggestions on the task
CSSBB 2014

The Recorder Role

The team recorder/secretary is normally a full-fledged
team member. The recorder maintains the team’s
minutes and agendas. He/she also distributes relevant
materials to team members. The recorder:
C May or may not participate as a member

C Takes clear notes including project responsibilities
C Publishes and distributes the minutes
C May ask for clarification of issues (for the record)
The Timekeeper Role

The timekeepers role is an optional responsibility. This
function sometimes becomes the responsibility of the
facilitator, when a facilitator is assigned to a team. The
timekeeper:
C Advises the team of the remaining meeting time

C Enforces any time “norms” of the team
CSSBB 2014

The Process Owner Role

A process owner coordinates process improvement
activities and monitors progress on a regular basis.
Process owners work with black belts to improve the
processes for which they are responsible. This upper
level manager should:
C Be comfortable with the team’s capabilities

C Believe in the team’s objectives
C Support team members with resources
C Share information with the team
C Understand the team’s mission
C Participate in project reviews
C Align personal goals with the team’s goals
C Be knowledgeable of six sigma core elements
CSSBB 2014

TEAM FORMATION/MEMBER SELECTION
Selecting Team Members

When selecting a team, upper management identifies
those parts of the organization that are associated most
closely with the problem. There are four places to look:
C Where the problem is observed or the pain is felt

C Where causes of the problem might be found
C Among those with special knowledge or skill
C In areas that can help develop remedies
Often a cross functional team is assembled to

accomplish significant results in a short period of time.
What is required are people who believe that two or
more minds are better than one, and who will contribute
a diversity of perspective, experience and knowledge.
CSSBB 2014

Adding New Team Members

Care must be taken when adding new people to existing
teams. The rule is to not impose an individual on a
team. This can be handled by involving the entire team
in the selection process. When the team has a
significant role in deciding on any new team member,
the team will be much more committed to making sure
the decision was the right decision.
Removing Team Members

Sometimes a team member may need to be taken off the
team. There are any number of reasons why this could
occur. Perhaps one of the members lacks the required
skills and shows little interest in developing them.
Personality conflicts may exist between team members.
Perhaps a team member is stretched or stressed by
other projects or personal problems, and can’t keep
his/her commitments to the team.
The result is a very delicate situation for the team leader

or sponsoring manager. Conversations should center
on what’s expected and what’s not happening that
needs to happen.
CSSBB 2014

Team Size
A team can consist of members from only one area, or
it can be made up of a group of representatives from
different parts of the organization. Each person may be
a subject matter expert who understands the processes
and activities at issue. It is usually impractical to
include every person who could be involved.
Conventional wisdom is that teams over 20 people,

some think over 15, become too unwieldy and lose the
active participation of all team members. Teams of 4
people or less may not generate enough ideas. A major
change management principle embraces the notion that
people will more readily accept and support a change, if
they are included in the development of the solution.
CSSBB 2014

Team Diversity
To achieve optimum performance, a team often needs
diversity in the orientation of its individual team
members. Some team members are needed who are
primarily oriented towards task and target date
accomplishment. Other team members will be needed
who hold process, planning, organization and methods
in the highest regard. Teams also need members who
nurture, encourage and communicate well. Teams need
some members who are creative and innovative.
CSSBB 2014
IV. TEAM MANAGEMENT III.A.2, 3

Belbin Team Roles

Meredith Belbin describes a pattern of behavior that
characterizes relationships in team performance.
People often have a mixture of roles.
Action-oriented roles:
C Shapers
C Implementers
C Completers
People-oriented roles:
C Coordinators
C Team workers
C Investigators
Problem solving roles:
C Plants (Innovators)
C Evaluators
C Specialists
CSSBB 2014
IV. TEAM MANAGEMENT III.A.2, 3

Belbin Team Roles (Continued)

Listed below are typical strengths and weaknesses.
Role Strengths Weaknesses

Shaper Shapers bring the drive and courage to Shapers offend people and will
overcome obstacles. The shaper is display aggression in the
committed to achieving ends. pursuit of goals. Two or three
shapers in a group can lead to
conflict.
Implementer Implementers turn ideas into practical Implementers are conservative,

actions. They tend to work for the team in a inflexible, and slow to respond
practical and realistic way. to new possibilities.
Completer Finishers finds errors and omissions. They Finishers worry unduly and are
Finisher deliver their contributions on time and pay reluctant to delegate. They tend
attention to details. to be over anxious.
Coordinator The coordinator is a positive thinker who Coordinators can be seen as

supports goal attainment and effort in others. manipulative. They might not
They clarify goals and delegate well. stand out in a team.
Team Team workers tend to keep team spirit up and They tend to be indecisive in
worker allow other members to contribute. They moments of crisis and are
bring cooperation and diplomacy to a team. reluctant to offend.
Resource The resource investigator explores They are over-optimistic and

Investigator opportunities and develops contacts. They may lose interest quickly. They
are good negotiators. are not the sources of original
ideas.
Plant A plant brings creativity, ideas, and Plants ignore incidentals and
Innovator imagination to a team. They can solve may be too preoccupied to
difficult problems. communicate effectively.
Monitor The monitor evaluator is not deflected by The evaluator may appear dry,
Evaluator emotional arguments. They are boring, and overcritical. They
serious-minded and bring objectivity and are not good at inspiring others.
judgment to options.
Specialist Specialists bring dedication and initiative. They may contribute only on a
They provide needed knowledge and narrow front and dwell on
technical skills. technicalities.
CSSBB 2014

TEAM FORMATION/SUCCESS FACTORS
Team Success Factors

A participative style of management is the best
approach to ensure employee involvement in the
improvement process. Today’s workforce has higher
educational levels and are eager to participate in the
decision-making process that affects them. There is no
better way of motivating employees than to provide
them with challenging jobs which make use of their
talents and abilities.
CSSBB 2014

Company Team Benefits

Improvement teams:
C Can usually tackle larger issues than individuals
C Can build a fuller understanding of the process
C Can have access to the technical skills and

knowledge of all team members
C Can rely on the mutual support and cooperation that

arises among team members
CSSBB 2014

Team Member Benefits

Teamwork offers the following benefits:
C An opportunity for greater understanding of the

issues affecting their work
C A chance to be creative and share ideas
C The opportunity to forge stronger working

relationships with colleagues
C The opportunity to learn new skills
C A chance to work on a project with the support and

interest of management
C The satisfaction of solving a chronic problem
C The opportunity to please customers, increase

revenues, and reduce costs
CSSBB 2014

Team Resources
Resources are time, talent, money, information, and
materials. The development of productive teams will
use considerable resources. Management must
optimize the resources available to teams. The team
charter is the best place to establish the team’s
expectations concerning available resources.
CSSBB 2014

Management Support
Management must give more than passive team support.
This means that management, especially mid-
management, must be educated to the degree that they
are enthusiastic about the team concept. In order for
teams to be successful, management must recognize
that there will be additional work created by their efforts.
Leaders, facilitators, and team members should be
thoroughly trained in six sigma and other improvement
techniques.
There are reasonable arguments that can be expressed

either for or against teams. The important questions
that need answers are: (1) Does the company have the
proper environment in which teams can survive and
thrive? and (2) Does management fully comprehend the
value of teams, to make them work?
CSSBB 2014

The Steering Committee Role

Establishing a steering committee is a logical first step
when an organization launches a six sigma
improvement initiative. The steering committee is
usually composed of upper management. In some
companies, middle management and hourly employees
are also represented.
Some of the steering committee key roles include:
C Setting goals
C Identifying projects
C Selecting teams
C Supporting project teams
C Monitoring progress
CSSBB 2014

Team Building Activities

Three key characteristics of effective team building are
mutual trust, respect, and support. Team members
encourage each other to stretch beyond their comfort
zone by offering advice or assistance when asked or
when it is obvious that the fellow team member needs it.
Teams must strive to improve the quality of their

teamwork as well as the quality of their output.
Activities that improvement teams may undertake
include:
C Awareness and education

C Data collection and presentation
C Problem solving and decision-making
C Organizing breakthroughs
CSSBB 2014

Initial Project Selection

Management may define the team project. When the
team chooses the improvement project, then the
following considerations are important:
C It should have appeal to members and management

C It should be fairly simple - but not trivial
C It should show quick benefits (3 to 4 months)
C It should be within the group’s control
C It should consider time and resource constraints
The two major activities are project resolution and

learning teamwork.
CSSBB 2014

The Problem Must be Properly Defined

C Problem statements are often fuzzy. Consider the
following examples:
C Poor communications
C Excessive downtime
C Low recovery
C Too much scrap
C The problem must be defined. There is a tendency

to work on a down stream symptom of an upstream
problem.
C A problem is the gap between current results and

desired results.
C A clearly defined problem statement that is

measurable should be the initial product.
C Consider the team to be working on a problem that

is scheduled for solution.
CSSBB 2014
IV. TEAM MANAGEMENT III.B.1

TEAM FACILITATION/MOTIVATIONAL TECHNIQUES
Team Motivation
The team process itself can be a highly effective,
people-building, potential-releasing, goal-achieving
social system that is characterized by:
C A climate of high support

C An open communication process
C Organizational goal achievement
C Commitment
C Individual achievement
C Creative problem-solving
The fundamental purpose of establishing teams is to

improve the internal and external efficiencies of the
company.
CSSBB 2014

Team Motivation (Continued)

If teams are properly functioning, they will:
C Improve employee morale

C Remove areas of conflict
C Develop creative skills of members
C Improve communication and leadership skills
C Develop problem solving techniques
C Improve both management and member attitudes
C Indicate that management will listen
C Demonstrate that employees have good ideas
C Improve management/employee relationships
CSSBB 2014

Motivational Techniques
Probably the most challenging management
responsibility is how to both sustain and increase
internal motivation in the work group. Management
should recognize that people do have certain needs in
common, which may often be met in basically the same
way. For example, two such common needs are that of
being needed and of being treated with dignity and
respect.
CSSBB 2014

The Hawthorne Studies

In 1924, Western Electric Company started research on
individual productivity at its Chicago Hawthorne facility.
They allowed Elton Mayo of Harvard to conduct studies
on the effects of worker fatigue and output. Two
important points were revealed in this study:
1. Group behavior has a powerful influence upon

individual members.
2. The work group is a social group which fulfills

certain human needs.
One of the primary objectives of the study was to

determine the effect of illumination on productivity. The
study actually revealed a far more important and
insightful factor -- so long as people are treated as
human beings, giving due consideration to individual
needs, they tend to cooperate in increasing productivity.
CSSBB 2014

Abraham Maslow
A major theory on the needs and motivation of an
individual is based on Abraham Maslow’s theory of
human needs based on research conducted during
World War II. He stated that there are five levels of
human needs, and they are listed below from the highest
to the lowest:
C Self-actualization: Maximum achievement

C Esteem: Respect, prestige, and recognition
C Social: Love, affection, relationships
C Safety: Security, protection, and stability
C Physiological: Basic needs; food, water, housing
CSSBB 2014

Abraham Maslow (Continued)

Maslow’s theory affirmed the view that individuals are
motivated to lower-order needs until these are satisfied
and then higher-order needs must be met.
GENERAL ORGANIZATIONAL
EXAMPLES EXAMPLES
ACHIEVEMENT SELF
ACTUALIZATION CHALLENGING
NEEDS JOB
STATUS JOB
ESTEEM NEEDS
TITLE
FRIENDS
FRIENDSHIP SOCIAL NEEDS IN WORK
GROUP
STABILITY SECURITY NEEDS PENSION

PLAN
SHELTER PHYSIOLOGICAL NEEDS BASE

FOOD SALARY
Maslow’s Hierarchy of Needs

CSSBB 2014

Douglas McGregor
During the mid 1950s, Douglas McGregor began to
introduce new theories, Theory X and Theory Y, to his
students at the MIT. McGregor contended that
traditional management practices were rooted in certain
basic negative assumptions about people (Theory X):
C Are fundamentally lazy

C Avoid responsibility
C Lack integrity
C Are not very bright
C Cannot direct their own behavior
C Are indifferent to organizational needs
C Prefer to be directed by others
C Are not interested in achievement
CSSBB 2014

Douglas McGregor (Continued)

By contrast, Theory Y contains the following important
points:
C The expenditure of physical effort in work is as

natural as play or rest
C The threat of punishment is not the only means to

achieve objectives
C Man can exercise self-direction and self-control
C Commitment to objectives is a function of the

associated rewards
C The average human can learn to accept and seek

responsibility
C Imagination, ingenuity, and creativity are widely, not

narrowly, distributed
C Only a fraction of the intellectual potential of

workers is utilized
CSSBB 2014

Frederick W. Herzberg
Frederick W. Herzberg and his colleagues at Western
Reserve Institute conducted studies on the motivation
to work. He proposed that motivation can be divided
into two factors, which have been referred to by a
variety of names such as:
C Dissatisfiers and satisfiers, or

C Maintenance factors and motivators, or
C Hygiene factors and motivators
The dissatisfiers or hygiene factors do not provide

strong motivation, but do cause dissatisfaction if they
are not present. On the other hand, satisfiers do provide
strong motivation and satisfaction when they are
present. For the manager, this is a reminder that salary
is not a motivator, but lack of an appropriate salary can
dissatisfy an employee.
CSSBB 2014

Motivation and Hygiene Factors
Hygiene Factors Motivation Factors

Supervision Achievement
Working conditions Recognition
Relationships The work itself
Pay and security Responsibility
Company policies Advancement
CSSBB 2014

Job Enrichment
Herzberg’s theory of satisfiers and hygiene factors for
understanding job satisfaction and performance are
linked to job enrichment. In this job design, jobs that
are boring or lack content are given elements that will
satisfy and enrich the employee. The following job
characteristics are important.
C Skill variety: a variety of tasks, and types of skills

C Task identity: an ability to do the complete job
C Task significance: the impact of the job on the firm
C Autonomy: freedom from supervision
C Job feedback: Seeing the direct results of work
CSSBB 2014

Employee Empowerment
Organizations have been searching for higher
performance for many years. French provides the
following definition of empowerment:
“To empower is to give someone power, which is

done by giving individuals the authority to make
decisions, to contribute their ideas, to exert influence,
and to be responsible.”
Management can do a better job in erasing barriers to

empowerment through:
C Providing more support

C Providing training
C Being role models
C Managing by walking around
C Being facilitators
C Responding quickly to recommendations
C Recognizing employee accomplishments
CSSBB 2014

Organizational Empowerment
Management has the responsibility of setting the tone
for organizational empowerment. The purpose of this
action is to engage the entire workforce in the activity of
making things better. The need for continuous
improvement includes: productivity, cost containment,
product and service quality, outstanding customer
treatment, and respect for all employees of the
company.
CSSBB 2014

Organizational Empowerment (Cont’d)

Organizational empowerment generally consists of
steps or stages. Very few companies would attempt to
advance from employee awareness directly to employee
ownership. Refer to the Figure below.
100
OWNERSHIP
90
80
COMMITMENT
70
60
IMPACT SUPPORT
50
40
UNDERSTANDING
30
20
AWARENESS PROBLEM SELF
10 OPINION SUGGESTION SOLVING REGULATING
SYSTEMS TEAMS WORK TEAMS
SURVEYS
0
INFORMATION EMPLOYEE TASK FORCE KAIZEN SELF MANAGED
SHARING TRAINING COMMITTEES CI TEAMS PLANT DESIGN
APPROACHES
Obviously, six sigma improvement teams would easily

fit into the support/commitment levels of empowerment.
CSSBB 2014

Management Actions
Without consideration of the quality of work life (by
enriching it, enlarging it, or other methods) the
individual worker will probably not be satisfied. Efforts
at higher levels of empowerment may be doomed.
Among the various concepts that management can use
are:
C Getting others involved in their own assignments

C Encouraging others to obtain the results of work
C Providing an environment of cooperation
C Sharing the ownership of results
C Encouraging others to take workplace initiative
C Allowing others to make decisions
C Letting others implement their ideas
C Recognizing successes and contributors
CSSBB 2014

Management Actions (Continued)

Kinni suggests that management cannot directly cause
an employee to become motivated. The best that they
can do is follow the concepts listed below to create an
environment for individuals to motivate themselves:
1. Know thyself
2. Know your employees
3. Establish a positive attitude
4. Share the goals
5. Monitor progress
6. Develop interesting work
7. Communicate effectively
8. Celebrate success
The above concepts work at all organizational levels.

CSSBB 2014

TEAM FACILITATION/TEAM STAGES
Team Stages
Most teams go through four development stages before
they become productive: forming, storming, norming,
and performing.
Forming
Forming is the beginning of team life. Expectations are

unclear. Members test the water. Interactions are
superficial. This is the honeymoon stage. When a team
forms, its members typically start out by exploring the
boundaries of acceptable group behavior.
Storming
The second phase consists of conflict and resistance to

the group’s task and structure. There are healthy and
unhealthy types of storming. Conflict often occurs in
the following major areas: authority issues, vision and
values dissonance, and personality and cultural
differences. However, if dealt with appropriately, these
stumbling blocks can be turned into performance
enhancers later. This is the most difficult stage for any
team to work through.
CSSBB 2014

Team Stages (Continued)

Norming
During the third phase, a sense of group cohesion

develops. Team members use more energy on data
collection and analysis as they begin to test theories
and identify root causes. Members accept other team
members and develop norms for resolving conflicts,
making decisions, and completing assignments.
Performing
This is the payoff stage. The group has developed its

relationships, structure, and purpose. The team begins
to tackle the tasks at hand. The team begins to work
effectively and cohesively. During this stage, the team
may still have its ups and downs. Occasionally, feelings
that surfaced during the storming stage may recur.
CSSBB 2014


A graphical display of the performance of teams as they
advance through the team evolutionary stages is shown
below.
Performing
Members:
show maturity
focus on the process
achieve goals
operate smoothly
Norming
Members:
cooperate
talk things out
focus on objectives
Performance
have fewer conflicts

Storming
Members:
have confrontation
think individually
are learning roles
have divided loyalties
Forming
Members are:
inexperienced
excited
anxious
proud
Time
CSSBB 2014


Stage 1 FORMING Stage 2 STORMING
BEHAVIORS BEHAVIORS
Lack of task focus Problem solving is superficial
Difficulty in defining problems There is petty arguing
Uneven participation Hidden agendas and cliques emerge
Ineffective decision-making Decisions don't come easily
Resistance to team building Plenty of uncertainties persist
FEELINGS FEELINGS
Excitement, anticipation, and pride Resistance is seen
Shaky alliance to the team Individual attitudes vary widely
Suspicion, fear, and anxiety Anger and jealousy abound
Roles and responsibilities are unclear
HOW TO IMPROVE
HOW TO IMPROVE Follow a problem solving format
Take time to become acquainted Clearly define roles
Establish mission and goals Debrief meetings for content and process
Establish team ground rules Deal openly with conflict
Add structure to meetings Work to expose hidden agendas
Train members in team concepts Focus team on goals
Encourage equal participation
Stage 3 NORMING Stage 4 PERFORMING
BEHAVIORS BEHAVIORS
Attitudes improve Members to work through problems
Trust and commitment grow Members manage the group process
Some goals and objectives are achieved There is creativity and informality
Feedback becomes regular and objective High levels of unity and spirit are seen
Conflicts are dealt with and resolved Close bonds form
The leader receives respect
Some leadership is shared by the team FEELINGS
Self improvement is noted
FEELINGS Acceptance of weakness
Comfort with giving feedback Appreciation of strengths
Comfort with receiving feedback Satisfaction with team progress
Sense of cohesion and spirit Team knows clearly what it is doing
Friendlier and more open exchanges
HOW TO IMPROVE
HOW TO IMPROVE Promote openness
Evaluate team performance Permit more self direction
Periodic summaries of progress Establish new goals
Create ties outside of the team
CSSBB 2014

Team Life Cycle Characteristics

Shown below is another representation of team
development stages.
Effectiveness Optimize
Develop
Build
Time
CSSBB 2014

TEAM FACILITATION/TEAM COMMUNICATIONS
Team Communications
For any organization to be successful there must be
effective internal and external communications. This
concept certainly extends to team communications as
well. Among other items, management must relay
information regarding:
C The vision and mission of the company

C Team-based policies and procedures
C Team performance feedback
C The extent of project support
C Satisfaction or dissatisfaction with performance
C Any shifts in priorities
Teams are responsible for communicating:
C The project’s progress and current status

C Any requests for any change in resources
C Items that may impact other teams or departments
C Any significant shifts in project direction
C Any pertinent stakeholder information
C Any requests for assistance
CSSBB 2014

Team Communications (Continued)

Oral communication can take a variety of forms such as
the telephone, face-to-face meetings, formal briefings,
and the internet. Examples of written communications
include letters, reports, computer messages and e-
mails. The written forms can be described as one-way
channels. Face-to-face meetings generally allow for
immediate feedback (two-way communications).
The skillful use of questioning is of great value. The

following key questions should be asked:
C Why? Ask “why” five times.

C What is the purpose?
C What will it take to accomplish the project?
C Who will care or benefit if the project is completed?
C What data is available?
C Where did the data come from?
CSSBB 2014


Some additional ideas for the art of asking questions:
C Avoid leading questions: let the group or individual

draw their own conclusion
C Phrase questions in a positive manner
C Prepare questions in advance whenever possible
The use of open-ended questions will allow for some

discussion and probing rather than just a simple “yes”
or “no” answer.
CSSBB 2014


Listening, the other half of the communication concept,
often receives far too little attention. Verbal information
can often be very difficult to understand, even when
active listening takes place.
Tips for good listening include:
C Put the message sender at ease

C Show that you want to listen
C Remove listening distractions
C Empathize with the speaker
C Be patient with your own response
C Hold your own temper
C Avoid arguing and criticism
C Ask questions
C Stop talking
Many individuals would rather hear themselves speak

than listen to another person. The good news is
listening skills can be developed.
CSSBB 2014

A3 Report
The A3 report is a concise summary of a project
described in five or six parts on one piece of paper. The
name A3 relates to the metric size paper used for the
report, approximately equivalent to an 11" x 17" sheet
of paper. This can be the final report presented to
management for approval of a project. The process
removes the fluff and flashy showmanship from the
presentation. It may only take 3 to 5 minutes for the
speaker to present the business case. A project is thus
judged on its merits.
CSSBB 2014

A3 Report (Continued)
There are four basic types of story lines used in the A3
report. Each type is similar in structure, but the
purposes are different. The basic types are:
C Proposal stories: a business case for a new venture
C Problem solving stories: an analysis and

implementation report
C Status stories: a project status report
C Information stories: a transmittal of information

CSSBB 2014

A3 Report (Continued)
The report format fits the problem solving stages onto
an 11" x 17" sheet. For example, the problem definition
and analysis stages would appear on the left side of the
paper, while the action plans, results of activities, and
future steps will appear on the right side. The following
are some key concepts for A3 report writing:
C Use concise wording
C Use charts to illustrate points
C Restrict statements to 3 or 4 bullets per section
C Provide substance, not fluff
A hypothetical A3 report is presented in the Primer.

There is certainly flexibility in the format. The concept
is to present a summary of the project in a concise,
clear, and understandable way.
CSSBB 2014

Team Leadership Models

The four basic leadership styles are:
C Directing
C Supporting
C Coaching
C Delegating
Directing
The directive approach is characterized by:
C An autocratic approach
C Typically, one way communication
C Directive as to who, what, when, and where
C Tight supervision of subordinates/staff
C Assigned job priorities
C Preplanned work
C Step by step instructions
C Close follow-up on assignments
CSSBB 2014

Team Leadership Models (Continued)

Supporting
The supporting approach exemplifies supporting

behavior such as:
C Listening to the suggestions of subordinates

C Providing very little directive behavior
C Building the confidence of others
C Facilitating interactions with other departments
C Helping staff reach their own solutions
C Encouraging staff risk taking
C Asking questions that expand the thinking of others
CSSBB 2014


Coaching
The coaching approach is one in which there are

components of both direction and support. This
approach is typified by:
C Engaging in two way communications

C Teaching others how to evaluate their own work
C Reinforcing correct initiatives
C Encouraging prudent risk taking
C Practicing good listening skills
C Supporting good ideas
C Making the final decision
CSSBB 2014


Delegating
The delegating approach is principally the opposite of

the direct approach. This management style:
C Allows subordinates to make day to day decisions

C Encourages problem solving at the lowest level
C Demonstrates little (if any) directive behavior
C Delegates most decisions to others
C Encourages subordinates to get on with the work
C Provides consultation if the need arises
CSSBB 2014


The four leadership styles can be summarized as
follows:
A - Directing I’ll decide I’ll tell you

B - Coaching We’ll talk I’ll decide
C - Supporting We’ll talk We’ll decide
D - Delegating You decide You tell me
A manager does not use one style for all occasions. The
task at hand will impact the required style.
CSSBB 2014

(high)
C. Supporting B. Coaching
Leader makes decisions with team Leader explains decisions and
members and helps with goal solicits suggestions from team
Supportive
Behavior
accomplishment if needed member but still directs goals
D. Delegating A. Directing
Leader turns over decisions to the Leader gives specific instructions

group and closely supervises the group
(Low) Directive Behavior (high)
The Four Leadership Styles

CSSBB 2014
IV. TEAM MANAGEMENT III.C.1

TEAM DYNAMICS/GROUP BEHAVIORS
Group Behavior
A variety of group dynamic challenges and techniques
to overcome them are presented in this Section element.
Groupthink
One aspect of group cohesiveness can work to a team’s
disadvantage. Members of highly cohesive groups may
publicly agree with actual or suggested courses of
action, while privately having serious doubts about
them. Strong feelings of group loyalty can make it hard
for members to criticize and evaluate other’s ideas and
suggestions. Desiring to hold the group together and
avoid disagreements may lead to poor decision-making.
Psychologist Irving Janis calls this phenomenon
“groupthink.”
CSSBB 2014

Groupthink (Continued)
Eight symptoms of groupthink include:
1. Illusion of invulnerability
2. Belief in inherent morality of group
3. Collective rationalization
4. Out-group stereotypes
5. Self-censorship
6. Illusion of unanimity
7. Direct pressure on dissenters
8. Self-appointed mind guards
CSSBB 2014

Risky - Shift
Many people think that the proposed solutions to team
projects would be fairly conservative. That is, any
proposed solution is an “averaged” remedy. However,
those experienced with team mechanics and dynamics
have found the opposite to be the case. Teams often get
swept up with expansive and expensive remedies.
There are ways to combat this tendency. One way is to

discuss risky-shift openly in the initial training. Another
approach is to ask the question, “If this were our
personal money would we still risk it on the proposed
solution.”
CSSBB 2014

Common Team Problem Areas

The following team problem areas are frequently
encountered and must be addressed by team leaders,
facilitators, sponsors, and upper management:
C Waning management support

C Inadequate meeting documentation
C Inadequate time or training
C Teams expose problems that are viewed as threats
C Facilitator and leader controversies can develop
C Good facilitation skills may be hard to find
C Conflicts may arise with suggestion programs
C Potential resistance may come from labor unions
C Teams may tackle problems outside their area
C Crisis management creates scheduling problems
C The reward system may be inconsistent
C Unproductive competition and conflict may occur
C Idea generation and evaluation are not separated
C Facts and opinions are not distinguished
C There is a failure to assign team responsibilities
CSSBB 2014

Common Team Problem Areas (Cont’d)

Problem Examples How to Fix
Floundering C Team direction is unclear C Leader must provide clarity
C Members seem overwhelmed C Review the team purpose
C Decisions are postponed C Ask “How can we proceed?”
Dominant C Members interrupt others C Promote equal participation

Participants C Members dominate the conversation C Structure the discussion
Overbearing C A member has excessive influence C Reinforce team concepts

Participants C A member has legitimate authority C Ask the expert to lead the group
C A member is an “expert” C Have a private discussion with “expert”
Negative C Members say “We tried that already” C Reinforce the positive
Nellies C Members defend their turf C Ask for other points of view
C Members are negative of suggestions C Separate idea generation from criticism
Opinions as C Members present opinions as facts C Ask for support data

Facts C Members make unfounded assumptions C Question opinions and assumptions
C Self assurance seen as unquestionable C See groupthink discussion
Shy Members C Members are reluctant to speak C Structure group participation

C Members afraid of making mistakes C Direct conversation their way
Jump to C Members rush to accomplish something C Reinforce the need for data analysis
Solutions C Members avoid data collection and C Ask for alternate solutions
analysis C Slow the process down
C Members want immediate decisions
Attributions C Members make casual inferences C Challenge assumptions

C Members don’t seek real explanations C Challenge judgments
C Members make psychological judgments C Ask for data to support conclusions
Put-downs C A member’s comments are ignored C Encourage active listening

(Discounts & C Members are not listening C Encourage equal participation
Plops) C The meaning of a suggestion is missed C Talk to parties privately
C Sarcasm is noted C Promote uniform idea consideration
Wanderlust C Conversations stray from the main topic C Follow a written agenda
(Tangents & C Sensitive issues are avoided C Reinforce team operating guidelines
Digressions) C Group pursues tangents C Redirect the discussion
Feuding C Win-lose hostilities emerge C Confront the adversaries alone

C The team takes entrenched sides C Reinforce team operating guidelines
C Some members become spectators C Replace the guilty parties if necessary
Risky-Shift C Expansive and expensive remedies are C Ask “If this were my personal money
suggested (using company money) would I still spend it?”
CSSBB 2014

Conflict Resolution
The following guidelines can be used by project leaders
to resolve conflict:
C Determine how important the issue is to all involved

C Determine if the issue can be discussed
C Select a private meeting place
C Parties must understand their responsibilities
C Let parties deal with the problem and solution
C Let all parties make opening comments
C Let parties express concerns, feelings, and ideas
C Guide all parties toward a clear problem definition
C Encourage participants to propose solutions
C Examine the problem from a variety of perspectives
C Discuss any and all proposed solutions
C Evaluate the costs versus the gains for proposals
C Choose the best solution
C Asking how the process might be improved
CSSBB 2014

Conflict Resolution (Continued)

Conflict is the result of mutually exclusive objectives or
views, manifested by emotional responses such as
anger, fear, frustration, and elation. Some conflicts are
inevitable in human relationships. Common causes of
conflict include:
• Organizational structure
• Value differences
• Role pressures
• Perceptual differences
• Divergent goals
• Status threats
• Personality clashes
• Differences in ideals
• Changes in procedures
• Discrepancies in priorities
CSSBB 2014


Negative conflicts result in:
• Hostility
• Win - lose situations
• Lose - lose situations
• Undesirable consequences
• Isolation
• Loss of productivity
CSSBB 2014


Positive conflicts result in:
• A combined desire to unite and improve

• Win - win situations
• Creative ideas brought forth
• Better understanding of tasks, problems
• Better understanding of other’s views
• Wider selection of alternatives
• Increased employee interest and participation
• Increased motivation and energy
CSSBB 2014


The ways of dealing with conflicts can be depicted in a
two dimensional model for conflict handling behavior,
adapted from the Thomas-Kilmann Conflict Mode
Instrument:
Unassertive º Assertive
Competing Collaborating
Assertiveness
Compromising
Avoiding Accommodating
Uncooperative º Cooperative
CSSBB 2014


• Avoiding is unassertive and uncooperative - the
individual withdraws from the situation. (You lose,
I lose).
• Accommodating is unassertive but cooperative - the

individual yields to the wishes of others. (You win,
I lose).
• Competing is assertive and uncooperative - the

individual tries to win, even at the expense of
others. (You lose, I win).
• Collaborating is assertive but cooperative
• Compromising is intermediate in both assertiveness

and cooperativeness
CSSBB 2014


There is no specific right or wrong method for handling
conflicts. The following are general applications for the
various conflict handling methods:
• Avoiding is appropriate for less important issues or

when the potential damage from conflict outweighs
the benefits of the goal.
• Accommodating is suitable when one party is

wrong or the issue is more important to the others
than it is to yourself.
• Competing is applicable when quick decisions are

needed and a stronger influence is held by one side.
• Collaborating is used when both views are

important and an integrated solution is desired.
• Compromising is used when two opponents have

equal power and the goals are not worth the effort
or disruption of mutually exclusive solutions.
CSSBB 2014

Negotiation Techniques
Nierenberg states that negotiating is the act of
exchanging ideas or changing relationships to meet a
need. As common and as important as negotiating is in
everyday life, most people learn to negotiate through
trial and error.
Negotiating should not be a process of using

overwhelming and irresistible force on the other party.
Some degree of cooperation must be employed in the
negotiating process. In dealing with people in a
business context, the best approach is to think win-win.
Both sides should emerge with a successful deal.
CSSBB 2014

TEAM DYNAMICS/MEETING MANAGEMENT
Team Time Management

Effective improvement teams manage their resources
well. One of the most valuable of these resources is
time. Many of the successful time management
elements are detailed or implied in the following
discussions of meeting structure, operating guidelines,
and sample meeting forms.
CSSBB 2014

Team Meeting Structure

Any effective team meeting needs logical structure for
many reasons including time management. Listed
below is an example format.
1. Develop an agenda
C Define goal(s)
C Identify discussion items
C Identify who should attend
C Allocate time for agenda items
C Set time and place
2. Distribute the agenda in advance
3. Start on time
4. Appoint a recorder to record minutes
5. Use visual aids liberally (flip chart, chalkboard)
6. Reinforce:
C Participation
C Consensus building
C Conflict resolution
C Problem solving processes
7. Summarize and repeat key points throughout
8. Put unfinished items on the next agenda
9. Review assignments and completion dates
10. Finish on time
11. Distribute minutes promptly
12. Critique meeting effectiveness periodically
CSSBB 2014

Typical Team Operating Guidelines
Topic Ideas
Team Who sets? When published? Input
agenda invited? A rolling agenda with
priorities? Recorder to publish.
Attendance Excused absences only? How are

latecomers handled? Minimum
members to conduct business?
Excused attendance permitted?
Latecomers to be updated?
Meetings Time, frequency, place? Which meeting

room? The time and frequency must be
determined.
Decision Consensus, collaborative, majority?

process Can one person remove an item from
the agenda?
Minutes/ Select a recorder. How are minutes

reports approved? Where posted? Who types?
How distributed? Is the recorder a
volunteer or appointed? Timekeeper to
maintain agenda timing?
CSSBB 2014

Typical Guidelines (Continued)

Leader role Expectations? How selected?
Behavioral Listening; interruptions; cell phones,
norms pagers, and radios off; no smoking;
breaks called at members discretion;
empathetic listening; feedback should
be constructive, specific, and timely.
Confidential What goes outside the group?
Guests How invited? How excused?
Audits How frequent? Who is responsible?
Facilitator How selected? Expectations? How will
this role differ from the leader?
Conflict Expected? How managed?
Results How initiated? How routed? Who is
informed?
Follow Follow through on commitments,
through analysis, word processing, etc.
CSSBB 2014

Sample Meeting Forms

Some simplified team-based forms are shown in the
Primer. A group accomplishment record is very
beneficial, since team progress in a single session may
be slow.
CSSBB 2014

Team Performance Evaluation

Teams are established to accomplish something within
a time frame. A clear understanding of the team’s
objectives is a very important element of creating a
successful team. A team can and should be expected to
develop and refine its objectives and measures of
performance.
Teams are often chartered to improve performance in

some way. Performance is associated with speed,
quality, cost and effectiveness. Finding good measures
on these variables is not always easy. Most
organizations count what can be easily counted, without
regard to the organization’s performance. Getting a
better handle on performance usually means starting
with the customer’s point of view.
CSSBB 2014

Team Performance Checklist

An example team performance checklist is shown
below.
Element Positive Negative

Establish Clarity about goals Uncertainty about
agenda and content goals and content
Stick to Always to the point Frequently gets
subject off track
Only one No interruptions Multiple
person talks conversations
Build on Comments are Focuses on what’s
positive additive wrong
Active No Comments that
listening misunderstandings show lack of
clarity
Participation Active participation Opinions only, no
data
Consensus Buy-in on all issues People stampeded
CSSBB 2014

Team Performance Factors

Listed below are some illustrative factors that can be
used to evaluate team performance.
Well Functioning Teams Poorly Functioning Teams

Relationship Factors
There is a firm team identity Members do not identify with the team
Conflict is openly discussed There are open or covert personality conflicts
Team members support each other Relationships are competitive
Members enjoy each other Members are defensive
Process Factors
Decisions are made by consensus Decisions are made by a few members
Meetings are efficient and task oriented Meetings are unproductive
There is growth and learning Minor points are debated
All members participate in discussions Members are late, passive, or do not attend
Members are informed and minutes are kept No record of progress is kept
There is ongoing performance feedback Feedback awaits the end of the project
Members listen well There are frequent interruptions
Goal Factors
Team members help set objectives Most goals are predetermined for teams
Objectives are understood by all members Goals are unclear or poorly communicated
Objectives are realistically set and met Members are oblivious to team goals
Environmental Factors
Team members are in close physical proximity Physical separation prevents attendance
There are adequate skills and resources Resources are inadequate
There is management and member support There is a lack of organizational recognition
Role Factors
There is strong effective leadership No clear leader is identified
Clear responsibilities are defined Members engage in power plays
Roles are understood and supported by all There is buck-passing of responsibility
Members work as a team Members act independently
CSSBB 2014

Team Reward and Recognition

Recognition and rewards for teams and team members
can be grouped into the following types:
C Material items of significant value or equivalent:
C Material items of incidental value or equivalent:
C Intangible items such as satisfaction and thanks

CSSBB 2014

Team Reward and Recognition (Cont’d)

Team rewards, which are material items and are given to
the team, should be the same for all members of the
team. Many intangible rewards are not formally given by
someone, but people can still receive and appreciate
them.
The ultimate reason that rewards and recognition are

given is to provide positive reinforcement for good
performance, with the expectation that this performance
will be repeated in the future.
Rewards and recognition are best received when they

are personal to the individual receiving them. This is
more difficult to achieve as the team size increases.
Probably one of the best rewards is “thank you” when it

is sincerely meant.
CSSBB 2014

Management Presentations
One way to offer recognition and support for teams is to
encourage them to present their accomplishments to
upper management. Presentation objectives include:
C Displaying skills
C Showing accomplishments
C Summarizing project results
C Gaining necessary approvals
C Opening lines of communication
C Knowing the customer’s true needs
Presentation guidelines include many of the following:
C Have a handout report and a leader

C Introduce and give everyone a chance to speak
C Capture key problems and action steps
C Indicate costs and benefits of recommendations
C Present an implementation plan
C Use visual aids - be professional
C Emphasize achievements and accomplishments
C Start and end on time (about one-half hour duration)
C Present no sudden surprises
CSSBB 2014

TEAM DYNAMICS/DECISION - MAKING METHODS
Team Decision-Making Methods

Teams must reach a decision or resolve a problem using
a variety of helpful decision-making techniques.
Brainstorming
Brainstorming is an intentionally uninhibited technique
for generating creative ideas when the best solution is
not obvious. The brainstorming technique is widely
used with the cause-and-effect diagram.
Generate a large number of ideas: Don’t inhibit anyone.

Just let the ideas out. The important thing is quantity,
but record the ideas one at a time.
Free-wheeling is encouraged: Even though an idea may

be half-baked or silly, it has value. It may provoke
thoughts from others.
Don’t criticize: There will be ample time after the

session to sift through the ideas for the good ones.
During the session, do not criticize ideas because that
might inhibit others.
CSSBB 2014

Brainstorming (Continued)
Encourage everyone to participate: Everyone thinks
and has ideas. So allow everyone an opportunity to
speak. Speaking in turn helps.
Record all the ideas: Appoint a recorder to write down

everything suggested. Don’t edit the ideas. Keep a
permanent record that can be read later.
Let ideas incubate: One must free the subconscious

mind to be creative. Let it do its work by giving it time.
Don’t discontinue brainstorming sessions too soon.
Select an appropriate meeting place: A place that is

comfortable, casual and the right size will greatly
enhance a brainstorming session.
Group size: The ideal group size is 4-10 people.
Brainstorming does not necessarily solve problems or

create a corrective action plan. It can be effectively
used with other techniques such as multivoting to arrive
at a consensus.
CSSBB 2014

Team Consensus
Unlike majority rule, there is no team vote with
consensus. Consensus implies that the proposed
action has general team support. The decision may not
be every team member's first choice. It is a course of
action that all can live with and not die over. Note that
multivoting and nominal group techniques have
elements of consensus built into them.
CSSBB 2014

Nominal Group Technique

The nominal group technique (NGT) brings people
together to solve problems but limits initial interaction
among them. The concept is to prevent social pressures
from influencing the generation of ideas. To conduct a
NGT problem solving meeting:
C A facilitator or moderator leads the discussion

C A group of five to nine individuals are assembled
C A problem is presented
C Members create ideas silently and individually
C The facilitator then requests ideas in sequence
C Each idea is recorded until ideas are exhausted
C No discussion is allowed at this point
C Ideas are then clarified and evaluated
C Expanding on the ideas of others is encouraged
C Voting for the best solution idea is then conducted
C Several rounds of voting may be needed
The facilitator should allow about 60 to 90 minutes for a

problem solving session.
CSSBB 2014

Voting
Voting is similar to the multivoting approach except that
only one vote is permitted per team member. Voting can
result in majority or unanimous decisions. In some
immature team environments, voting can lead to
conflict. This is why consensus decisions are usually
preferred.
CSSBB 2014

Multivoting
Multivoting is a popular way to select the most
potentially most important items from a previously
generated list. Often, there are too many items for a
team to work on at a single time. It may be worthwhile
to narrow the field to a few items worthy of immediate
attention. Multivoting consists of the following steps:
C Generate and number a list of items

C Combine similar items, if the group agrees
C If necessary, renumber the list
C Allow members to choose several items
C Members may make their initial choices silently
C Then the votes are then tallied
C Eliminate those items with the fewest votes
Members normally have a number of choices equal to

one-third of the listed items. Voting can be conducted by
a show of hands. The items receiving the largest
number of votes are usually worked on first.
CSSBB 2014

Effort/Impact
One of the most viable methods of deciding on an
acceptable course of action is by determining and
comparing the impact of that action with the effort (or
expense) to accomplish it. Usually some form of a
matrix or modified Johari window is used. Refer to
Figure below.
Effort (Expense)
Impact Little Great
Little This is low hanging A low priority. Shelve it
fruit. It should be until all other options are
done when completed.
convenient.
Great A top priority. It An important long-term
should be given objective. Needs a
immediate attention. champion to keep it active.
CSSBB 2014

Force Field Analysis

Another tool often used for problem identification and
resolution is force field analysis. A description of the
process used to perform a force field analysis follows:
1. A desire to understand the forces acting on a

problem to be resolved
2. Determine the forces favoring the desired goal

(driving forces)
3. Determine the opposing forces to the desired goal

(restraining forces)
4. Add to the driving forces to overwhelm the

restraining forces, or
5. Remove or weaken the restraining forces, or
6. Do both (strengthen driving forces and weaken

restraining forces)
The Primer contains an example of a force field analysis

on reducing teenage smoking:
CSSBB 2014
IV. TEAM MANAGEMENT III.EXTRA

Team Problem Solving Methodologies

A green or black belt directed project team should select
the approach and tools appropriate to the task. Five
methodologies are reviewed in this Primer Section:
PDCA, PDSA, a classical approach, DMAIC, IDEA. These
methodologies generally use the seven original quality
tools: checksheets, Pareto diagrams, process flow
diagrams, scatter diagrams, run charts, histograms, and
fishbone diagrams.
CSSBB 2014

PDCA
The PDCA cycle is very popular in many problem
solving situations because it is a graphical and logical
representation of how most individuals already solve
problems.
4. ACT 1. PLAN
3. CHECK 2. DO
The PDCA Cycle

CSSBB 2014

PDCA (Continued)
Plan (P): Establish a plan for achieving a goal.
Do (D): Enact the plan.
Act (A): Implement necessary reforms when the

results are not as expected.
Check(C): Measure and analyze the results.

CSSBB 2014

PDSA
Deming was somewhat disappointed with the Japanese
PDCA adaption. Deming proposed a Plan-Do-Study-Act
continuous improvement loop (actually a spiral), which
he considered principally a team oriented, problem
solving technique.
1. Plan - What could be the most important

accomplishment of this team? What changes
might be desirable? What data is needed?
2. Do - Carry out the change or test decided upon,

preferably on a small scale.
3. Study - Observe the effects of the change of the

test.
4. Act - Study the results. What was learned? What

can one predict from what was learned?
5. Repeat step 1 with new knowledge accumulated.
6. Repeat step 2 and onward.

CSSBB 2014

Classic Team Problem Solving Steps

1. Identify business or customer problems; select one
to work on.
2. Define the problem; if it is large, break it down to

smaller ones and solve these one at a time.
3. Investigate the problem. Collect data and facts.
4. Analyze the problem. Find all the possible causes;

decide which are major ones.
5. Solve the problem. Select the solution that has the

greatest organizational benefit.
6. Confirm the results.

CSSBB 2014

Team Problem Solving Steps (Cont’d)
ACT PLAN
CHECK DO
SELECT CONDITION(S)
NEEDING IMPROVEMENT
STUDY CURRENT PROCEDURES
NO IS THE PROJECT
MISSION & SCOPE
APPROPRIATE?
YES
DETERMINE THE POTENTIAL CAUSES
(CANDIDATE FOR DATA COLLECTION)
CATEGORY FREQUENCY
COLLECT DATA A
B
C
ANALYZE DATA
YES
NO MORE NO DOES
DETAIL THE ANALYSIS
NECESSARY? POINT TO
SOLUTION?
YES
IMPLEMENT SOLUTION
ACT PLAN
CHECK DO
CSSBB 2014

DMAIC Process
Each step in the cyclical DMAIC process is required to
ensure the best possible results from lean six sigma
team projects. The process steps are detailed below:
Define the customer, their critical to quality (CTQ)

issues, and the core business process involved.
C Define who the customers are

C Define customer requirements and expectations
C Define project boundaries
C Define the process to be improved by mapping
Measure the performance of the core business process

involved.
C Develop a data collection plan

C Collect data from many sources
C Collect customer survey results
CSSBB 2014

DMAIC Process (Continued)

Analyze the data collected and process map to
determine root causes of defects and opportunities for
improvement.
C Identify performance gaps

C Prioritize opportunities to improve
C Identify excessive sources of variation
C Identify objective statistical procedures
Improve the target process by designing creative

solutions to fix and prevent problems.
C Create innovative solutions

C Develop and deploy improvement plans
Control the improvements to keep the process on the

new course.
C Prevent reverting back to the “old way”

C Develop an ongoing monitoring plan
C Institutionalize the improvements
CSSBB 2014

IDEA Process
The IDEA problem solving loop is similar in nature to the
PDCA and DMAIC process cycles. IDEA stands for
Investigate, Design, Execute, and Adjust. The detailed
IDEA process steps are:
C Investigate: Provide a definition of the problem,

provide some facts about the problem, and provide
a root cause.
C Design: Envision the idealized future state and

create a list of options to achieve the idealized state.
C Execute: Establish the specific metrics for success,

test the best solution, and determine a measurable
project impact.
C Adjust: Reflect on the outcome of the project. This

is a post-action review and is also conducted for
successful projects.
CSSBB 2014
IV. TEAM MANAGEMENT III.D.1

TEAM DYNAMICS/NEEDS ASSESSMENT
Training Needs Assessment

Training needs assessment is a critical step in the team
training process. It is usually used to uncover the gaps
between adequate and inadequate job performance. The
results of a training needs assessment are twofold:
1. What is being done now

2. What should be done
Conducting a training needs analysis requires diligent

work and effort on behalf of the analyst.
CSSBB 2014

Training Needs Assessment (Continued)

Surveillance
An organizational trainer should gather information in

regards to many key company needs. These facts,
figures, documents, data, etc., should include: policy
statements, policies, procedures, attitudes, feelings, and
contacts. These and other factors can provide sensory
data for training efforts.
CSSBB 2014


Investigation
In this stage, a possible performance gap may be

suspected. The next step would be to gather more
details in that specific area using techniques like:
1. Observation by individuals
2. Interviews with individuals
3. Questionnaires with ratings
4. Records of activities
5. Work sampling
6. Performance appraisals
7. Work studies
8. Personnel testing
CSSBB 2014


Analysis
At the analysis stage, the gathered information is sorted

and examined for validity. It is then summarized with
conclusions and recommendations. Strategic planning
considerations include the following:
C Organizational level: Determine the critical business

issues or opportunities
C Process level: Identify the process areas with
greatest business impact
C Job/performer level: Identify the areas that the
performer needs to address
A training analysis seeks to place various root causes

for inadequate performance into appropriate categories
and floor level observations.
CSSBB 2014

Resources for Training

Environmental and location factors to consider include:
C Good lighting
C Room temperature and ventilation
C Acoustics
C Good line of sight
C Comfortable seating
C Convenient accommodations
C Appropriate timing
C Central physical location
CSSBB 2014

Resources for Training (Continued)

The quality of training, to a high degree, is determined
by the training department personnel.
The training department staffing requires three types of

people:
C Experts in the subject matter and content resource

materials
C Technical writers, editors, training professionals,

etc., who design and produce the materials
C Teachers, instructors, and presenters who deliver

the training.
CSSBB 2014

Curriculum Development
Consideration should be given to the four parts of a
learning experience: the process, the content, the
trainer, and the student.
The Process
The training method can be termed trainee-centered or

trainer-centered models. In the trainee-centered
learning model, the trainee has most of the
responsibility for learning.
In trainer-centered learning, the trainer provides the

direction and pace of the learning for the student.
CSSBB 2014

Curriculum Development
The Content
The skill or knowledge that the learner has to acquire is
the content phase of training. The content may be
simple or complex.
The Trainer
As the complexity of the training content rises, the
trainer’s skill level or ability to provide guidance to the
learner must rise.
The Student
In an industrial setting most of the trainees will be
adults. The adult learner has a different learning
maturity that should be considered.
CSSBB 2014

Classical Training Tools

Types of classical training tools Includes:
C Lectures, courses, classes

C Tutoring
C Mentoring
C Books
C Case studies
C Homework
C Student presentations
C Demonstrations
C Hands on learning
C Quizzes, exams, tests
C Overhead projection of information
C Slides and movies
C Audio and cassette tapes
C Work shops
C Trade schools
C Seminars
C Speeches
C Home study and correspondence courses
C On the job training
CSSBB 2014

Classical Training Tools (Continued)

Applications of Classical Training Tools Include:
C Elementary and supplemental education

C Preparation of basic skills
C Continuing training and education
C Expansion of skills and knowledge
C Retraining and cross training
C Hobby or pleasure interests
CSSBB 2014


Advantages of Classical Training Tools Include:
C Preparation and training materials have low cost

C The instructor can answer specific or unique
questions
C Adjustments can be made in course materials
C Portions of the material can be repeated
C The instructor has immediate or short term
feedback
CSSBB 2014


Disadvantages of Classical Training Tools:
C Coordination of schedules to accommodate a fixed

time
C Cost per individual is relatively high
C Desired training may not be available
C The number of qualified instructors may be limited
C Personal attention decreases with class size
CSSBB 2014

Technological Training Tools

The use of technological tools, such as videos,
computers, etc., in the training process is increasing.
Types of Technological Training Tools:
C Training at home or office via internet

C Teleconferences via remote communication
C Video tapes, CD ROMs, video disks
C Interactive computer training and simulators
C Reduced scale models
C Instrumented training aids
C Device sensors as performance feedback
CSSBB 2014

Technological Training Tools (Continued)
Applications of Technological Training Tools:
C Large audiences - Connected through satellite

broadcasts.
C Software which can be customized.
C Simulators - Can simulate real problems
C Computer analysis of physical performance
C Artificial Intelligence to assist training
CSSBB 2014

Advantages of Technological Training Tools:
C The individual can control learning rates

C No instructor is required
C Material can be repeated
C The material can be used many people
C Low per student costs for most applications
C Use is usually not limited by location
C Can be used to examine “what-if” situations
CSSBB 2014

Disadvantages of Technological Training Tools:
C Higher costs to develop

C Entry barriers for required equipment
C Longer development time required
C Cannot accommodate unique questions
C The user learning curve is longer
C Special simulator equipment may be costly
C May be more difficult to modify training content
CSSBB 2014

Training Delivery
The tools of training and education programs are the
methods and medium used to deliver the information.
These methods can be categorized into two broad types:
C Classical Tools - The more classical methods use

lectures, workbooks, case studies, and on the job
training.
C Technological Tools - The newer, technology driven

methods include videos and computer delivered
instruction.
Neither method is necessarily superior to the other.

CSSBB 2014

Adult Learning Principles

Adult learners are the typical students in the industrial
sector. The key characteristics of adult education
include:
1. Adults have more life experiences

2. Adults are motivated to learn
3. Adults have more responsibilities in their lives:
4. Adults may lack self-confidence for learning:
5. Adults vary more than young people
The best learning styles for adults:
1. Help set goals for learning

2. Have an appropriate setting
3. Be problem oriented
4. The pace is set by the adult
5. Provide feedback
CSSBB 2014

General Learning Principles

Goetsch summarizes the following principles of
learning:
C People learn best when they are ready to learn

C People learn more easily when a new topic relates
to a known topic
C People learn best in a step-by-step manner
C People learn by doing (very important)
C Retention is best when people use the new
knowledge frequently
C Success in learning tends to stimulate additional
learning
C People need both immediate and continual progress
feedback
CSSBB 2014

General Learning Principles (Continued)

Smith provides the following learning principles:
1. Whole or part learning

2. Spaced learning
3. Active learning
4. Feedback
5. Overlearning
6. Reinforcement
7. Primacy and Recency
8. Meaningful material
9. Multiple-sense learning
C 80% if information is by sight
C 11% if information is by hearing
C 9% if information is by other senses
10. Transfer of learning
CSSBB 2014

General Learning Principles (Continued)

Another source provides the following percentages
regarding what learners retain from instruction they
receive:
C 10% of what is read

C 20% of what is heard
C 30% of what is seen
C 50% of what is seen and heard
C 70% of what is seen and spoken
C 90% of what is said while doing
Transferring learning to the workplace will depend on

the closeness of training to the workplace and to the
workplace design.
CSSBB 2014

Training Program Evaluation

Pure training results are often clouded by complications
like:
C Unknown equipment conditions

C Changing customer preferences
C Emerging competition
C Step-changes in technology
Potential internal or external interest areas is provided

below:
C Has the percentage of on-time deliveries increased?

C Has downtime decreased?
C Have the number complaints decreased?
C Has the number of subscribers increased?
C Has the death rate, during surgery, decreased?
C Do employees feel more empowered?
C Has employee turn-over decreased?
CSSBB 2014

Trainee Feedback and Training Effect

The students or trainees should be given an opportunity
for evaluation feedback regarding the training program.
This evaluation includes:
C Instructor knowledge, style, thoroughness, and

presentation
C Content and applicability of training materials
C Perceived value of the training
C Rating of environmental conditions
C What was least liked about the class
C What was most liked about the class
C A request for suggested training changes or
improvements
The instructor and other interested parties should also

be queried for their inputs on training program
improvement.
CSSBB 2014

Training Program Evaluation (Continued)

Evaluation of Training Effect on the Company
The benefits of training include hard factors (easy to

measure) such as reduced scrap, increased
productivity, and soft factors (subjective in
measurement) such as improved morale, reduced
employee turnover, and investment in human resource
development.
Post Training Reinforcement
It is best to apply the concepts learned during training

to simple or small scale problems so that fundamentals
can be understood. The longer the new techniques lie
dormant, the higher the probability that they will be
forgotten. Some certifications and job skills require re-
training at fixed intervals to retain proficiency.
CSSBB 2014

Training Costs and Benefits

The trainer may need to gather data in the following
areas:
C Costs of training
C Organizational benefits
C Cost-benefit comparisons
Costs of Training
The costs involved in training include:
C Trainer costs
C Administrative costs
C Materials
C Travel expenses
C Overhead
C Costs of trainee time
Cost - Benefit Comparison
This is generally an after-the-fact type of comparison.

Various costs will be gathered on the training effort,
either before or after the event. The benefits of the
training may be known quite soon after the training, or
months or years later.
CSSBB 2014
IV. TEAM MANAGEMENT

QUESTIONS
4.2. Many training instructors have developed approaches to

emphasize multiple sense learning. Which of the following options
would be generally recognized to best foster student retention?
a. Reading and hearing

b. Hearing and seeing
c. Seeing and speaking
d. Reading and seeing
4.6. In most cases, an improvement team receives the least control and
direction during which of the following stages:
a. Building
b. Storming
c. Performing
d. Alarming
4.9. The team leadership model characterized by “we’ll talk but I’ll
decide” is called:
a. Directing
b. Coaching
c. Supporting
d. Delegating
Answers: 4.2. c, 4.6. c, 4.9. b

CSSBB 2014
IV. TEAM MANAGEMENT

QUESTIONS
4.16. Project improvement team members normally have:
a. Narrow skills and experiences

b. Diverse skills and narrow experiences
c. Diverse experience and narrow skills
d. Diverse skills and experiences
4.19. Divided member loyalties, having personal confrontations, thinking

individually, and learning their role on the team are indicators that
a team is in which stage of its evolution?
a. Forming
b. Storming
c. Performing
d. Norming
4.21. In order for a team to understand and anticipate the resistance that
other members of the organization may have for a proposed
recommendation, what team tool would be most useful?
a. Multivoting
b. Groupthink
c. Force field analysis
d. Nominal group think
Answers: 4.16. d, 4.19. b, 4.21. c

CSSBB 2014
IV. TEAM MANAGEMENT

QUESTIONS
4.26. Which of the following is the most beneficial reason for training
program evaluation?
a. To reduce the training content

b. To compensate the instructors
c. To measure the perceived training value
d. To improve the training conditions
4.29. It is noted that the involvement and participation of which of the

following team roles expands and increases over time?
a. Team members
b. Facilitator
c. Sponsor
d. Leader
4.32. The situational leadership model which exhibits both low

supportive and low directive behavior is called?
a. Directing
b. Coaching
c. Supporting
d. Delegating
Answers: 4.26. c, 4.29. a, 4.32. d

CSSBB 2014
IV. TEAM MANAGEMENT

QUESTIONS
4.38. When the team is creating their guidelines, the issue of what
information will be shared outside the team is called:
a. Brainstorming
b. Conflict resolution
c. Confidentiality
d. Empathic listening
4.43. A skilled facilitator will:
a. Intervene when progress ceases

b. Correct the group when their ideas are wrong
c. Take sides when one side is correct
d. Make sure the opinions of influential members are heard
4.47. Identify the facilitation technique that would most likely result in
team conflict:
a. Multivoting
b. Voting
c. Consensus
d. Nominal group technique
Answers: 4.38. c, 4.43. a, 4.47. b

© QUALITY COUNCIL OF INDIANA V-1 (307)
CSSBB 2014
V. DEFINE
OUR PLANS MISCARRY BECAUSE

THEY HAVE NO AIM. WHEN A
MAN DOES NOT KNOW WHAT
HARBOR HE IS MAKING FOR, NO
WIND IS THE RIGHT WIND.
SENECA (4 B.C. - 65 A.D.)
CSSBB 2014
V. DEFINE IV.A.1
VOICE OF THE CUSTOMER/IDENTIFICATION
Six Sigma Methodology - Define

The Define element is presented in the following topic
areas:
C Voice of the customer

C Business case and project charter
C Project management tools
C Analytical tools
Voice of the Customer

Voice of the customer is reviewed in the following topic
areas:
C Customer identification
C Customer feedback
C Customer requirements
CSSBB 2014
V. DEFINE IV.A.1
Customer Identification
Six sigma quality is built around the customer. They
define quality and set expectations. They rightfully
expect performance, reliability, competitive prices, on-
time delivery, service, and clear and accurate
transactions.
At times, the customer of the project may not be as

evident as initially thought. The receiver of the next
operation, an internal department, could be thought of
as a customer.
The primary customer is of utmost importance to the

process.
CSSBB 2014
V. DEFINE IV.A.1
Customer Identification (Continued)

External customers are the most important part of any
business. If one can identify them and understand their
requirements, goods and services can be designed that
they will want to buy. Customers evaluate goods and
services by considering cost, quality, features, and
availability factors (CQFA).
Pande suggests that a flaw with customer data would be

“getting information from the wrong customers.”
Customer satisfaction surveys may result in information
that reflects the thoughts of a market segment for which
the product was not intended.
The largest portion of unhappy customers may come

from the unintended end user. The company will then
react to the bad data. The organization may fix the
wrong problem.
CSSBB 2014
V. DEFINE IV.A.1

Customers can constitute:
C Current, happy customers

C Current, unhappy customers
C Lost customers
C Competitor’s customers
C Prospective customers
The following methods can collect information and data

from customers or would-be customers:
C Surveys
C Focus groups
C Interviews
C Complaint systems
C Market research
C Shopper programs
CSSBB 2014
V. DEFINE IV.A.1

The traditional methods of obtaining customer
information could also include:
C Targeted and multi-level surveys

C Targeted and multi-level interviews
C Customer scorecards
C Data mining and warehousing
C Customer audits
C Supplier audits
C Quality function deployment
Gathered information should allow the organization to

identify customer requirements and to spot trends.
Consider how both internal and external customers can

be identified, and what their requirements might be in
order to understand and improve the business process.
CSSBB 2014
V. DEFINE IV.A.1
Project Stakeholders
A six sigma project with high impact will bring about
major changes to a system or to the entire company.
The change can affect various people inside and outside
of the system. Major resistance to the change can
develop. As part of the define process, attempts to
remove or reduce the resistance must be made.
Stakeholders can be identified as:
C Managers of the process

C People in the process
C Finance
C Suppliers
C Customers
A stakeholder communication plan should be developed

to reduce or remove the resistance to change.
CSSBB 2014
V. DEFINE IV.A.1
Internal Customers
An internal customer can be defined as anyone in the
company who is affected by the product or service as it
is being generated. The internal customer is sometimes
forgotten in the effort to produce an item or service for
the external customer.
Internal customers are often employees of the company.

Kaoru Ishikawa coined “The next operation as
customer” in order to remove the sectionalism of
departments toward each other. The idea is to enable all
employees to come together to solve problems.
Within a company, the staff should consider what kind

of work they can perform for the line departments.
Research has shown that management practices relate

to employee satisfaction, which also impacts customer
satisfaction.
CSSBB 2014
V. DEFINE IV.A.1
Internal Customers (Continued)

Internal employee communications for customer
satisfaction can be improved through the following
options:
C Company newsletters
C Story boards
C Team meetings
C Posting customer letters
C Staff meetings
C Display of goals
C Quality awards
To stay competitive, a constant schedule of training for

the entire workforce is required. Employee training
must focus on helping them do their job better.
CSSBB 2014
V. DEFINE IV.A.1
External Customers
External customers are not part of the organization but
are impacted by it. End users, Intermediate customers,
and impacted parties are important. Generally, external
customers play a critical role by providing the major
portion of company revenues.
End Users
The category of external customers includes those that
purchase a product or service for their own use. In the
case of a CSSBB Primer, the end users would be:
C Potential test takers

C People interested in self development
C People who need to know about quality
C Course or university instructors
C People who buy from the bookstores
CSSBB 2014
V. DEFINE IV.A.1
Intermediate Customers
Intermediate customers purchase the product or service
and then resell, repackage, modify, or assemble the
product for sale to an end user. The intermediate
customers for CSSBB Primers could include:
C Training and consulting firms

C University and commercial bookstores
C Foreign distributors
Examples of intermediate customers in other areas

include:
C Retailers
C Distributors
C Wholesalers
C Transport companies
CSSBB 2014
V. DEFINE IV.A.1
Impacted Parties
The third external customer category are those who do
not purchase or use the product, but are impacted by it.
Certainly the families of the CSSBB Primer authors
would be in this category. The impacted parties for an
educational system might include:
C Parents
C Communities and civic groups
C City governments
C Service companies who hire students
C Vocational schools
C Colleges and universities
C Service companies who are suppliers
Meeting external customers’ needs can become a

complex process.
CSSBB 2014
V. DEFINE IV.A.1
External Customer Segmentation

External customers may be sorted in many ways in an
attempt to better understand their requirements and
identify possible market niches. Business customers
can include for-profit and not-for-profit enterprises.
Examples of not-for-profits might include schools,
hospitals, public agencies, etc. The various customer
groups could also be reviewed for:
C High profit margin

C Risk of market
C Competition in market
C Growth in market
CSSBB 2014
V. DEFINE IV.A.1
External Customer Segmentation (Cont’d)

The consumer customer market differs from the
business market as follows:
C The consumer market has a large number of

customers
C The majority of consumer purchases are small in

actual dollar amounts
C The transaction is usually a simple purchase
C Most consumers are not very knowledgeable about

the product
C The supplier does not share proprietary information

with the consumer
CSSBB 2014
V. DEFINE IV.A.1
External Customer Segmentation (Cont’d)

In contrast, the business customer acts in the following
manner:
C There are a very small number of business

customers
C The amount purchased per transaction is quite large
C The purchase is handled through specialized

personnel
C The customer may know more about the

requirements than the producer
C The supplier may allow the customer access to all

sorts of information
It is also important to look at the market for the next two

to five years and estimate how it will change and grow.
CSSBB 2014
V. DEFINE IV.A.1
Customer Service
The customer driven company is beginning to emerge in
America. The public demands and expects better
quality products and service. One sample program
follows:
C Listen to the customer and determine needs

C Define a service strategy
C Set standards of performance
C Select and train the right employees
C Recognize and reward accomplishment
There is the need to listen to the customer, provide a

vision, provide training, improve the process, find or
develop response metrics, and measure the results.
About 70% of customers who leave a company do so
because of service quality.
CSSBB 2014
V. DEFINE IV.A.1
Customer Retention
Most organizations spend the bulk of their resources on
attaining new customers and smaller amounts on
retaining customers. High customer satisfaction
numbers do not necessarily mean the company has
good customer retention and good customer loyalty. It
has been found that current customers are worth as
much as five times more than new customers. The cost
of retaining a current customer is only one-fourth the
cost of acquiring a new customer.
Another study showed that companies will boost profits

by about 100% by just retaining 5% more of their
customers.
CSSBB 2014
V. DEFINE IV.A.1
Customer Retention (Continued)

Furlong lists some techniques for getting to know
customers better:
C Don’t use your own instincts as data

C See the world from the customer’s side
C People high in the organization are out of touch
C Get customers to talk
C 90% of unhappy customers won’t complain
C Do research to retain customers
C Determine how satisfied customers are
C Conduct research on customer expectations
C Develop a customer profile
C Share the results of customer research studies
C Don’t go overboard on the details and measurement
C Coordinate and use research efforts
C Understand that sometimes research does not help
CSSBB 2014
V. DEFINE IV.A.1
Customer Loyalty
The value of a loyal customer is not measured on the
basis of one gigantic purchase, but rather on his/her
lifetime worth. Loyal customers account for a high
proportion of sales and profit growth. Customer
retention generates repeat sales, and it is cheaper to
retain customers. Customer loyalty is something that
must be demonstrated through an act of execution,
trust, or delightful service. Customers become partners.
CSSBB 2014
V. DEFINE IV.A.2
VOICE OF THE CUSTOMER/DATA COLLECTION
Customer Metrics Selection

The primary and more detailed metrics are developed,
but are not finalized in the definition step. That is left to
the measure step. The primary metrics for a project
could come from several sources:
C Suppliers
C Internal processes
C Customers
The metrics that will affect projects involving suppliers,

internal processes, and customers would be: quality,
cycle time, cost, value, and labor.
CSSBB 2014
V. DEFINE IV.A.2
Customer Metrics Selection (Continued)

Garvin and Besterfield suggest nine dimensions of
quality measurement:
C Performance: Primary features of the product

C Features: Secondary added features added
C Conformance: Meeting fit, form, and function
C Reliability: The quality of a product over time
C Durability: Useful life
C Service: Ease of repair
C Response: Human interface
C Aesthetics: Product appearance
C Reputation: Based on past performance
CSSBB 2014
V. DEFINE IV.A.2
Customer Metrics Selection (Continued)

Secondary or consequential metrics could be derived
from the primary metrics. For example, if cycle time was
determined to be a key metric, the next step would be to
establish the numerical measurement. Examples of
measurements include:
C Defects per unit (DPU)

C Defects per million opportunities (DPMO)
C Average age of receivables
C Lines of error free software code
C Reduction in scrap
CSSBB 2014
V. DEFINE IV.A.2
Collecting Customer Data

When collecting data from customers, it helps to
consider the levels where customers impact the
business.
Business Level
Customers at this level are primarily shareholders and

top management employees. The data of interest is
primarily financial data such as stock price, market
share, revenues, earnings, return-on-investment (ROI),
return-on-net assets (RONA), etc.
Operations Level
Customers at this level are primarily those who

purchase the product (external) and those who manage
production operations (internal). Data of interest
measures overall process performance with the focus
on customer satisfaction (external measures of
operational effectiveness), and internal operations
efficiency (internal measures such as rolled throughput
yield, sigma levels, WIP inventory, etc.).
CSSBB 2014
V. DEFINE IV.A.2
Collecting Customer Data (Continued)

Process Level
Customers at this level are primarily internal, including

employees and the “next process” in the operation.
Data of interest primarily involves key process variables.
Typical analysis tools are statistical methods for
process control, capability, and improvement.
Since employees are customers, they should be

surveyed on a regular basis. Jack Welch feels the
survey factors should include:
C Job satisfaction
C Training
C Pay
C Advancement fairness
C Treatment: respect & dignity
C Company’s interest in well-being
CSSBB 2014
V. DEFINE IV.A.2
Voice of the Internal Customer

Surveys can serve as a tool for overall improvement.
Information should be gathered on improvement efforts
and some of the following factors:
C State of the company: What is the employee’s

perception of the company?
C State of quality efforts: Are the quality efforts

worthwhile?
C State of the processes: Are there improvements?
C Reaction to policies: What dumb things have been

implemented?
C Rating of job satisfaction: Do I like my job, my

boss, etc.?
C Rating of company satisfaction: Is the company a

good place to work?
CSSBB 2014
V. DEFINE IV.A.2
Voice of the External Customer

The voice of the customer is an expression for listening
to the external customer. It is necessary to have
constant contact with the customer. Among the ways
that a company can listen to the external customer are:
C Immediate customer surveys

C Customer follow-up surveys
C Community surveys
C Personal customer contact
C Customer contact reports
C Focus groups
C Customer interviews
C Electronic mail
C Test marketing
C Quality guarantees
C Inspectors
C Ombudsmen
C Use of toll-free phone numbers
C Use of suggestion boxes
CSSBB 2014
V. DEFINE IV.A.2
Customer Surveys
Research on customer satisfaction can be worthwhile in
helping the company efforts. The objectives of
customer research vary, but a few major themes are
noted below:
C To determine what quality is

C Find out what competitors are doing
C Define quality performance measures for use
C Identify factors to give a competitive edge
C Identify urgent problems
In the evaluation of customer information, not all

attributes and transactions should be treated equally.
Some are much more important than others. As
customers’ needs change, the evaluations will change.
CSSBB 2014
V. DEFINE IV.A.2
Customer Surveys (Continued)

Customer survey sample sizes and frequency can have
significant cost implications, and should be chosen to
balance business resources and the need to monitor
changes in the business environment.
Surveys can be developed in questionnaire form. An

adequate number would range from 25 to 30 questions.
One survey matrix is shown below:
Customer Satisfaction
Very Dissatisfied Very Satisfied
Task 1 2 3 4 5 6 7 8 9 10
On Schedule
Good Product
Friendly
Prompt
CSSBB 2014
V. DEFINE IV.A.2
Survey Pitfalls
There can be problems with the use of surveys:
C Improper survey form design

C Poorly defined survey issues
C Sampling errors or poor sampling techniques
C Ignoring nonresponses
C Treating customer perceptions as measures
C Using incorrect analysis methods
C Treating surveys as an event, not a process
C Asking nonspecific questions
C Failing to ask the right questions
C Ignoring the results or using them incorrectly
C Failing to provide feedback when necessary
C Using too many questions (25-30 questions limit)
C Using a temporary employee to conduct interviews
CSSBB 2014
V. DEFINE IV.A.2
Instruments to Gather Data

Some of the common customer information instruments
are described below:
C Surveys: A properly designed questionnaire

gathers data using a set of standard questions.
C Focus groups: A small group (3 to 12) is assembled

to explore specific topics.
C Face-to-face interviews: Individual interviews of 30-

60 minutes in length may be used.
C Satisfaction/complaint cards: The return of a card

prompts a reaction by the company.
C Dissatisfaction sources: Some methods that voice

dissatisfaction include: complaints, claims,
refunds, recalls, and returns.
C Competitive shoppers: Shoppers evaluate a

company and their competitors.
CSSBB 2014
V. DEFINE IV.A.2
Customer Service Measurement

Customer service measurements can be obtained
through the various instruments mentioned above.
Some of the more common techniques include:
C Customer surveys
C Customer visits
C Service engineer feedback
C Complaint analysis: rejects, Pareto analysis
Customers interact with the business at many levels,

each with their own specific interests.
CSSBB 2014
V. DEFINE IV.A.2
Customer Data Analysis

Comparing customer attitudes over time or between
groupings can provide insights into market niches and
changes. The results of customer feedback can be
analyzed using a variety of tools:
Statistical tests: A large number of nonparametric tests

and contingency tables can be used to determine, with
identified confidence levels, whether customer
preferences have shifted.
Line graphs: Line graphs can graphically show whether

either discrete or continuous characteristics of a
product or service are changing. In most cases, a visual
assessment can be made.
Control charts: A variety of variable or attribute charts

can also be used to display customer feedback data.
CSSBB 2014
V. DEFINE IV.A.2
Customer Data Analysis (Continued)

Matrix diagrams: A variety of matrix diagrams can be
used for examination of customer defects or complaints.
Data from matrix diagrams can be used to create a
Pareto chart or can be used directly.
Pareto analysis: Snapshots of customer defects or

rejects (Pareto charts) can be displayed at selected time
intervals to answer such questions as: Are reject
categories still of the same magnitude? Are reject
categories still in the same sequence of magnitude?
Other comparative analyses: Comparative Pareto

analysis is a powerful tool for analyzing customer data.
Charts can be compared from one time period to
another, from one supplier to another, etc. to provide
real insight into the needs of the customer.
CSSBB 2014
V. DEFINE IV.A.3
VOICE OF THE CUSTOMER/REQUIREMENTS
Critical Customer Requirements

Customers ultimately determine the value of any
product (goods and services) with their decision to buy
or not buy. These decisions are made based on a
complex system of critical customer requirements. One
must be able to determine the critical customer
requirements that influence these decisions.
Customer value consists of cost, quality, features, and

availability factors (CQFA). To prosper, a business
process must do well in at least one of these four areas
while at least meeting acceptable levels in the others. If
a business can be “best in class” in one of these four
quadrants, or above average in more than one quadrant,
they can thrive. However, the level of the bar in all of the
quadrants is constantly changing.
In addition to looking at customer’s CQFA preferences,

it also helps to understand the entire system of
customer expectations, needs, and priorities.
CSSBB 2014
V. DEFINE IV.A.3
Customer Expectations
A deep knowledge and understanding of the customer
is required in order to properly serve them. There is a
need to go beyond the sale, to uncover the subjective
factors of why the product was purchased. The
customer’s expectations can be described through an
analogy similar to Maslow’s hierarchy of human needs.
C Basic: The bare essential attributes of the product

or service should be present.
C Expected: Some attributes will be provided as a

part of the product.
C Desired: These are attributes that are worthwhile to

have, but not necessarily provided as part of the
package.
C Unanticipated: These are surprise attributes that go

beyond what the customer expects from a
purchase.
CSSBB 2014
V. DEFINE IV.A.3
Customer Needs
Customer’ needs change continuously. A product or
service that satisfied a certain need may generate new
needs for the customer. Juran lists customer needs as
follows:
C Stated needs: What the customers say they want
C Real needs: What the customer really wants
C Perceived needs: What the customer thinks is

desired
C Cultural needs: Status of the product
C Unintended needs: The customer uses the product

in an unintended manner
CSSBB 2014
V. DEFINE IV.A.3
Customer Needs
There are customer needs related to the use of a
product:
C Convenience
C Safety needs
C Product simplification features
C Communications
C Service for product failures
C Customer service
Customers’ needs change rapidly. They require new

products or services to take the place of existing or
inadequate ones.
CSSBB 2014
V. DEFINE IV.A.3
Customer Priorities
The customer will have priorities as to which of their
many expectations will be met. Services or products
that are of high priority today, may be unimportant 5
years later.
Companies can make use of customer interviews,

surveys, focus groups, phone surveys, mail surveys,
audits, sales reports, or other data gathering tools to
identify customer needs and expectations. The tools do
not have to be complicated but should ask the right
questions:
C What attributes are of value?

C How desirable is each attribute
C How do we compare with competitor’s products?
C What other features or services would be of value?
The use of priorities for customer needs can enable a

company to respond in a more timely manner.
CSSBB 2014
V. DEFINE IV.A.3
Voice of the Customer (VOC)

A detailed plan to gather and collect customer needs
and customer perceptions can be described as listening
to the voice of the customer (VOC). This enables the
organization to:
C Make decisions on products and services

C Identify product features and specifications
C Focus on improvement plans
C Develop baseline metrics on customer satisfaction
C Identify customer satisfaction drivers
Rath & Strong suggests the following process for

collecting VOC data:
C Identify customers and their needs

C Collect and analyze reactive data
C Convert collected data into customer needs
C Sort out the most important attributes
C Obtain specifications using the critical characteristics
VOC can be critical in six sigma team project selection

and measurement.
CSSBB 2014
V. DEFINE IV.A.3
Critical to Quality (CTQ) Tree

One technique that is useful in the six sigma problem
definition stage is the construction of a critical to quality
tree. This tool focuses on the key metrics of customer
satisfaction. A CTQ tree will translate the initial
customer requirements to numerical or quantified
requirements for the product or service.
Generally, it will require two or three levels to proceed

from need to drivers to CTQs. See the Figure below:
General Specific
Need Drivers CTQs

CSSBB 2014
V. DEFINE IV.A.3
Critical to Quality Tree (Continued)

The creation of the CTQ tree involves the following
steps:
1. Identify the customer.
2. Identify the customer’s need.
3. Identify the first set of basic requirements of the

customer.
4. Progress further with more levels as needed.
5. Validate the requirements with the customer.

CSSBB 2014
V. DEFINE IV.A.3
Critical Requirements
Projects can be directed at any number of CTX (critical
to X) requirements. Examples are listed below:
CTQ
Critical to quality improvement projects may include:

C Simplifying product designs
C Aligning designs with customer requirements
C Meeting current marketplace quality levels
C Exceeding current marketplace quality levels
C Exceeding reliability requirements
C Exceeding product appearance expectations
C Meeting technical requirements
C Providing products that are more durable
COQ
Cost of quality improvement projects may include:

C Reducing internal and external rejections
C Minimizing salvage and sorting operations
C Reducing warranty claims
C Reducing product and process variation
C Reducing various forms of waste
C Eliminating unnecessary inspections
CSSBB 2014
V. DEFINE IV.A.3
Critical Requirements (Continued)

CTD
Critical to delivery improvement projects may include:

C Providing exact amounts of product
C Providing service within a specific time interval
C Ensuring responses to customer questions
C Providing a product or service at the proper time
C Providing more rapid field service
C Providing cost-effective delivery methods
C Meeting customer packaging requirements
C Minimizing shipping damage
CTP
Critical to process improvement projects may include:
C Minimizing changeover times
C Reducing in-process inventories
C Minimizing product touch times
C Optimizing work cell design
C Streamlining internal work flows
C Enhancing process velocity
C Eliminating redundant operations
C Maximizing product yields
C Reducing cycle times
C Minimizing equipment downtime
C Performing value stream mapping
CSSBB 2014
V. DEFINE IV.A.3
Critical Requirements (Continued)

CTS
Critical to safety improvements may include:
C Simplifying tasks
C Mistake proofing operations
C Providing operator visual prompts
C Providing safety cut-off devices
C Using warning alarms
C Providing adequate employee training
C Providing clear written instructions
C Protecting both operators and equipment
C Making products that are user-friendly
C Preventing incorrect product use
C Providing back-up redundancies when needed
C Conducting safety reviews
C Expanding prototype testing
C Providing protective devices when applicable
C Eliminating failure-prone elements
C Meeting product disposal requirements
CSSBB 2014
V. DEFINE IV.A.3
Kano Model
The Kano model is also referred to as Kano analysis. It
is used to analyze customer requirements. The model
is based on 3 categories of customer needs:
1. Dissatisfiers (basic requirements, or “must be’s”)
2. Satisfiers (variable requirements, or “more is

better”)
3. Delighters (latent requirements)
Competition in today’s environment raises the basic

expectations of all customers. The standards of a happy
customer continue to rise. What was once considered
a delighter, may in time turn into a basic satisfier.
Improvement projects can often be selected from among

the satisfier and delighter categories. Most companies
in a competitive environment would not be around long
enough to tackle a basic requirement issue.
CSSBB 2014
V. DEFINE IV.A.3
Quality Function Deployment

Quality function deployment (QFD) is a tool that is
sometimes referred to as the “voice of the customer,” or
as the “house of quality.” Quality function deployment
has been described as a process to ensure that the
customers’ wants and needs are heard and translated
into technical characteristics.
The technical characteristics are handled by the

company through the design function, or better still,
through a cross functional team that includes sales,
marketing, design engineering, manufacturing
engineering, and operations. QFD is a tool for the entire
organization to use. It is flexible and customized for
each case and works well for manufactured products
and in the service industry.
CSSBB 2014
V. DEFINE IV.A.3
QFD (Continued)
QFD provides a graphic method of expressing
relationships between customer wants and design
features. It is a matrix that lists the attributes a
customer wants and compares it to the design features
(services that satisfy customer wants).
The collection of customer wants and expectations are

expressed through the methods available to most any
organization: surveys, focus groups, interviews, trade
shows, hot lines, etc. The house of quality is so named
because of the image used in its construction. The use
of matrices is the key to the building of the house. The
primary matrix is the relationship matrix between the
customer needs or wants and the design features and
requirements.
CSSBB 2014
V. DEFINE IV.A.3
QFD (Continued)
The construction of the house follows:
C The left side of the house has the customer needs

C The ceiling has the features and requirements
C The right side contains the customer priorities
C The foundation contains the target values
C The roof contains a design feature matrix
INTER-RELATIONSHIP
OF DESIGN FEATURES
HOW TO MEET THE

CUSTOMER WANTS
(DESIGN FEATURES)
CUSTOMER COMPARISON
NEEDS OF CUSTOMER
(WANTS) PRIORITIES
BENCHMARKED
TARGET VALUES
CSSBB 2014
V. DEFINE IV.A.3
QFD (Continued)
The benefits for using the QFD process are:
C Creates a customer driven environment

C Reduces the cycle time for new products
C Uses concurrent engineering methods
C Reduces design to manufacture costs
C Increases communications through teamwork
C Creates data for proper documentation
C Establishes priority requirements
CSSBB 2014
V. DEFINE IV.A.3
QFD (Continued)
P
N
N P P = POSITIVE INTERACTIONS
N N = NEGATIVE INTERACTIONS
P
DESIGN FEATURES
ADAPTABLE CONTENT
MOVABLE SECTIONS
COMPETITIVE
RANKINGS COMPARISON
5 = MOST IMPORTANT
COMPACT SIZE
CURRENT BOK
FEW ERRORS
0 = NO IMPORTANCE
DURABILITY
LOW PRICE
COMPETITIVE
RANKING
ASSESSMENT
BETTER
WORSE
5 = BEST
SAME
CUSTOMER 3 = AVERAGE
NEEDS 1 = WORST
COMPREHENSIVE 4 5 1 2 1 1 3 5 5 4 3 4 4 3
LOW COST 2 0 4 3 1 5 3 0 4 3 4 3 3 3
UP-TO-DATE 5 4 1 0 3 0 2 5 5 3 3 4 3 3
EASILY AVAILABLE 4 0 1 3 0 4 0 0 4 4 4 5 3 3
TEST QUESTIONS 5 5 0 2 5 1 5 5 5 4 2 5 4 4
RATINGS 65 21 36 46 35 53 70
ADVERTISING
SALES METHODS
Y MANUAL
TECHNICAL SUPPORT
PRIMER
X BOOK
TARGET VALUES
3 ERRORS/SEC
INSTRUCTOR
<$85
REVIEWER
3 HOLES
-30 F/+140 F
<6LBS
PRIMER 89 98 5.2 2 80 3 91
ASSESSMENT
COMPETITIVE
OBJECTIVE
CUSTOMER
MEASURES X BOOK 77 95 3.9 5 99 0 79

Y MANUAL 81 76 4.5 4 80 3 83
TECHNICAL ABSOLUTE 83 27 44 91 89 15 72
IMPORTANCE RANKING 3 6 5 1 2 7 4
The house of quality is flexible and customized to each

situation. Each organization will develop guidelines.
CSSBB 2014
V. DEFINE IV.A.3
QFD (Continued)
After setting the primary design characteristics the
“hows” from the house of quality can become the
“whats” of another house that depicts detailed product
design. This process is repeated with a process
planning house and then production planning house. In
this way, the voice of the customer is carried through
from design to manufacturing. See the Figure below:
Engineering Parts Key process Production

characteristics characteristics operations requirements
characteristics
characteristics
Engineering
Key Process
Operationss
Attributes
Customer
Parts
House of Parts Process Production

quality deployment planning planning
CSSBB 2014
V. DEFINE IV.A.3
Simplified Approaches
The quality function deployment (QFD) approach is a
rigorous process that requires considerable time and
effort to implement. Time pressures may force a quicker
resolution. In these cases, a simplified approach using
a cause and effect matrix or perceptual map can be
effective.
Cause-and-Effect Matrix
A cause-and-effect matrix can help prioritize the

importance of key process input variables (KPIVs). This
technique is described in the Primer.
CSSBB 2014
V. DEFINE IV.A.3
Simplified Approaches (Continued)

Perceptual Maps
The following steps describe how to determine the

appropriate questions to help quantify and prioritize the
needs of the customer.
C Conduct brainstorming sessions to identify a wish

list of features and/or problem resolutions.
C Rank the brainstorming session items and consider

the highest ranking items for possible customer
survey questions.
C Construct the set of questions, being careful not to

bias responses on customer satisfaction or
customer importance with the wording.
C Collect a numerical ranking (e.g. 1-5 on both

satisfaction and importance) for each item and plot
them on a perceptual map.
An example is shown in the Primer.

CSSBB 2014
V. DEFINE IV.A.3
Input and Output Variables

Measurements of process inputs and outputs can be
used to optimize the process being measured. Process
inputs may be raw materials, human resources, or
services. All inputs have some quantifiable
measurement, including human effort and skill level.
Process input requirements should be stated so that key

measures of input quality can be controlled.
Measurements within the process can also be used as
effective controls. Once process capabilities are known,
output measures can be used to monitor if the process
has remained in control.
Feedback from downstream process measurements can

be used to improve an upstream process.
CSSBB 2014
V. DEFINE IV.A.3
Input and Output Variables (Continued)

A key concept in six sigma methodology is the SIPOC
high level process map in which SIPOC stands for
suppliers, inputs, process, outputs, and customers.
This diagram identifies major core processes.
SYSTEMS & PROCESSES
INPUTS OUTPUTS
SUPPLIERS CUSTOMERS
FEEDBACK FEEDBACK
KNOWLEDGE
BANK
The advantages of using a SIPOC model include:
C A single display of cross functional activities

C Provides a “big picture” perspective
C A framework applicable to all organizations
The diagram can also be adapted to a number of

essential support processes.
CSSBB 2014
V. DEFINE IV.A.3
SIPOC
The SIPOC process map is designed to be a high level
process view with 4 - 7 displayed steps. This is a flow
chart viewed at the 50,000 foot level. The map enables
all team members to view the process in the same light.
The following development steps can be used:
C Have the team create the process map

C The process may have 4 or 5 key steps
C List the outputs of the process
C List the customers of the output of the process
C List the inputs of the process
C List the suppliers of the process
C Identify some preliminary customer requirements
C Involve the team leader and other stakeholders
CSSBB 2014
V. DEFINE IV.A.3
SIPOC (Continued)
An example of a SIPOC diagram for the CSSBB Primer
follows:
Suppliers Inputs Process Outputs Customers

Paper suppliers BOK from ASQ See below Primers ASQ members
Binder suppliers Written and Solution texts Libraries

electronic texts
Tab suppliers Charts and graphics Question CDs Test takers
Outside Printers Reviews Instructor PDF Education chairs

CDs
Authors & Reference materials Company trainers

Reviewers
Process
Authors Publisher Materials Orders are Materials

submit edits the assembled processed are sent to
materials materials into a book customer
CSSBB 2014
V. DEFINE IV.B.1, 2, 3
PROJECT BUSINESS CASE/CHARTER CONTENT
Project Charter
Project Charter is presented in the following topic areas:
C Charter content
C Charter negotiation
C Project management
C Project performance measures
Charter Content
A charter is a written document that defines the team’s
mission, scope of operation, objectives, time frames,
and consequences. Charters can be developed by top
management and presented to teams, or teams can
create their own charters and present them to top
management.
CSSBB 2014
Charter Content (Continued)

The charter begins with a purpose statement. This is a
one or two line statement explaining why the team is
being formed. The charter should also identify the team
objectives. Objectives should always be stated in
measurable terms.
The charter should also define the operating scope,

identifying the organizational or operational boundaries.
A charter provides the following advantages:
C Eliminates any confusion

C Defines the subject boundaries
C Identifies areas which should not be addressed
C Identifies the deliverable product
C Provides a basis for team goal setting
C Authorizes the team to collect relevant data
C Provides access to necessary resources
C Approves time for a team to address problems
CSSBB 2014
Charter Content (Continued)

Moen suggests that a team project charter should
contain the following key points:
C Business case (financial impact)

C Problem statement
C Project scope (boundaries)
C Goal statement
C Role of team members
C Milestones/ deliverables (end products)
C Resources required
Identifying the above details, in written form, will

provide a constant and consistent target for the team.
CSSBB 2014
Business Case
The business case is a short summary of the strategic
reasons for the project. The general rationale for a
business case would normally involve quality, cost, or
delivery of a product with a financial justification. Moen
suggests that there are four basic activities:
C Design of a new product

C Redesign of an existing product
C Design of a new process
C Redesign of an existing process
A project improvement team should follow typical

financial department justification guidelines. Projects
which do not show a significant financial impact to the
company should be stopped or eliminated as soon as
possible.
CSSBB 2014
Problem Statement
A problem statement will detail the issue that the team
wants to improve. The problem statement should be
crafted to be as descriptive as possible. That is, how
long has the problem existed, what measurable item is
affected, what is the business impact, and what is the
performance gap.
The problem statement should be neutral, to avoid

jumping to conclusions. The problem statement should
include a reference to a baseline measure for guidance.
The collection of good data and process performance

measurements will provide a picture of the areas in the
company that are in greatest need of improvement.
CSSBB 2014
Project Scope
The project scope refers to the boundaries of the
project. It is an attempt to outline the range of the
team’s activities.
Goal Statement
The goal statement will be created and agreed to by the
team and team champion. The goals will should be
achievable within a 120 to 160 day period.
CSSBB 2014
Milestones/Deliverables
For any well managed project, a set of stages or
milestones are used to keep the project on track and to
help bring a project to completion. A typical milestone
chart might be:
Day 0: Start team activities

Day 1: Start the define portion of the project
Day 40: Begin the measure portion of the project
Day 80: Start the analysis portion of the project
Day 120: Start the improvement phase of the project
Day 160: Conclude the project with a management
presentation
>Day 160: Control elements in progress
CSSBB 2014
Resources Required
The resources required for a project must be detailed.
Typical resources required for a project could include:
C Qualified people C Machine time

C Equipment C Phones and faxes
C Machinery C Computer equipment
C Lab or office space C Utilities, etc.
The six sigma define phase should provide the following

information for the team members:
1. Importance of the project

2. Goals of the project
3. Knowledge of the team members
4. Scope of the project (time and resources)
5. The key process involved
6. Current baseline metrics
7. What the customer requirements are
CSSBB 2014
PROJECT BUSINESS CASE/CHARTER NEGOTIATION
Charter Negotiation
The project charter can and presented to the team by
upper management. However, the project team might be
closer to the facts and might propose a different mode
of attack than envisioned. Hence, charter negotiation
may be required.
Consider the following examples:
1. Objectives: The final customer product or internal

process may require a substantial redesign.
2. Scope: The boundaries of the project could require

expansion .
3. Boundaries: The project team may discover that

additional areas should be included.
4. Resources: Excessive project resources are rarely

provided.
CSSBB 2014
PROJECT BUSINESS CASE/CHARTER NEGOTIATION
Charter Negotiation (Continued)

5. Project transition: The transition of a project to
normal company controls may necessitate either a
time extension or additional monitoring.
6. Project closure: The improvement team could

discover that related processes or products need
the same type of effort.
Obviously, both the project team and upper

management are interested in success, not failure.
There should be a willingness on the part of both parties
to negotiate a number of project details.
CSSBB 2014
PROJECT BUSINESS CASE/PROJECT MANAGEMENT
Project Management and Benefits

Breyfogle states that the successful black belt candidate
should have the “ability to manage projects and reach
closure: a persistent drive toward meaningful bottom-
line results and timely completion of projects.”
Project management roles and responsibilities of six

sigma black belts include the following:
C Lead the (cross functional) team

C Possess interpersonal and meeting facilitation skills
C Develop and manage a detailed project plan
C Schedule and lead team meetings
C Sustain team motivation and stability
C Communicate project benefits to stakeholders
C Track and report milestones and tasks
C Interface with organizational departments
CSSBB 2014
PROJECT BUSINESS CASE/PROJECT MANAGEMENT
Project Management and Benefits (Cont’d)

Projects need charters, plans, and boundaries. Projects
may be selected from a broad range of areas including:
C Improved process capabilities

C Customer complaints
C Reduction of internal defects
C Cost reduction opportunities
C Supplier related improvements
C Lean manufacturing principles
C Improved work flows
C Administrative/service improvements
C Cycle time reductions
C Market share growth
The actual project should be consistent with company

strategies and be rather specific. Examples include:
C Reduce scrap costs in the finishing department for

closure defects by 20% by May 2013.
C Improve market share for left handled widgets by

10% for the 2013 fiscal year.
CSSBB 2014
V. DEFINE IV.B.4
PROJECT BUSINESS CASE/GOALS
SMART
The acronym S.M.A.R.T. is an outline used for project
management and personal development. The original
five elements were directed at achieving corporate and
departmental goals:
C Specific: the specific targeted area for improvement

C Measurable: quantify an indication of improvement
progress
C Assignable: indicate the responsible individual or
team
C Realistic: state realistic results using available
resources
C Time-related: specify when the results can be
achieved
CSSBB 2014
V. DEFINE IV.B.4
SMART (Continued)
See the table below for more detail:
Letter Common Term Other Terms

S Specific Significant, sustainable, simple
M Measurable Meaningful, manageable,
motivational
A Assignable Achievable, attainable, ambitious,
acceptable, aggressive,actionable
R Realistic Relevant, resourced, results-
based, reasonable, results-
oriented
T Time - related Time-framed, time-based, time-
specific, time sensitive, time
limited, time boxed
CSSBB 2014
V. DEFINE IV.B.4
SMART (Continued)
Specific: A specific goal will usually answer five
questions:
C Who is involved?
C What will be accomplished?
C Why is this to be done?
C Where is the location?
C Which requirements and constraints are apparent?
Measurable: A measurable goal will answer questions

like:
C How much?
C How many?
C How will one know when the goal is accomplished?
C How will the indicators be quantified?
CSSBB 2014
V. DEFINE IV.B.4
SMART (Continued)
Achievable: An achievable goal answers the question:
C How can the goal be accomplished?
Relevant: A relevant goal answers the following

questions:
C Is the goal worthwhile?

C Is this the right time?
C Are we (or I) the right person?
C Does the goal align with our other needs?
C Is it applicable in the current environment?
Time-Bound: Goals should be grounded with a time

element like:
C What can be done immediately?

C What can be done in a designated time?
C When can it be totally accomplished?
CSSBB 2014
V. DEFINE IV.B.5
PROJECT BUSINESS CASE/PROJECT MEASUREMENTS
Project Performance Measurements

It should be noted that key project measurements are
also difficult to determine until the project selection and
charter processes are complete.
Careful selection of project measures ensure the overall

success of the six sigma project. most six sigma
projects deal with time and money issues. Project
measures for both cost and revenue factors must be
included.
The project budget must be reasonable, attainable, and

based on estimates of the tasks to be accomplished.
CSSBB 2014
V. DEFINE IV.B.5
Project Performance Measurents (Contd)

Revenue factors include:
C Income from additional generated sales

C Reduced losses in a number of areas
Cost factors included in the budget are:
C Manpower and labor costs

C Materials
C Equipment costs, rentals, leases, etc.
C Subcontracted work or fees
C Overhead or consulting charges
C Reserve or contingency funds
CSSBB 2014
V. DEFINE IV.B.5
Project Performance Measurents (Contd)

The budget becomes the measurement standard for
project costs. Some organizations use a budget that,
once approved, is fixed for the duration of the project.
Estimates of project revenues and costs are described
by four types of measurements:
C Budget: The approved written plan of the total

costs expressed in dollars.
C Forecast: The predicted total revenues and costs.
C Actual: Revenues and costs that have occurred.
C Variance: The difference between the budgeted and

actual revenues and costs.
During project implementation, the actual cash flows are

documented by the accounting department and the
information is provided to the team leader.
CSSBB 2014
V. DEFINE IV.B.5
Project Decision Analysis

In addition to a benefit cost analysis for a project, a
decision to proceed must also include an evaluation of
the risks associated with the project. Risk areas
include:
Business risks Insurable risks
C Technology changes C Property damage

C Competitor actions C Indirect loss
C Material shortages C Legal liability
C Health & safety issues C Personnel injury
C Environmental issues
After the risk areas are identified, each is assigned a

probability of occurrence and the consequence of risk.
The project risk factor is then the sum of the probability
of occurrence and the consequence of risk.
CSSBB 2014
V. DEFINE IV.B.5
Project Portfolio Analysis

When there is a portfolio of project opportunities and
limited resources, management must make decisions to
approve, postpone, or reject project proposals. These
decisions are based on the project benefit cost analysis,
and the evaluation of the risks.
Regardless of the method(s) used for evaluating the

project(s), ROA, ROI, NPV, IRR, payback period, risk
factor, etc., each project will be compared against the
other projects in the portfolio, and against a criteria
amount established by an organization.
Typically, the project with the highest financial benefit,

shortest payback period, or lowest risk factor is chosen
for implementation first.
CSSBB 2014
V. DEFINE IV.B.6
PROJECT BUSINESS CASE/PROJECT REVIEWS
Project Reviews
The project review is a formal and documented critique
conducted by a committee of qualified company
personnel. The project review extends over all phases
of development, from inception to completion. Some of
the fundamental review topics include:
C The adequacy of personnel, time, equipment and

money
C The project effectiveness as determined by internal

and external information
C The effectiveness and reliability of corrective

actions
C The true quality level of the delivered product and/or

service
Results of project reviews will be retained along with the

other project documentation and archived for future
reference.
CSSBB 2014
V. DEFINE IV.B.6
Measurement of Project Activity

During the planning stage of project management, the
monitoring and measuring requirements should be
defined. In most cases, upper management requires
scheduled briefing sessions during the project. These
sessions range in depth from an overview of the project
milestones to comprehensive reports.
CSSBB 2014
V. DEFINE IV.B.6
Measurement of Project Activity (Cont’d)

The project monitoring plan should address the
following areas:
C What is being monitored

C The purpose of the monitoring
C Timing or frequency of reporting
C Method of reporting
C Procedure for indicating a need for assistance
C Criteria for reporting of unusual events
C The channel for feedback
C Assignment of feedback loop responsibilities
C Action to be taken when exceptions occur
CSSBB 2014
V. DEFINE IV.B.6
Measurement of Project Activity (Cont’d)

The feedback loop defines the methods for monitoring
and adjusting the process if results are different than
desired. The success or failure of a project is measured
in the following dimensions:
C Were the specified goals and objectives achieved?

C Within the time deadlines?
C At or below cost constraints?
C Utilizing the allocated resources?
Well executed project plans meet all of the above

criteria. However, it is possible for a project to be
considered a success, even when the project is late,
over budget, or does not meet the stated objectives.
Nearly every project encounters unanticipated events or

problems, but this is not an acceptable excuse for
failure to meet the performance standards.
CSSBB 2014
V. DEFINE IV.C
PROJECT MANAGEMENT TOOLS
Project Management Tools

Project management tools is presented in the following
areas:
C Project plan elements

C Work breakdown structure
C Project documentation
Project Plan Elements

A project is a series of activities and tasks with a
specified objective, starting and ending dates, and
necessary resources. Resources consumed by the
project include time, money, people, and equipment.
Project management includes project planning and
implementation to achieve:
C The specified goals and objectives,

C At the desired performance or technology level,
C Within the time and cost constraints,
C While utilizing the allocated resources.
CSSBB 2014
V. DEFINE IV.C
PROJECT MANAGEMENT TOOLS
Project Plan Elements

The stages of project management are:
C Planning - deciding what to do

C Scheduling - deciding when to do it
C Controlling - ensuring the desired results
Key project management elements include:
C Identifying schedule time limits

C Allocation of functional responsibilities
C Establishing continuous reporting methods
C Selecting applicable tradeoff methodologies
C Measuring accomplishments against plans
C Identifying problems early
C Applying corrective action to problems
C Knowing when objectives will be met or exceeded
C Improving capabilities for future projects
CSSBB 2014
V. DEFINE IV.C.3
PROJECT MANAGEMENT TOOLS/WBS
Work Breakdown Structure

The work breakdown structure (WBS) is a detailed plan
which expands the project (statement of work) into a
detailed listing of activities required to complete the
project. The project team leader is usually responsible
for completion of the work breakdown structure.
Each project task is divided into smaller activities, until

the level is reached in which each element is under one
identifiable, individual or group responsibility. Each
activity is assigned a duration, along with the
interrelationships between activities. After the material,
equipment, and personnel needs are established for
each activity, the project network can be scheduled.
The schedule is a balance between time, resource, and
cost constraints.
CSSBB 2014
V. DEFINE IV.C.3
PROJECT MANAGEMENT TOOLS/WBS
Work Breakdown Structure (Continued)

If the time constraints are fixed, such as a set deadline,
then the resource constraints must be flexible to
accommodate variations in the project. Time delays
during the project require offsetting increases in
resources and costs to maintain a set deadline.
Time constrained projects are assumed to have

unlimited resources. In a practical sense, this means
that resources must be acquired until they reach a level
that will ensure an on time project completion.
Most projects have relatively fixed levels of resources

including manpower and equipment. In resource
constrained projects, the objective is to meet the project
duration requirements, without exceeding resources.
For those resources with time conflicts, it may become

necessary to schedule planned parallel activities as
sequential tasks, using existing slack time and possibly
delaying the project completion date.
CSSBB 2014
V. DEFINE IV.C
PROJECT MANAGEMENT TOOLS/PLANNING TOOLS
Planning Tools
Project planning tools include developing and analyzing
the project time line, determining required resources,
and estimating costs. Common techniques for
evaluating project time lines include PERT charts, Gantt
charts, and critical path method (CPM). The work
breakdown structure (WBS) helps identify detailed
activities for the plan.
CSSBB 2014
V. DEFINE IV.C
Network Planning Rules

Common applications of network planning include the
program evaluation and review technique (PERT), the
critical path method (CPM), and Gantt charts. The
following network rules are widely followed:
C Before an activity may begin, all activities preceding

it must be completed.
C Arrows imply logical precedence only.
C Any two events may be directly connected by only

one activity.
C Event numbers must be unique.
C The network must start at a single event, and end at

a single event.
CSSBB 2014
V. DEFINE IV.C
PERT
The program evaluation and review technique (PERT)
requirements are:
C All individual project tasks must be included.
C Events and activities must be sequenced in the

network to allow determination of the critical path.
C Time estimates must be made for each activity in

the network, and stated as three values: optimistic,
most likely, and pessimistic elapsed times.
C The critical path and slack times for the project are
calculated. The critical path is the sequence of
tasks which requires the greatest expected time.
The slack time, S, for an event is the latest date an event

can occur or can be finished without extending the
project, (TL), minus the earliest date an event can occur
(TE). For events on the critical path, TL = TE, and S = 0.
S = TL - TE
CSSBB 2014
V. DEFINE IV.C
PERT (Continued)
Advantages of using PERT include:
C Planning can identify interrelationships between

tasks and problem areas.
C The probability of achieving the project deadlines

can be determined.
C Changes in the project can be evaluated to

determine their effects.
C A large amount of project data can be organized and

presented in a diagram for use in decision making.
C PERT can be used on unique.
Disadvantages of using PERT include:
C The complexity of PERT presents problems.
C More data is required as network inputs.

CSSBB 2014
V. DEFINE IV.C
PERT (Continued)
Each starting or ending point for a group of activities on
a PERT chart is an event, called a node, and is denoted
as a circle with an event number inside. Events are
connected by arrows with a number indicating the
amount of time required to go between events. An event
at the start of an arrow must be completed before the
event at the end of the arrow may begin. The expected
time between events, te is given by:
t o + 4t m + tp
te =
6
Where: to is optimistic time, tm is most likely time, and

tp is pessimistic time.
CSSBB 2014
V. DEFINE IV.C
PERT Chart Example

An example of a PERT chart for a company seeking ISO
9001 certification is shown below.
2 6
6
4
4 1
0 1 5 7 8 9 10
3 1 3
4 8 4 6 2
3
12
4
TASK ACTIVITY DURATION

0 ISO 9001 Certification Objective
0-1 Planning 4 weeks
1-2 Select Registrar 4 weeks
1-3 Write Procedures 8 weeks
1-4 Contact Consultant 3 weeks
2-6 Schedule Audit 6 weeks
3-5 Write Quality Manual 4 weeks
4-7 Consultant Advising 12 weeks
5-6 Send Manual to Auditor 1 week
5-7 Perform Training 6 weeks
6-8 Auditor Review Manual 4 weeks
7-8 Internal Audits 2 weeks
8-9 ISO Audit 1 week
9 - 10 Corrective Action 3 weeks
10 Certification Milestone
CSSBB 2014
V. DEFINE IV.C
PERT Chart Example (Continued)

To calculate the critical path, add the durations for each
possible path through the network. Which path is the
critical path? The possible paths and total times are:
PATH TOTAL TIME

0-1-2-6-8-9-10 22 weeks
0-1-3-5-6-8-9-10 25 weeks
0-1-3-5-7-8-9-10 28 weeks
0-1-4-7-8-9-10 25 weeks
The critical path is 0-1-3-5-7-8-9-10. Projects not on the

critical path may be delayed by an amount equal to the
slack time without delaying project completion.
What is the slack time for event 6? First, one observes

that event 6 is not on the critical path calculated above.
The earliest, TE, that event 6 can occur is at the 17th
week, found by path 0-1-3-5-6. Event 8 is on the critical
path and occurs at the 24th week. Since task 6-8 takes
4 weeks, the latest, TL, that event 6 can take place is the
20th week. Using the formula for slack time:
S = TL - TE = 20 - 17 = 3 weeks for event 6.

CSSBB 2014
V. DEFINE IV.C
Critical Path Method (CPM)

The critical path method (CPM) is very similar to PERT,
except PERT is event oriented, while CPM is activity
oriented. Unique features of CPM include:
C The emphasis is on activities
C The time and cost factors for each activity are

considered
C Only activities on the critical path are contemplated
C Activities with the lowest crash cost (per

incremental time savings) are selected first
C As an activity is crashed, it is possible for a new

critical path to develop
For each activity there is a normal cost and time

required for completion. To crash an activity, the
duration is reduced, while costs increase. Crash, in this
sense, means to apply more resources to complete the
activity in a shorter time.
CSSBB 2014
V. DEFINE IV.C
CPM (Continued)
Using information from the PERT chart example, and
adding crash times and costs, yields:
B.4 E.6 J.4
H.1
A.4 C.8 F.4 I.6 K.2 L.1 M.3
D.3 G.12
Note that each activity arrow on the PERT chart example

becomes a circle on the CPM example. The letter
indicates the activity and a number. The number, in this
example, is the normal activity duration in weeks. The
critical path is indicated by the thicker arrows, along
path A-C-F-I-K-L-M. A CPM example with crash cost
data is shown in the Primer.
CSSBB 2014
V. DEFINE IV.C
CPM Example (Continued)

The CPM time-cost trade-off for the Primer example can
be represented graphically as shown below:
65000
60000
Cost ($)
55000
50000
45000
18 20 22 24 26 28 30
Time (weeks)
The graph illustrates that crashing activities beyond a

certain level, increases costs without time reductions.
CSSBB 2014
V. DEFINE IV.C.1
PROJECT MANAGEMENT TOOLS/GANTT CHARTS
Gantt Charts (Bar Charts)

Gantt charts (bar charts) display activities or events as
a function of time (or cost). Each activity is shown as a
horizontal bar with ends positioned at the starting and
ending dates for the activity.
Advantages of Gantt charts include:
C The charts are easy to understand.

C Each bar represents a single activity.
C It is simple to change the chart.
C The chart can be constructed with minimal data.
C Program task progress versus date is shown.
CSSBB 2014
V. DEFINE IV.C.1
PROJECT MANAGEMENT TOOLS/GANTT CHARTS
Gantt Charts (Continued)

Disadvantages of Gantt charts include:
C They do not show interdependencies of activities.
C The effects of early or late start of an activity are not

shown.
C There is no means to indicate the variation in

expected time to complete an activity.
C The details of an activity are not indicated.
C There is little predictive value to this presentation.
The bar charts indicate only an ambiguous description

of how the project as a system reacts to changes. The
network relationship between activities which are
indicated in PERT and CPM charts, are not shown in the
Gantt chart. An example of a Gantt is shown in the
Primer.
CSSBB 2014
V. DEFINE IV.C.2
PROJECT MANAGEMENT TOOLS/PROJECT DOCUMENTATION
Project Documentation
The project time line is the most visible yardstick for
measurement of project activities.
Both early and late projects have the opportunity for

poor quality compared to projects on schedule.
Methods for planning, monitoring, and controlling

projects range from manual techniques (using plain
paper, graph paper, grease boards, and colored
magnetic markers) to computer software.
CSSBB 2014
V. DEFINE IV.C.2
Project Documentation (Cont’d)

Advantages of manual project management methods
include:
C Ease of use
C Low cost
C Best for monitoring schedules and timing of events
C A hands-on feel for the status of the project
C Easily customized for the specific project needs
C Minimal training requirements
Advantages of automated project management methods

include:
C Able to model what-if scenarios

C Able to show the impact of alternate options
C Can present information in a variety of formats
C Schedules are automatically calculated
C Variances from plan are known in almost real time
C Project status reports are easier to generate
C People at different locations can share data
C Projects can be easily summarized
C Some data collection activities can be automated
CSSBB 2014
V. DEFINE IV.C.2
Milestones Reporting
Milestones are significant points in the project which are
planned to be completed at specific points in time.
Milestones typically occur at points where they act as a
gate for a go/no-go decision to continue the project.
The project team leader would be expected to make a

presentation to management at each major milestone.
The date and time for the milestone and the milestone
activity are set very early on in the project planning
phase. Once set and approved, the milestones are not
normally subject to change or negotiation.
CSSBB 2014
V. DEFINE IV.C.2
Project Report
The final report is the report card on project
performance for completion of objectives, comparison
of actual benefits and costs with budgets, and measures
of major activity completion dates versus milestones.
The next step of project closure is postmortem analysis.
An analysis of what went well and what went wrong is
used as a learning tool for future projects. The intent is
to avoid making the same mistakes, and to benefit from
effective processes.
CSSBB 2014
V. DEFINE IV.C.2
Document Archiving
The final project stage is document archiving. This
includes test data, traceability of materials, key process
variables, and reports generated during the project. The
documents must be complete and organized. Storage
requirements include:
C Protection from damage and other deterioration

C Security of access
C Retrievability within a reasonable period
C Adequate identification
C Consideration of duplicate copies at different sites
C Use of a medium with a sufficiently long life
CSSBB 2014
V. DEFINE IV.C.4
PROJECT MANAGEMENT TOOLS/RACI MODEL
RACI Model
One project management tool is RACI. The key
responsibility roles in the RACI methodology are:
Responsible: Who will achieve the task.
Accountable (also Approve): The person who is

ultimately responsible for the task completion.
Consulted: These are the individuals whose opinions

are sought.
Informed: These are the individuals who are kept

abreast of progress.
CSSBB 2014
V. DEFINE IV.C.4
PROJECT MANAGEMENT TOOLS/RACI MODEL
RACI Model (Continued)

There are other alternative acronym forms of this model:
RAM: Responsibility Assignment Matrix
RASCI: RACI with the term Support added
RAPID: Recommend, Agree, Perform, Input, Decide
RACI-VS: A version with Verify and Signatory added
CAIRO: The O stands for Omitted (out of the loop)
DACI: Driver, Approver, Contributor, Informed

CSSBB 2014
V. DEFINE IV.D
ANALYTICAL TOOLS
Analytical Tools
Analytical tools include:
C Affinity diagrams
C Tree diagrams
C Prioritization matrices
C Matrix diagrams
C Activity network diagrams
CSSBB 2014
V. DEFINE IV.D.1
ANALYTICAL TOOLS
Affinity Diagrams
Affinity diagrams uses an organized method to gather
facts and ideas to form developed patterns of thought.
It can be widely used in the planning stages of a
problem to organize the ideas and information.
The steps can be organized as follows:
C Define the problem under consideration

C Have 3" x 5" cards or Post-it® notes available
C Enter ideas, facts, and opinions, on the cards
C Place the cards or notes on a wall
C Arrange the groups into similar thought patterns
C Develop a category for each similar group of ideas
C Borders can be drawn around each affinity group
The affinity diagram is also referred to as the KJ Method.

CSSBB 2014
V. DEFINE IV.D.1
ANALYTICAL TOOLS
Example Affinity Diagram

H O W T O PREPARE FO R A CSSBB EX AM
RESOURCES
OBTAIN KNOWLEDGE
GET CSSBB Primer
WATCH VIDEO PRESENTATION
TAKE OTHER CSSBB COURSES

ENROLL IN
AN ON-LINE COURSE
STUDY RELATED MATERIALS

GET OTHER
CSSBB TEXTBOOKS STUDY IN GROUPS HAVE A TUTOR
TAKE UNIVERSITY LEVEL

COURSES IN SIX SIGMA
PREPARATION
HAVE A Q & A SOURCE
TEACH CSSBB SUBJECTS
USE PRACTICAL MOTIVATION

EXPERIENCE
MOTIVATE SELF GET BONUS
STUDY INTENSIVELY
START EARLY LISTEN TO SUCCESSFUL

CSSBB STUDENTS
STUDY 1 SUBJECT BE AROUND OTHERS

AT A TIME WHO ARE POSITIVE
PRIDE
STUDY CSSBB TESTS
MAKE YOUR OWN PUMP YOURSELF UP

CSSBB EXAMS
CSSBB 2014
V. DEFINE IV.D.2
ANALYTICAL TOOLS
Tree Diagrams
The tree diagram is a systematic method to outline all
the details needed to complete a given objective. The
tree diagram can also be referred to as a systematic
diagram. It is an orderly structure similar to a family tree
chart or an organization chart. The organization is by
levels of importance (i.e., why - how, goals - means).
The tree diagram can be used to:
C Develop the elements for a new product

C Show the relationships of a production process
C Create new ideas in problem solving
C Outline the steps to implement a project
One way to organize a tree diagram is provided in the

Primer.
CSSBB 2014
V. DEFINE IV.D.2
ANALYTICAL TOOLS
An Example Tree Diagram

ASK FOR
BONUS
BE AROUND LISTEN TO ASK FOR

OTHERS WHO SUCCESSFUL HELPFUL TIPS
EXAMINE ARE POSITIVE CSSBBS
MOTIVATION
REWARD
YOURSELF FOR EACH PRIDE
MOTIVATE
STEP
YOURSELF
STUDY OLD MAKE UP YOUR

CSSBB TESTS OWN CSSBB EXAMS
USE START
TEACH CSSBB STUDY EARLY
NEED TO PRACTICAL
SUBJECTS BOK 1 - 2 YEARS
PREPARE EXPERIENCE
TAKE UNIVERSITY
PASS THE HAVE A CONTACT
LEVEL COURSES
CSSBB EXAM SOURCE FOR Q/A
IN SIX SIGMA
HAVE A STUDY VIA

TUTOR TUTOR
OBTAIN
KNOWLEDGE ATTEND CSSBB STUDY IN STUDY
REFRESHER A GROUP AT HOME
TAKE CSSBB RESTUDY SEMINAR

SEMINARS MATERIALS
OBTAIN CRITIQUE STUDY

VIDEOS VIDEOS VIDEOS
NEED GET CSSBB CALL ASQ GET OTHER

RESOURCES PRIMER FOR BOK TEXTBOOKS
CSSBB 2014
V. DEFINE IV.D.4
ANALYTICAL TOOLS
Prioritization Matrices
To counter the heavy mathematical emphasis of the
matrix data-analysis, the prioritization matrices
approach was developed. There are several matrices to
use in this approach. The answers that result from
these matrices will be the basis for decision-making.
To use the prioritization matrices, the key issues and

concerns have been identified and alternatives have
been generated. The need is to determine the option to
use. Three types of prioritization matrices can be used:
C The full analytical criteria method
C The consensus criteria method
C The combination I.D./matrix method

CSSBB 2014
V. DEFINE IV.D.4
ANALYTICAL TOOLS
Prioritization Matrices Example

The Criteria Composite Ranking (4 People)
Total
A. Work Experience 0.05 + 0.10 + 0.10 + 0.20 = 0.45
B. Have Tutor 0.10 + 0.20 + 0.30 + 0.10 = 0.70
C. Study In Group 0.15 + 0.10 + 0.05 + 0.20 = 0.50
D. Attend Refresher 0.25 + 0.20 + 0.20 + 0.30 = 0.95
E. Study Tests 0.15 + 0.15 + 0.25 + 0.10 = 0.65
F. High Motivation 0.30 + 0.25 + 0.10 + 0.10 = 0.75
1.00 1.00 1.00 1.00 = 4.00
Completed Rank Order Scores
Criteria 0.45 0.70 0.50 0.95 0.65 0.75

Work Have Study Attend Study High Total
Factors Experience Tutor Group Refresher Tests Motivation
Deployment 2(0.45) 1(0.70) 2(0.50) 1(0.95) 2(0.65) 1(0.75) 5.6
Business
4(0.45) 2(0.70) 1(0.50) 3(0.95) 3(0.65) 2(0.75) 10
Management
Team
6(0.45) 4(0.70) 6(0.50) 5(0.95) 5(0.65) 6(0.75) 21
Management
Define
5(0.45) 3(0.70) 3(0.50) 2(0.95) 1(0.65) 3(0.75) 10.7
Six Sigma
*Measure
10(0.45) 9(0.70) 9(0.50) 7(0.95) 9(0.65) 8(0.75) 33.8
Six Sigma
*Analyze
7(0.45) 8(0.70) 8(0.50) 10(0.95) 10(0.65) 9(0.75) 35.5
Six Sigma
*Improve
9(0.45) 10(0.70) 10(0.50) 9(0.95) 8(0.65) 7(0.75) 35.1
Six Sigma
Control
1(0.45) 5(0.70) 4(0.50) 4(0.95) 4(0.65) 4(0.75) 15.4
Six Sigma
Lean
3(0.45) 6(0.70) 5(0.50) 6(0.95) 7(0.65) 5(0.75) 22.1
Enterprise
*DFSS 8(0.45) 7(0.70) 7(0.50) 8(0.95) 6(0.65) 10(0.75) 31

CSSBB 2014
V. DEFINE IV.D.3
ANALYTICAL TOOLS
Matrix Diagrams
Matrix diagrams show the relationship between
objectives and methods, results and causes, tasks and
people, etc. The objective is to determine the strength
of relationships between a grid of rows and columns.
The intersection of the grid will clarify the problem
strength. There are several basic types of matrices:
C L-type...elements on the Y-axis and the X-axis
C T-type...2 sets of elements on the Y-axis, split by a set

of elements on the X-axis.
C X-type...2 sets of elements on both the Y and X-axis
C Y-type...2 L-type matrices joined at the Y-axis to

produce a matrix design in 3 planes
C C-type (3-D matrix)...2 L-type matrices joined at the Y-

axis, but with only 1 set of relationships indicated in
3-dimensional space (the use of a computer software
package is recommended for this type)
Variations of the above matrices exist.

CSSBB 2014
V. DEFINE IV.D.3
ANALYTICAL TOOLS
X-Type Matrix Example
 Cause 1 Î Æ
Æ Î Cause 2 Î
Cause 3  
Cause 4
Mach 1 Mach 2 Mach 3 Dept 1 Dept 2 Dept 3 Dept 4 Dept 5
Problem 1
 Problem 2 
Problem 3 Î Î
Æ Î  Problem 4 Æ
Problem 5
Problem 6 Æ
Æ Strong Relationship
 Relationship
Î Possible Relationship
CSSBB 2014
V. DEFINE IV.D.5
ANALYTICAL TOOLS
Activity Network Diagrams

The arrow diagram is the original Japanese name for
this tool. The activity network diagram incorporates a
lot of PERT and CPM techniques in its usage. The
activities, milestones, and critical times must be
developed and then drawn onto a chart. The chart will
then provide a tool to help monitor, schedule, modify,
and review the project.
An implementation sequence is described in the Primer.

CSSBB 2014
V. DEFINE IV.D.5
ANALYTICAL TOOLS
Activity Network Diagrams Example

Key 1
Earliest 0 10
Task Start Finish 10 Determine need
Length 0 10 for CSSBB
(days)
Latest
Start Finish
2
10 20
10 Obtain CSSBB BOK
10 20
Critical
Path 3
20 40 Locate five
20
20 40 outside authors
40 50 Make
4 10 external
50 60
assignments
Make internal assignments
Move in
40 80 existing
40 5
40 45 material
40 80
5 External
45 50 authors
complete
45 75
6 30 material
50 80
50 70
20
60 80
Internal authors 7
complete material 80 90
10 Proof major
80 90 content
90 120 8
Develop 30 Create 90 100
test 90 120 index 10
questions 110 120
9 10
0 0
11 120 125 Obtain

5 approvals
120 125
12
Publish material
CSSBB 2014
V. DEFINE
QUESTIONS
5.3. A six sigma project directed at critical to safety (CTS) concerns

could include:
a. Increasing process flow velocity

b. Reducing various forms of waste
c. Exceeding product appearance expectations
d. Providing operator visual prompts
5.6. Which of the new quality management tools is used to organize

facts and data about an unfamiliar subject or problem?
a. The affinity diagram

b. The header technique
c. The activity network diagram
d. Matrix diagram
5.12. The project charter will contain a business case, which can be
defined as:
a. A reasoning for the redesign of a process or product

b. The full arguments for the project
c. A short summary of the strategic reason for the project
d. A case study of the project situation
Answers: 5.3. d, 5.6. a, 5.12. c

CSSBB 2014
V. DEFINE
QUESTIONS
5.18. The RACI model is considered:
a. A project management tool

b. A rapid deployment approach
c. An analytical definition treatment
d. A SMART problem methodology
5.22. Focus groups can best be defined as:
a. Small groups with a specific topic

b. A segmented group of suppliers
c. A segmented group of intermediate users
d. A pre-selected group of users
5.25. Lessons learned for a project would typically include which of the
following?
a. How to secure increased funding for similar projects

b. The effectiveness of the entire project
c. The advantages of manual project management methods
d. Elimination of milestone reporting
Answers: 5.18. a, 5.22. a, 5.25. b

CSSBB 2014
V. DEFINE
QUESTIONS
5.27. The key difference between internal and external customers is:
a. Their interest in the product or service

b. Internal customers can influence the design of the product
c. External customers usually influence the design of the product
d. External customers best determine the true quality of the product
5.32. One limitation of Gantt charts would be that:
a. They are easy to understand

b. Changes can be made easily
c. Each bar represents a single activity
d. Estimates of optimistic and pessimistic times can't be included
5.36. An approved written plan of the total costs and cash inflows for a
project is called:
a. A budget
b. A forecast
c. Actual costs and revenues
d. A variance
Answers: 5.27. d, 5.32. d, 5.36. a

CSSBB 2014
V. DEFINE
QUESTIONS
5.42. A project review would concern itself with which of the following?
a. The effectiveness of the entire project

b. The storage media for document archiving
c. What tracking software worked best
d. What project team members fell below expectations
5.44. In the preparation of a project, efforts should be made to identify

and involve various parties affected by the planned changes.
These other parties are known as:
a. Process owners
b. Champions
c. Team leaders
d. Stakeholders
5.45. The SMART approach focuses on:
a. Individual involvement
b. Time boxed results
c. Project management
d. Process management
Answers: 5.42. a, 5.44. d, 5.45. c

© QUALITY COUNCIL OF INDIANA VI-1 (427)
CSSBB 2014
VI. MEASURE - DATA
MAKE A HABIT OF DISCUSSING A

PROBLEM ON THE BASIS OF THE
DATA AND RESPECTING THE
FACTS SHOWN BY THEM.
KAORU ISHIKAWA
CSSBB 2014
VI. MEASURE - DATA V.A.1

PROCESS CHARACTERISTICS/FLOW METRICS
Measure
The Measure element is presented in the following topic
areas:
C Process characteristics
C Data collection
C Measurement systems
Process characteristics are discussed in the following

areas:
C Process flow metrics

C Process analysis tools
CSSBB 2014

Lean Glossary
In order to introduce process flow metrics, the following
definitions are useful.
Andon board - A visual control device in a production

area. It is typically a lit overhead display.
Continuous flow manufacturing (CFM) - A process

whereby material moves one piece at a time, at a rate
determined by the needs of the customer without WIP.
Cycle time - The time required to complete one cycle of

an operation.
Inventory turns - The number of times inventory is

consumed in a given period.
Just-in-time (JIT) - A system for producing the right

items at the right time in the right amounts.
Level loading - The smoothing or balancing of the work

load in all steps of a process.
CSSBB 2014

Lean Glossary (Continued)

Muda - A Japanese term meaning any activity that
consumes resources but creates no value.
Non-value added - Any activity that does not add value

to the product or service.
Perfection - The complete elimination of muda so that all

activities along a value stream create value.
Point of use inventory - Inventory that is delivered

directly to the location where it will be consumed.
Poka-yoke - A mistake-proofing device or procedure to

prevent or detect an error.
Process flow chart - A diagram of the flow or sequence

of events in a process.
Pull - A system in which nothing is produced by the

upstream supplier until the downstream customer
indicates a need.
CSSBB 2014


Queue time - The time a product spends awaiting the
next processing step.
Single minute exchange of dies (SMED) - A series of

techniques for rapid changeovers of production
machinery. The long-term objective is zero setup time.
Single-piece-flow - A situation in which one complete

product proceeds through various operations without
interruptions or scrap.
Skills matrix - A work cell visual control depicting all

work activities. It provides assistance in the cross
training of team members.
Small lot principle - Effectively reducing lot size until the

optimum of one piece flow is realized.
Standard work - A description of each work activity,

specifying cycle time, takt time, the work sequence and
the minimum inventory.
CSSBB 2014


Takt time - The available production time divided by the
rate of customer demand.
Value stream - The specific activities required to design

and provide a product from order to delivery.
Visual control - The placement, in plain view, of all the

tools, parts, production activities, and indicators of
production system performance.
Waste - All overproduction waiting, unnecessary

transport, excessive inventories, unnecessary
movements, and production of defective parts.
Work cell - The layout of machines or business

processes of different types, performing different
operations in a tight sequence, typically a U or L shape.
Work center - One process station in a work cell.

CSSBB 2014

Lean Thinking
Womack offers five guiding principles:
C Specify value by product

C Identify the value stream for each product
C Make value flow
C Let the customer pull value from the producer
C Pursue perfection
CSSBB 2014

Value
Value is defined by the customer. Quite often,
manufacturers cannot provide appropriate customer
value. In many cases, senior business managers in the
United States have lost touch with the customer. They
desire customers to purchase products at reasonable
prices in order to keep the company profitable. They
attempt to increase the performance of the firm, while
reducing costs.
The German mind set is more product feature and

process oriented.
The Japanese define value in the context of where value

is created.
CSSBB 2014

Value (Continued)
In the long run, value must be defined by the customer.
The customer wants specific products, with specific
capabilities, at specific prices. Specifying value is the
first step in lean thinking.
The target cost of the product may be determined after

defining customer value. This target cost is more than
the “market cost” of the product. The market cost is
typically the manufacturing cost of the product plus the
selling expenses and profit. In lean thinking, the target
cost is the mixture of the current selling prices by
competitors and the examination of and the elimination
of muda (waste) through lean methods.
CSSBB 2014

Value Stream (Value Chain)

The benefits of reducing waste can be magnified many
times by concentrating on the set of activities that link
a process together. There are three streams or chains
that should be considered:
C Problem solving
C Information management
C Physical transformation
The problem solving stream for a business includes:

solve the concept design, develop the prototype, plan
reviews, and determine the mechanism for product
launch. The information management stream consists
of customer order taking, raw material sequencing from
suppliers, in-house scheduling, and delivery to the
customer. The physical transformation stream proceeds
from the conversion of raw materials to finished goods.
CSSBB 2014

Value Stream (Continued)

The value chain, as described by Porter, is at a high
operating level. A generic value chain involves the key
components: human resources, technology,
procurement, inbound logistics, operations, outbound
logistics, marketing, sales, and service. The discussion
and analysis centers on methods to differentiate the
firm. An organization gains competitive advantage by
performing certain strategic activities better or at a
lower cost than its competitors.
The value stream (or chain), as used in lean

manufacturing efforts, goes into greater detail. It
involves a single product stream, which is analyzed for
reduction of waste, reduction in cycle time, or
improvement in quality.
CSSBB 2014

Value Stream (Continued)

A value stream map is created to identify all of the
activities involved in the product. This value stream can
include the various suppliers, production activities, and
the final customer. The activities are viewed in terms of
the following criteria:
C It adds value, as perceived by the customer

C It adds no value, but is required by the process
C It adds no value, and can be eliminated
CSSBB 2014

Value Flow
Traditional mass production is often accomplished by
the batch technique. The objective is to produce many
units of a specific part at a given time, in order to
maintain the production efficiency of the machines and
the overall efficiency of the departments. However, this
leads to sub-optimization of the process as practiced by
non-lean companies.
The lean effort requires the conversion of a batch

process to a continuous flow process. In some cases,
converting the batch process to a one piece flow is
ideal. Some of the obstacles to overcome include:
C The plant has always done it in batches or lots

C The plant has a multitude of functions
C The plant can’t afford quick changeover tooling
C The plant has high momentum, inflexible machinery
C The plant machinery would be expensive to move
CSSBB 2014

Value Flow (Continued)

Ideally, in a continuous flow layout, the production steps
for single piece flow, without WIP, are arranged in a
straight line, U-shape or cellular sequence. Inside this
flow, the work of each station and operator must be
performed with reliability.
When the machinery is performing as expected, there

are zero breakdowns. This is the concept of TPM (total
productive maintenance). The quality level of each
operation is very high, near perfect, using a variety of
defect elimination and detection techniques.
The activities needed for production should be in a

steady, continuous flow, with no wasted motions, no
batches, and no WIP.
CSSBB 2014

Pull Value
Instead of creating product in response to an estimated
sales forecast, the plant manufactures product as the
customer requires it. This is the pull system in action.
This results in many positive things for the organization:
C Cycle times decrease

C Finished inventories are reduced
C Work-in-process (WIP) is reduced
C The customer stabilizes their ordering
C Pricing is stabilized
Most mass production manufacturing firms are in the

push production mode. Each operation produces as
much as possible and sends it onto the next operation.
The goal is to maximize machine efficiency with a
maximum amount of in-process inventory.
Contrast the above manufacturing firm with the factory

that is dependent on the pull of the market. The receipt
of a customer order initiates activities. Each operation
produces parts as needed through a signal from
downstream. There is a minimal amount of WIP.
CSSBB 2014

Perfection
The pursuit of the first four principles of lean thinking
allows the firm to move toward perfection. Solving
customer value problems, working the value stream,
converting to flow, and making pull occur, all help
eliminate muda (waste). As the process continues, more
muda is eliminated, and perfection seems possible.
Perfection is accomplished via:
C Teams working with customers to find ways to

specify value, enhance flow, and achieve pull
C Using collaboration between the value stream

partners to uncover more value
C Using technologies to eliminate muda
C Developing new products
Lean thinking principles are the cornerstones to higher

performance and economic growth. Perfection is a
journey.
CSSBB 2014

Perfection (Continued)
Some possible improvement results from lean thinking
are:
Improvement Area Reduction or

Improvement
Labor productivity 100% increase
Throughput times 90% reduction
Inventories 90% reduction
In-house scrap 50% reduction
Safety injuries 50% reduction
Development time 50% reduction
Capital investment Modest
It may take years to apply lean thinking principles in a

company, and even more time to apply lean thinking in
the entire value chain.
CSSBB 2014

Takt Time
Takt time is the available production time divided by the
rate of customer demand. In order to understand the
measurement of takt time, and how it can be improved,
a discussion of a hypothetical process is presented in
the Primer.
CSSBB 2014

Takt Time (Continued)

It should be emphasized that the use of lean
manufacturing principles should not result in the direct
layoff of employees. It is best to reassign the
employees to new responsibilities. Normal employee
attrition is another option. Expansion of the business
into new areas is also a great solution.
The takt time is defined as the available production time

divided by the rate of customer demand.
Net operating time per period

Takt time =
Customer requirements per period
CSSBB 2014


An example of the calculation of takt time (using Primer
data) is shown below:
430 minutes per day

Takt time =
300 units per day
Takt time = 1.433 minutes per unit
The takt time will be 1.433 minutes or about 86 seconds

per unit. The ideal pace of each operation is set at 86
seconds.
If a plant does not have the luxury of mass production to

establish a single takt time, the solution may be to
develop multiple takt times, breaking the requirements
into smaller components.
CSSBB 2014


Illustration of Batch vs. One-piece Flow
The following three cases assume a series of three

operations, with each processing one unit per minute.
Operation 1 Operation 2 Operation 3
Case 1: Orders are manufactured in batches of 100

units.
C Only batches of 100 units are transferred from

operation to operation
C The total processing time is 201 minutes before the

first unit is available
C Total order time through the entire process is 300

minutes for 100 units
CSSBB 2014


Illustration of Batch vs. One-piece Flow
Case 2: Orders are manufactured in batches of 10 units.
C Only batches of 10 units are transferred from

operation to operation


minutes for 10 units
Case 3: Orders are manufactured in batches of 1 units.
C Each unit is transferred from operation to operation


minutes for 1 unit
CSSBB 2014


The three cases illustrate the power of one-piece flow.
If a customer changes their requirements, the shop will
not have 300 units in queue in partial stages of
production. The shop will be able to shift production
requirements and provide the first units rapidly.
The necessary quality, machine, personnel, materials,

and supplier resources must be coordinated and made
available as needed. The layout of the line or cell is a
starting point. The line should be examined as
necessary to:
C Improve cycle times

C Reduce product defects
C Correct long changeover times
C Address equipment reliability issues
CSSBB 2014

Lean Metrics
TPM metrics include availability, operating speed rate
(OSR), net operating rate (NOR), performance efficiency
(PE), and overall equipment effectiveness (OEE).
Some additional lean measurements that can prove

beneficial:
1
Throughput Rate =
Cycle Time
Little’s law states:
WIP Inventory = Throughput x Flow Time

CSSBB 2014

Lean Metrics (Continued)

Since the throughput rate is equal to 1/(cycle time),
Little’s law can be written as:
Flow Time = WIP x Cycle Time
George further customizes Little’s Law as:
Number Items in Process

Total Lead Time =
Average Completion Rate
George also defines process cycle efficiency as:
Value Added Time

PCE =
Total Lead Time
CSSBB 2014


If there are 30,000 units scheduled for production by
ACME Stamping and 3,000 units can be produced per
day, what is the total required lead time?
TLT = WIP/Completion Rate
TLT = 30,000 units/(3,000 units/day) = 10 days
Quality Shipping has a backlog of 21 outstanding

quotes. They provide their customers 3 days maximum
quotation service. What must their completion rate be
to meet their internal goal?
WIP 21 quotes
Completion Rate = = = 7 quotes/day
TLT 3 days
CSSBB 2014


Better Brass Corporation requires 8 weeks to fulfill a
typical bearing order. The value added time is 18 hours.
Assuming the company works 24 hours/day, seven day
per week, what is the process cycle efficiency expressed
as a percentage?
VAT
PCE = x 100%
TLT
18 hours
PCE = x 100% = 1.34%
1,344 hours
If the throughput rate for an operation is 7,200 units per

hour, what is the cycle time in seconds?
1
Cycle Time =
Throughput Rate
1
Cycle Time = x 3,600 sec/hr = 0.5 sec
7,200/hr
Note that 1 hour equals 3,600 seconds.

CSSBB 2014

PROCESS CHARACTERISTICS/ANALYSIS TOOLS
Process Analysis Tools

Flow charts, process maps, written procedures, and
work instructions are tools used for process analysis
and documentation. Other lean techniques such as
value stream mapping and spaghetti diagrams are also
often used.
Value Stream Mapping

A value stream map is created to identify all of the
activities involved in product manufacturing from start
to finish. This value stream may include suppliers,
production operations and the end customer. VSMs
typically focus on material and information flow. For
product development, value stream mapping includes
the design flow from product concept to launch. This is
the large view, looking at the entire system for
improvement opportunities.
CSSBB 2014

Value Stream Mapping (Continued)

Benefits of a value stream map include:
C Seeing the complete process flow

C Identifying sources and locations of waste
C Providing common terminology for discussions
C Helping to make decisions about the flow
C Tying multiple lean techniques together
C Providing a blueprint for lean ideas
C Showing the motion and material flow linkage
C Describing how the process can change
C Determining effects on various metrics
CSSBB 2014

Value Stream Mapping (Continued)

The value stream mapping process is shown below.
Define product family Use a product equipment matrix
Draw current state map Do this personally
Create future state map Use creative concepts
Implementation planning Can take months or years

CSSBB 2014

VSM (Continued)
Define Product Family
The recommended value stream approach is to map one

product family. A product family is defined as a group
of products that pass through similar processing steps
and over common equipment. A product and equipment
matrix can be used to indicate common features. An
example matrix is shown in the Primer.
CSSBB 2014

VSM (Continued)
Current State Map
A current state map of the process is developed to

facilitate a process analysis. Basic tips on drawing a
current state map include:
C Start with a quick orientation of process routes

C Personally follow the material and information flows
C Map the process with a backward flow
C Collect the data personally
C Map the whole stream
C Create a pencil drawing of the value stream
Some of the typical process data included are: cycle

time (CT), changeover time (COT), uptime (UT), number
of operators, pack size, working time (minus breaks, in
seconds), WIP, and scrap rate. An analysis of the
current status can provide the amount of lead and value-
added time.
CSSBB 2014

VSM (Continued)
Value stream mapping definitions worth noting include:
Value-added time (VAT) - The amount of time spent

transforming the product, which the customer is willing
to pay for.
Lead time (L/T) - The time it takes one piece of product

to move through all of the processes.
Cycle time (C/T) - The time it takes a piece to complete

an individual process.
Reliability (REL) - Reliability of the process step. Some

current state maps show this as uptime %.
CSSBB 2014

VSM (Continued)
Future State Map
A future value stream map, is an attempt to make the

process lean. This involves creativity and teamwork by
the value stream manager and the lean team to identify
creative solutions. Questions to ask when developing
a future state map are:
C What is the required takt time?

C Do manufactured items move directly to shipping?
C Are items sent to a finished goods supermarket?
C Is continuous flow processing applicable?
C Where is the pacemaker process?
C Can the process be leveled?
C What is the increment of work to be released?
C What process improvements can be used?
CSSBB 2014

VSM (Continued)
Implementation Planning
The final step in the value stream mapping process is to

develop an implementation plan for establishing the
future state. This includes a step-by-step plan,
measurable goals, and checkpoints to measure
progress. The plan could take months or years to
complete, and even then, there may be a need to
improve upon it in the future.
Examples of VSM icons and hypothetical current and

future state maps are shown in the Primer.
CSSBB 2014

Value Stream Mapping Icons
Finished Goods
Electronic Flow FIFO Movement Go See
Inventory Kaizen Burst Kanban Batches Kanban Post
Kanban Production Kanban Signal Kanban Load Leveling

Withdrawal
Manual Information
Flow Operator Process Box Pull Arrow
Pull Circle Push Arrow Schedule Box Source
Supermarket Truck Shipment Buffer Stock Data Box

0.5 days
2.7 min.
Time Line Lead Time Cycle Time

CSSBB 2014

Flow Charts/Process Mapping

A flow chart, or process map, is useful to people familiar
with a process, as well as those that have a need to
understand a process. Flow charts can be used to
identify improvement opportunities as illustrated by the
following sequence:
C Organize a team to examine a process

C Construct a flow chart to represent each step
C Discuss and analyze each step in detail
C Ask the key question, “Why do we do it this way?”
C Compare the actual process to a “perfect” process
C Is there unnecessary complexity?
C Does duplication or redundancy exist?
C Are there control points to prevent errors?
C Is this process being run the way it should?
C Ideas may come from different processes
CSSBB 2014

Flow Charts (Continued)

Listed below are examples of flow chart applications.
Purch. Processing purchase orders, placing

actual purchases, vendor contract
negotiations
Mfg. Processing returned goods, handling

internal rejections, production
processes, training new operators
Sales Making a sales call, taking order

information, advertising sequences
Admin. Correspondence flow, processing

times, correcting mistakes, handling
mail, typing letters, hiring employees
Maint. Work order processing,

p.m. scheduling
Lab. Delivery of samples, testing steps,

selection of new equipment, personnel
qualification sequence, management of
work flow
CSSBB 2014

Flow Charts (Continued)

There are advantages to depicting a process in a
schematic format. The major advantage is the ability to
visualize the process being described.
Process mapping or flow charting has the benefit of

describing a process with symbols, arrows, and words
without the clutter of sentences. Some common
mapping symbols are shown below:
Process Alternate
Process Decision Data
Predefined Process Document Terminator Decision or

Preparation
Manual
Manual Operation Display
Input Off Page
Connector Connector
Merge Magnetic
Delay Disk or
Storage
Extract
CSSBB 2014

Flow Chart Example

Start
Material
received
Visual
inspection
Inform purchasing
No Of rejection.
Visual Yes
Generate
Defects?
Corrective
action report
Return to
Dimensional Yes supplier
Dimensional
Inspection
inspection
Required? End
No
No
Acceptable?
Yes
Place in
inventory
End
CSSBB 2014

Procedures
A procedure is a document that specifies the way to
perform an activity. For most operations, a procedure
can be created in advance by the appropriate
individual(s). Consider the situation where a process
exists, but has not been documented. The procedure
should be developed by those having responsibility for
the process of interest.
An example is described in the Primer.

CSSBB 2014

Nonconforming Material Flow Chart

Production or Inspection
discovers NC product
Product is held In special

holding area
Inspection creates a
Deviation Report (DR)
to document condition
Quality supervisor
reviews product
Deviation
Report noted
with action
MRB
Scrap
product
Rework product
product
Identify as
scrap
MRB makes
disposition
No Rework
successful
Use as is Repair Scrap Send to scrap
product product disposal area
Yes
Return to production
for repair
Return product
Follow scrap
to flow
instructions
Repair
Close DR Yes successful No
Clear DR and File
File in DR
Quality Records
CSSBB 2014

Work Instructions
Procedures describe the process at a general level,
while work instructions provide details and a step-by-
step sequence of activities. Flow charts may also be
used with work instructions to show relationships of
process steps. Controlled copies of work instructions
are kept in the area where the activities are performed.
Some discretion is required in writing work instructions,

so that the level of detail included is appropriate for the
background experience and skills of the personnel that
would typically be using them. Similar to writing
procedures, the people that perform the activities
described in the work instruction should be involved in
writing the work instruction.
CSSBB 2014

Spaghetti Diagrams
Spaghetti diagrams can be useful in describing the flow
of people, information, or material in almost any type of
process. They are not as diagnostic or definitive as
VSM in traditional manufacturing operations, but
certainly have a utility for a number of service,
administrative, and light production situations.
Most applications consider people, information, or

material flows.
CSSBB 2014

Spaghetti Diagram Example
BINDER STORAGE
ACTIVE
DOORS
SHRINK WRAP PRINTER A
6
PACKING STATION
4
STAIRWELL
TAB STORAGE
BUFFER
WORK TABLE
A STORAGE
UNDERNEATH
SERVER
DOOR
2 B
UPS SCALE
STORAGE
MANUAL
3
1
PRINTER B
SHIPPING
DOOR SHIPPING
CART BOOKS
4 PAPER
5 STORAGE
BOOKS
MEDIA
CENTER
S.TEXT INACTIVE
BINDING DOOR
STORAGE
ACTIVE ACTIVE
DOOR DOOR
An example of information flow.

CSSBB 2014

Circle Diagrams
There are a large number of circle diagram applications.
See the examples below:
Pie Chart Dial Chart

50
40 60
B
30 70
C
D 80
20
E
10 90
F
0% 100%
Categories % Project Completion
CSSBB 2014

Circle Diagrams (Continued)
4. ACT 1. PLAN
3. CHECK 2. DO
The PDCA Cycle
Precise, but Unbiased, but Unbiased and

Biased not Precise Precise
Bias and Precision Distinction

CSSBB 2014

Circle Diagrams
On occasion, a circle diagram can help conceptualize
the relationship between work elements in order to
optimize work activities. Shown below is a hypothetical
analysis of the work load for a shipping employee using
a Venn (or circle) diagram.
Packing Shipping
0.30 Data
Entry
0.20
0.04
Pulling 0.06
Stock Making
0.25 CDs
0.10
CSSBB 2014

Circle Diagrams (Continued)

In this case, there is an overlap between packing and
data entry, as well as packing and pulling stock. Making
CDs is exclusive of other activities. If the sample space
equals 1.0 or 100%, then one can determine both the
busy time and idle time in an 8-hour shift.
Busy time = Packing + Data entry + Pulling stock +

Making CDs - Overlap
= 0.30 + 0.20 + 0.25 + 0.10 - 0.06 - 0.04
= 0.75
In an 8-hour shift, there are 6.0 hours of activity. By the

same logic, there are 2.0 idle hours. After deducting
customary break times, one can consider whether
additional duties can be assumed by this individual.
Venn diagrams are normally used to explain probability

theory. In the previous diagram, making CDs and
packing are mutually exclusive, but packing and pulling
stock are not.
CSSBB 2014

Gemba Walk
Gemba refers to the work area. In manufacturing, this is
the factory floor. In the lean manufacturing
environment, gemba walk takes members of
management to the floor level looking for signs of waste
and opportunities for improvement.
The term gemba walk was championed by Taiichi Ohno

while at Toyota.
Gemba walk also gives managers an opportunity to

communicate directly with employees and learn their
problems. The managers benefit from seeing the real
shopfloor practices. In addition to waste and efficiency,
housekeeping and potential safety issues are noted.
Any shopfloor improvement is the underlying objective.
A gemba walk should display mutual respect and true

interest in making things better, easier, and safer.
CSSBB 2014
VI. MEASURE - DATA V.B.1

DATA COLLECTION/TYPES OF DATA
Data Collection
Data collection is described in the following topic

areas:
C Types of data
C Measurement scales
C Sampling methods
C Collection methods
Types of Data
Data is objective information that everyone can agree
on. Measurability is important in collecting data. The
three types of data are attribute data, variables data, and
locational data. Of these three, attribute and variables
data are more widely used.
CSSBB 2014

Types of Data (Continued)

Attribute Data
Attribute data is discrete. This means that the data

values can only be integers, for example, 3, 48, 1029.
Counted data or attribute data are answers to questions
like “how many”, “how often”, or “what kind.”
Examples include:
C How many of the final products are defective?

C How many people are absent each day?
C How many days did it rain last month?
C What kind of performance was achieved?
Variables Data
Variables data is continuous. This means that the data

values can be any real number, for example, 1.037, -4.69,
84.35. Measured data (variables data) are answers to
questions like “how long,” “what volume,” “how much
time” and “how far.” This data is generally measured
with some instrument or device. Examples include:
C How long is each item?

C How long did it take to complete the task?
C What is the weight of the product?
CSSBB 2014

Types of Data (Continued)

Variables Data (Continued)
Measured data is regarded as being better than counted

data. It is more precise and contains more information.
For example, one would certainly know much more
about the climate of an area if they knew how much it
rained each day rather than how many days it rained.
Collecting measured data is often difficult or expensive,

so counted data must be used. In some situations, data
will only occur as counted data.
For information which can be obtained as either

attribute or variables data, it is generally preferable to
collect variables data.
CSSBB 2014

Data Comparison
Variables Attribute
Characteristics measurable countable
continuous discrete units or
may derive occurrences
from counting good/bad
Types of Data length no. of defects
volume no. of defectives
time no. of scrap items
Examples width of door audit points lost
lug nut torque paint chips per unit
fan belt defective lamps
tension
Data Examples 1.7 inches 10 scratches
32.06 psi 6 rejected parts
10.542 25 paint runs
seconds
Locational Data
The third type of data does not fit into either category
above. This data is known as locational data which
simply answers the question “where.” Charts that utilize
locational data are often called “measles charts” or
concentration charts.
CSSBB 2014

Data Conversion
Some data may only have discrete values, such as this
part is good or bad, or I like or dislike the quality of this
product. Since variables data provides more
information than does attribute data, for a given sample
size, it is desirable to use variables data whenever
possible.
When collecting data, there are opportunities for some

types of data to be either attributes or variables. Instead
of a good or bad part, the data can be stated as to how
far out of tolerance or within tolerance it is. The like or
dislike of product quality can be converted to a scale of
how much do I like or dislike it.
Consideration of the cost of collecting variables versus

attributes data should also be given when choosing the
method. Typically, the measuring instruments are more
costly for performing variables measurements and the
cost to organize, analyze and store variables data is
higher as well.
CSSBB 2014

Data Conversion (Continued)

Variables data requires storing of individual values and
computations for the mean, standard deviation, and
other estimates of the population. Attributes data
requires minimal counts of each category and hence
requires very little data storage space.
For manual data collection, the required skill level of the

technician is higher for variables data than for attribute
data. Likewise, the cost of automated equipment for
variables data is higher than for attributes data.
The ultimate purpose for the data collection and the type
of data are the most significant factors in the decision to
collect attribute or variables data.
CSSBB 2014

DATA COLLECTION/MEASUREMENT SCALES
Measurement Scales
Shown below are four measurement scales in increasing
order of statistical desirability.
Scale Description Example
Nominal Data consists of names or A bag of candy contained the
categories only. No ordering following colors:
scheme is possible. Yellow 15
Red 10
Orange 9
Green 7
Ordinal Data is arranged in some order Product defects, where A type
(Ranking) but differences between values defects are more critical than D
cannot be determined or are type defects are tabulated as
meaningless. follows:
A 16
B 32
C 42
D 30
Interval Data is arranged in order and The temperatures of three
differences can be found. ingots were 200EF, 400EF and
However, there is no inherent 600EF. Note, that three times
starting point and ratios are 200EF is not the same as 600EF
meaningless. as a temperature measurement.
Ratio An extension of the interval Product A costs $300 and
level that includes an inherent product B costs $600. Note,
zero starting point. Both that $600 is twice as much as
differences and ratios are $300.
meaningful.
CSSBB 2014

Measurement Scales (Continued)

A further description of the key characteristics of the
four measurement scale is provided in below.
Scale Central Dispersion Significance

Location Tests
Nominal Mode Information Chi square
Only
Ordinal Median Percentages Sign or Run
Test
Interval Arithmetic Standard or t test
Mean Average F test
Deviation Correlation
Analysis
Ratio Geometric Percent *
or Harmonic Variation
Mean
*Note: Many of the interval measures may be useful for

ratio data as well
CSSBB 2014

Data Examples
Examples of continuous data, discrete data, and

measurement scales:
1. Continuous data: A station wagon weighs 3478.6 lbs
2. Discrete data: On the last CSSBB exam, 400 people

failed
3. Ordinal scale: Defects are categorized as critical,

major A, major B, and minor
4. Nominal scale: A print-out of all shipping codes for

last week’s orders
5. Ratio scale: The individual weights of a sample of

widgets
6. Interval scale: The temperatures of steel rods (EF)

after one hour of cooling
CSSBB 2014

DATA COLLECTION/SAMPLING METHODS
Random Sampling
Random sampling requires giving every part an equal
chance of being selected for the sample.
The sample must be representative of the lot and not

just the product that is easy to obtain. Thus, the
selection of samples requires some up front thought
and planning.
Sampling without randomness ruins the effectiveness of

any plan. The product to be sampled may take many
forms: in a layer, on a conveyor, in sequential order, etc.
The sampling sequence must be based on an
independent random plan. The sample is determined by
selecting an appropriate number from a hat or random
number table.
CSSBB 2014

Sequential Sampling
Sequential sampling plans are similar to multiple
sampling plans except that sequential sampling can
theoretically continue indefinitely. Usually, these plans
are ended after the number inspected has exceeded
three times the sample size of a corresponding single
sampling plan. Sequential testing is used for costly or
destructive testing with sample sizes of one and are
based on a probability ratio test developed by Wald.
CSSBB 2014

Stratified Sampling
One of the basic assumptions made in sampling is that
the sample is randomly selected from a homogeneous
lot. When sampling, the “lot” may not be homogeneous.
The concept behind stratified sampling is to attempt to

select random samples from each group or process that
is different from other similar groups or processes. The
resulting mix of samples thus drawn can be biased if the
proportion of the samples does not reflect the relative
frequency of the groups.
To the person using the sample data, the implication is

that they must first be aware of the possibility of
stratified groups and second, phrase the data report
such that the observations are relevant only to the
sample drawn and may not necessarily reflect the
overall system.
CSSBB 2014

Data Accuracy and Integrity

Bad data corrupts the decision-making process. Some
considerations include:
C Avoid emotional bias relative to tolerances when

counting, measuring, or recording data.
C Avoid unnecessary rounding. Rounding often

reduces measurement sensitivity.
C If data occurs in time sequence, record the order of

its capture.
C If an item characteristic changes over time, record

the measurement after its manufacture, as well as
after a stabilization period.
CSSBB 2014

Data Accuracy and Integrity (Cont’d)

C To apply statistics which assume a normal
population, determine whether the expected
dispersion of data can be represented by at least 8
to 10 resolution increments. If not, the default
statistic may be the count of observations which do
or do not meet specification criteria.
C Screen or filter data to detect and remove data entry

errors such transposition or misplaced decimal
points.
C Avoid removal by hunch. Use objective statistical

tests to identify outliers.
C Each important classification identification should

be recorded along with the data.
It is important to select a sampling plan appropriate for

the purpose of the use of the data.
CSSBB 2014

DATA COLLECTION/COLLECTION METHODS
Data Collection Methods

To ensure that the collected data is relevant to the
problem, some prior thought must be given to what is
expected. Check sheets, tally sheets, and checklists are
data collection methods which are widely used. Other
data collection methods include automatic measurement
and data coding.
Some data collection guidelines are:
C Formulate a clear statement of the problem

C Define precisely what is to be measured
C List all the important characteristics to be measured
C Carefully select the right measurement technique
C Construct an uncomplicated data form
C Decide who will collect the data
C Arrange for an appropriate sampling method
C Decide who will analyze and interpret the results
C Decide who will report the results
Without an operational definition, most data is

meaningless. Both attribute and variable specifications
must defined.
CSSBB 2014

Data Collection Methods (Continued)

Data collection includes both manual and automatic
methods. Data collected manually may be done using
printed forms or by data entry at the time the
measurements are taken. Manual systems are labor
intensive and subject to human errors in measuring and
recording the correct values.
Automatic data collection includes electronic chart

recorders and digital storage. Automatic systems have
higher initial costs than manual systems, and have the
disadvantage of collecting both “good” and “erroneous”
data. Advantages to using automatic data collection
systems include high accuracy rates and the ability to
operate unattended.
CSSBB 2014

Automatic Measurement
Automatic sorting gages are widely used to sort parts by
dimension.
Computer controlled measurement systems may offer

distinct advantages over their human counterparts.
(Improved test quality, shorter inspection times, lower
operating costs, automatic report generation, improved
accuracy, and automatic calibration.) Automated
measurement systems have the capacity and speed to
be used in high volume operations.
Automated systems have the disadvantages of higher

initial costs, and a lack of mobility and flexibility
compared to humans. Automated systems may require
technical malfunction diagnostics.
CSSBB 2014

Automatic Measurement (Continued)

Applications for automatic measurement and digital
vision systems are quite extensive. The following
incomplete list is intended to show examples:
C Error proofing a process

C Avoiding human boredom and errors
C Sorting acceptable from defective parts
C Detecting flaws, surface defects, or foreign material
C Creating CAD drawings from an object
C Building prototypes by duplicating a model
C Making dimensional measurements
C Performing high speed inspections
C Machining, using laser or mechanical methods
C Marking and identifying parts
C Inspecting solder joints on circuit boards
C Verifying and inspecting packaging
C Providing optical and bar code recognition
C Identifying missing components
C Controlling motion
C Assembling components
C Verifying color
CSSBB 2014

Data Coding
The efficiency of data entry and analysis is frequently
improved by data coding. Problems due to not coding
include:
C Inspectors trying to squeeze too many digits into

small blocks on a form
C Reduced throughput and increased errors by

reading and entering large sequences of digits
C Insensitivity of analytic results due to rounding

CSSBB 2014

Data Coding (Continued)

Coding by adding or subtracting a constant or by
multiplying or dividing by a factor:
Let the subscript, lowercase c, represent a coded

statistic; the absence of a subscript represents raw
data; uppercase C indicates a constant; and lowercase
f represents a factor. Then:
Code: X C = X + C Decode: X = X C - C, σ = σC
Code: X C = X - C Decode: X = X C + C, σ = σC
Code: X C = fX Decode: X = X C /f, σ = σC /f
Code: X C = X/f Decode: X = fX C , σ = fσC
CSSBB 2014

Data Coding (Continued)

Coding by substitution:
Consider a dimensional inspection procedure in which

the specification is nominal plus and minus 1.25". The
measurement resolution is 1/8 of an inch and
inspectors, using a ruler, record plus and minus
deviations from nominal. A typical recorded
observation might be 32-3/8" crammed in a space that
was designed to accommodate three characters. The
data can be coded as integers expressing the number of
1/8 inch increments deviating from nominal.
Coding by truncation of repetitive place values:
Measurements such as 0.55303, 0.55310, 0.55308, in

which the digits 0.553 repeat in all observations, can be
recorded as the last two digits expressed as integers.
Depending on the objectives of the analysis, it may or
may not be necessary to decode the measurements.
CSSBB 2014

Check Sheets
Check sheets are great tools for organizing and
collecting facts and data. By collecting data, individuals
or teams can make better decisions, solve problems
faster, and earn management support.
Recording Check Sheets

A recording check sheet is used to collect measured or
counted data. The simplest form of the recording check
sheet is for counted data. Data is collected by making
tick marks on this particular style of check sheet.
The check sheet can be broken down to indicate either

shift, day, or month. Measured data may be summarized
by the means of a check sheet called a tally sheet. To
collect measured data, the same general check sheet
form is used. The only precaution is to leave enough
room to write in individual measurements.
CSSBB 2014

Recording Check Sheets (Continued)

Illustrated below is an example of a process check sheet
used to evaluate the productivity of a meeting. This is
subjective data, but is very useful in some
circumstances. Measured data is physically measured
information, such as: the pH, the air pressure in psi, or
the amount of downtime in hours.
Member Member Member Member Member

#1 #2 #3 #4 #5
On Track
Participation
Listening
Leadership
Decision
Quality
Scale: 1 = poor, 10 = excellent.

CSSBB 2014

Checklists
The second major type of check sheet is called the
checklist. A grocery list is a common example of a
checklist. On the job, checklists may often be used for
inspecting machinery or products. Checklists are also
very helpful when learning how to operate complex or
delicate equipment. An example is shown in the Primer.
Measles Charts
Not illustrated is a locational variety of check sheet
called a measles chart. This check sheet could be used
to show defect or injury locations using a schematic of
the product or a human.
CSSBB 2014
VI. MEASURE - DATA V.C.1

MEASUREMENT SYSTEMS/SYSTEM ANALYSIS
Measurement Systems
Measurement Systems is described in the following

topic areas:
C Measurement system analysis

C Enterprise measurement systems
C Metrology
CSSBB 2014

Measurement System Analysis

There are four characteristics to examine in a gage
system.
Sensitivity
The gage should be sensitive enough to detect
differences in measurement as slight as one-tenth of
the total tolerance specification or process spread,
whichever is smaller. Inadequate discrimination will
affect both accuracy and precision.
Reproducibility
Reproducibility is the “reliability” of the gage system
gage systems to reproduce measurements. The
reproducibility of a single gage is customarily
checked by comparing the results of different
operators taken at different times. Gage
reproducibility affects both accuracy and precision.
Accuracy
Accuracy is an unbiased true value and is normally
reported as the difference between the average of a
number of measurements and the true value.
CSSBB 2014

Measurement System Analysis (Cont’d)

Precision
In gage terminology, “repeatability” is often
substituted for precision. Repeatability is the ability
to repeat the same measurement by the same
operator at or near the same time.
Precise, but Accurate, but Accurate and

not accurate not precise precise
The calibration of measuring instruments is necessary

to maintain accuracy, but does not necessarily increase
precision.
CSSBB 2014

Repeatability and Reproducibility

There are three widely used methods to quantify
measurement error: the range method, the average and
range method, and the ANOVA method. A brief
description of each follows:
Range Method
The range method is a simple way to quantify the

combined repeatability and reproducibility of a
measurement system.
Average and Range Method
The average and range method computes the total

measurement system variability, and allows the total
measurement system variability to be separated into
repeatability, reproducibility, and part variation.
Analysis of Variance Method
ANOVA is the most accurate method for quantifying

repeatability and reproducibility and allows the
variability of the interaction between the appraisers and
the parts to be determined.
CSSBB 2014

Analysis of Variance Method (Continued)

The example in the Primer is for five parts, three
technicians, and two replications.
ANOVA TABLE α = 0.05

Source SS DF MS Fcal F(α) Var Adj Var %
Technician 0.6167 2 0.3083 1.28 3.68 0.0067 0.0067 1.08
Part No. 9.867 4 2.467 10.21 3.06 0.3708 0.3708 59.83
Interaction 1.633 8 0.2041 0.84 2.64 -0.0188 0 0
Error 3.625 15 0.2417 0.2417 39.03
Total DF 29 SIGe = 0.4916 Totals 0.4753 100
SIGtot = 0.7869
For this example, repeatability is the error variance and

contributes 39.03% of the total variation in the data.
Reproducibility is the variation among technicians

which contributes 1.08% of the variation in the data.
Process variation accounts for 59.83% of the total

variation in the data.
Hypothesis tests based on the F distribution are used to

determine if there are differences between technicians
or between processes.
CSSBB 2014

Measurement Correlation
Measurement correlation can have several meanings.
The most basic meaning is the correlation or
comparison of the measurement values from one
measurement system with the corresponding values
reported by different measurement systems.
A measurement system or device can be used to

compare values against a known standard. Both the
measurement system and the standard have variation.
The measurement system or device may also be

compared against the mean and standard deviation of
multiple other similar devices, all reporting
measurements of the same or similar artifacts.
Measurement correlation can also mean comparison of

values obtained using different measurement methods
used to measure different properties.
The Measurement Systems Analysis Reference Manual

classifies measurement system error into five
categories: bias, repeatability, reproducibility, stability,
and linearity. Each of these is a component whose
combined effect explains measurement correlation.
CSSBB 2014

Bias
Bias is the difference between the output of the
measurement system and the true value. It is often
referred to as accuracy. Bias is computed from the
expression:
n
x i
Bias = i=1
-
n
Where n is the number of times the standard is

measured, Xi is the i th measurement, and  is the value
of the standard.
Bias is usually reported as a percent of process

variation or as a percent of tolerance. If a gage
underestimates, the bias is negative. If a gage
overestimates, the bias is positive.
CSSBB 2014

Linearity
Measurement system linearity is found by obtaining
reference part measurement values throughout the
operating range of the instrument, and plotting the bias
against the reference values. The default procedure for
determining linearity is to measure 10 parts 5 times
each. The percent linearity is equal to the slope, b, of
the best-fit straight line through the data points, and the
linearity is equal to the slope multiplied by the process
variation (process spread or the tolerance):
L = bVp
The bias at any point can be estimated from the slope

and the y-intercept, (y0) of the best-fit line:
B = y0 + b x
CSSBB 2014

Percent Agreement
Percent agreement between the measurement system
and either reference values or the true value of a
variable being measured, can be estimated using a
correlation coefficient, “r”. If r = ±1.0, then there is 100%
agreement and if r = 0, then there is 0% agreement
between the measurement system variables and the
reference or true values.
CSSBB 2014

Precision/Tolerance (P/T)
The precision/tolerance ratio (P/T) is the ratio between
the estimated measurement error (precision) and the
tolerance of the characteristic being measured. The
precision/tolerance ratio is thus:
6σE
P/T Ratio =
Tolerance
Where 6σE is the standard deviation due to

measurement variability. The assumptions are:
C Measurement errors are independent

C Measurement errors are normally distributed
C Measurement error is independent
It is desirable for the P/T ratio to be small to minimize

the effect of the measurement error. AIAG defines
measurement error as a statistical term meaning the net
effect of all sources of measurement variability that
causes a deviation from the master value.
CSSBB 2014

Precision/Total Variation (P/TV)

The precision/total variation (P/TV) is given by:
6σE
P/TV Ratio =
Total variation
measurement variation
P/TV Ratio =
product variation+measurement variation
It is desirable to minimize the P/TV ratio to reduce the

effect of measurement variation on assessments of
process variation.
CSSBB 2014

ANOVA
The Measurement Systems Analysis (MSA) Reference
Manual provides a mathematical model for the variable
study using ANOVA. The observed value is given by:
Observed Value = Part Mean + Bias + Part Effect +

Appraiser Effect + Replication Error
In equation format: Yijm = xi + εijm
Where yijm is the mth measurement taken by appraiser j

on the ith part.
Assuming that the xi’s are independent and normally

distributed with mean μ and variance σ2, the total
variance is given by:
VAR(yijm) = σ2 + ω2 + α2 + τ2
Where ω2, α2 and τ2 are the variances due to the part

effect, the appraiser effect, and the replication error,
respectively.
CSSBB 2014

Control Chart Methods

The Measurement Systems Analysis (MSA) Reference
Manual provides a mathematical model for the variables
study using average and range. This method requires
either two or three replications, r, by two or three
appraisers, k, on 10 parts, n. The average of the ranges
is given by:
n k Rij
R=  nk
i=1 j=1
The average range value is proportional to the standard

deviation, and using a constant of proportionality, can
be used to estimate the process standard deviation.
CSSBB 2014

MEASUREMENT SYSTEMS/ENTERPRISE MEASUREMENT
Enterprise Measurement
Often overlooked are key enterprise measures that can
be service-oriented and/or transactional in nature.
These measures are often expressed in percentages or
presented to management in time line or graphical
formats.
Enterprise performance can be measured and presented

by using:
C Automatic counters
C Computer generated reports
C Internal and external audits
C Supplier assessments
C Management reports
C Internal and external surveys
C A variety of feedback reports
The following is a non-exclusive list of items that can be

measured:
CSSBB 2014

Enterprise Measurement (Continued)

Suppliers
C Number of product deviations

C Percentage of on-time deliveries
C Percentage of early deliveries
C Shipment costs per unit
C Shipment costs per time interval
C Percentage of compliance to specifications
C Current unit cost compared to historical unit cost
C Dollars rejected versus dollars purchased
C Timeliness of supplier technical assistance
Marketing/Sales
C Sales growth per time period

C Percentage of market compared to the competition
C Dollar amount of sales/month
C Dollar amount of an average transaction
C Time spent by an average customer on website
C Effectiveness of sales events
C Sales dollars per marketing dollar
CSSBB 2014


External Customer Satisfaction
C A weighted comparison with competitors

C Perceived value as measured by the customer
C Ranking of product/service satisfaction
C Evaluation of technical competency
C Percentage of retained customers
Internal Customer Satisfaction
C Employee rating of company satisfaction

C Rating of job satisfaction
C An indication of training effectiveness
C An evaluation of advancement fairness
C Feedback reaction to major policies and procedures
C Knowledge of company goals and progress
CSSBB 2014


Research and Development
C Number of development projects in progress

C Percentage of projects meeting budget
C Number of projects behind schedule
C Development expenses versus sales income
C Reliability of design change requests
Engineering
C Evaluation of product performance

C Number of corrective action requests
C Percentage of closed corrective action requests
C An assessment of measurement control
C Availability of internal technical assistance
Manufacturing
C Key machine and process capabilities

C Machine downtime percentages
C Average cycle times (key product lines)
C Measurement of housekeeping control
C Adequacy of operator training
CSSBB 2014

MEASUREMENT SYSTEMS/METROLOGY
Metrology
Metrology is the science of measurement. The word
metrology derives from two Greek words: matron
(meaning measure) and logos (meaning logic).
Metrology encompasses the following key elements:
C The establishment of measurement standards that

are both internationally accepted and definable
C The use of measuring equipment to correlate the

extent that product and process data conforms to
specification
C The regular calibration of measuring equipment,

traceable to established international standards
CSSBB 2014

Units of Measurement
The SI system classifies measurements into seven
distinct categories:
1. Length (meter)
2. Time (second)
3. Mass (kilogram)
4. Electric current (ampere)
5. Temperature (Kelvin)
Temp F = 1.8 (Temp C) + 32

Temp C = (Temp F - 32) ÷ 1.8
Temp K = Temp C + 273.15
6. Light (candela)
7. Amount of substance (mole)

CSSBB 2014

SI System Units
Listed below is a table of significant SI system units:
QUANTITY MEASURED UNIT SYMBOL FORMULA
FUNDAMENTAL UNITS
AMOUNT OF SUBSTANCE MOLE mol

LENGTH METER m
MASS KILOGRAM kg
TIME SECOND s
ELECTRIC CURRENT AMPERE A
TEMPERATURE DEGREE KELVIN K
LUMINOUS INTENSITY CANDELA cd
SUPPLEMENTARY UNITS
PLANE ANGLE RADIAN rad

SOLID ANGLE STERADIAN sr
DERIVED UNITS
AREA SQUARE METER m2

VOLUME CUBIC METER m3
FREQUENCY HERTZ (ONE CYCLE PER SECOND) Hz (s-1)
DENSITY KILOGRAM PER CUBIC METER kg/m3
VELOCITY METER PER SECOND m/s
ANGULAR VELOCITY RADIAN PER SECOND rad/s
ACCELERATION METER PER SECOND SQUARED m/s2
ANGULAR ACCELERATION RADIAN PER SECOND SQUARED rad/s2
FORCE NEWTON N (kg @ m/s2)
PRESSURE NEWTON PER SQUARE METER N/m2
KINEMATIC VISCOSITY SQ METER PER SECOND m2/s
DYNAMIC VISCOSITY NEWTON-SECOND PER SQ METER N @ s/m2
WORK, ENERGY, QUANTITY OF HEAT JOULE J (N @ m)
POWER WATT W (J/s)
ELECTRIC CHARGE COULOMB C (A @ s)
VOLTAGE (ELECTROMOTIVE FORCE) VOLT V (W/A)
ELECTRIC FIELD STRENGTH VOLT PER METER V/m
ELECTRIC RESISTANCE OHM Ω (V/A)
ELECTRIC CAPACITANCE FARAD F (A @ s/V)
MAGNETIC FLUX WEBER Wb (V @ s)
INDUCTANCE HENRY H (V @ s/A)
MAGNETIC FLUX DENSITY TESLA T (Wb/m2)
MAGNETIC FIELD STRENGTH AMPERE PER METER A/m
MAGNETOMOTIVE FORCE AMPERE A
LUMINOUS FLUX LUMEN lm (cd @ sr)
LUMINANCE CANDELA PER SQ METER cd/m2
ILLUMINATION LUX lx (lm/m2)
CSSBB 2014

Calibration
Calibration is the comparison of a measurement
standard or instrument of known accuracy with another
standard or instrument to detect, correlate, report or
eliminate by adjustment, any variation in the accuracy of
the item being compared. The elimination of
measurement error is the primary goal of calibration
systems.
Total Product Variability

The total variability in a product includes the variability
of the measurement process:
σ2Total = σ2Process + σ2Measurement

CSSBB 2014

Measurement Error
The error of a measuring instrument is the indication of
a measuring instrument minus the true value.
σ2Error = σ2Measurement - σ2True
or σ2Measurement = σ2True + σ2Error
The precision of measurement can best be improved

through the correction of the causes of variation in the
measurement process. The confidence interval for the
mean of measurements is reduced by obtaining multiple
readings according to the central limit theorem using
the following relationship.
Readings
 Measure =
n
The formula states that halving the error of

measurement requires quadrupling the number of
measurements.
CSSBB 2014

Measurement Error (Continued)

There are many reasons that a measuring instrument
may yield erroneous variation, including the following
categories:
Operator Variation: This error occurs when the operator

of a measuring instrument obtains measurements
utilizing the same equipment on the same standards and
a pattern of variation occurs.
Operator to Operator Variation: This error occurs when

two operators of a measuring instrument obtain
measurements utilizing the same equipment on the
same standards and a pattern of variation occurs
between the operators about the bias between them.
Equipment Variation: This error occurs when sources

of variation within the equipment surface through
measurement studies.
CSSBB 2014

Measurement Error (Continued)

Material Variation: This error occurs when the testing of
a sample destroys or changes the sample prohibiting
retesting.
Procedural Variation: This error occurs when there are

two or more methods to obtain a measurement resulting
in multiple results.
Software Variation: With software generated

measurement programs, variation in the software
formulas may result in errors, even after identical inputs.
Laboratory to Laboratory Variation: This error is

common when procedures for measurement vary from
laboratory-to-laboratory. The advent of standardized
testing such as the ASTM procedures have been
developed to correct this type of error.
CSSBB 2014

Calibration Interval
It is generally accepted that the interval of calibration of
measuring equipment be based on stability, purpose,
and degree of usage. The following basic calibration
principles must be applied:
1. The stability of a measurement instrument refers to

the ability of a measuring instrument to consistently
maintain its metrological characteristics over time.
2. The purpose or function of the measurement

instrument is important. In general, the critical
applications will increase calibration frequency and
minor applications would decrease calibration
frequency.
3. The degree of usage refers to the environment as a

whole. Thought must be given as to how often an
instrument is utilized and to what environmental
conditions an instrument is exposed.
CSSBB 2014

Calibration Interval (Continued)

Intervals should be shortened if previous calibration
records and equipment usage indicate this need. The
interval can be lengthened if the results of prior
calibrations show that accuracy will not be sacrificed.
Intervals of calibrations are not always stated in

standard lengths of time such as annually, bi-annually,
quarterly, etc. A method gaining recent popularity is the
verification methodology. This technique requires that
very short verification frequencies be established for
instruments placed into the system.
Measuring and test equipment should be traceable to

records that indicate the date of the last calibration, by
whom it was calibrated and when the next calibration is
due. Coding is sometimes used.
CSSBB 2014

Calibration Standards
Any system of measurement must be based on
fundamental units that are virtually unchangeable.
Today, a master international kilogram is maintained in
France. In the SI system, most of the fundamental units
are defined in terms of natural phenomena that are
unchangeable. This recognized true value is called the
standard.
When a measurement of a characteristics is made, the

dimension being measured is compared to a standard.
The standard may be a yardstick, a pair of calipers, or
even a set of gage blocks, but they all represent some
criteria against which an object is compared ultimately
to national and international standards.
CSSBB 2014

Calibration Standards (Continued)

Linear standards are easy to define and describe if they
are divided into functional levels. There are five levels
in which linear standards are usually described.
Working Level
This level includes gages used at the work center.
Calibration Standards
These are standards to which working level standards
are calibrated.
Functional Standards
This level of standards is used only in the metrology
laboratory of the company for measuring precision
work and calibrating other standards.
Reference Standard
These standards are certified directly to the NIST and
are used in lieu of national standards.
National and International Standards

This is the final authority of measurement to which all
standards are traceable.
CSSBB 2014

Calibration Standards (Continued)

Since the continuous use of national standards is
neither feasible nor possible, other standards are
developed for various levels of functional utilization.
National standards are taken as the central authority for
measurement accuracy, and all levels of working
standards are traceable to this “grand” standard. The
downward direction of this traceability is shown as
follows:
1. National Institute of Standards and Technology

2. Standards Laboratory
3. Metrology Laboratory
4. Quality Control System (Inspection Department)
5. Work Center
The calibration of measuring instruments is necessary

to maintain accuracy, but does not necessarily increase
precision. Precision most generally stays constant over
the working range of the instrument.
CSSBB 2014

ISO 10012 Synopsis

The following are basic summaries of what must be
accomplished to meet ISO requirements.
C All measuring equipment must be identified,

controlled, and calibrated. Records must be kept.
C The system for evaluating measuring equipment to

meet the required sensitivity, accuracy, and
reliability must be defined in written procedures.
C The calibration system must be evaluated on a

periodic basis by audits and reviews.
C The actions involved with the entire calibration

system must be planned.
C The uncertainty of measurement must be

determined.
C The methods used to confirm the measuring

equipment and devices must be documented.
CSSBB 2014

ISO 10012 Synopsis (Continued)

C Records must be kept on the methods used to
calibrate measuring and test equipment.
C Suitable procedures must be in place to ensure that

nonconforming measuring equipment is not used.
C A labeling system must be in place that shows the

identification of each piece of measuring equipment
or device and its status.
C The frequency of recalibration for each measuring

device must be established, documented, and be
based upon.
C Where adjustments may be made that may logically

go undetected, sealing of the adjusting devices or
case is required.
C Procedures must define controls when any outside

source is used regarding the calibration or supply
of measuring equipment.
CSSBB 2014

ISO 10012 Synopsis (Continued)

C Calibrations must be traceable to national
standards. If no national standard is available, the
method of establishing and maintaining the
standard must be documented.
C Measuring equipment will be handled, transported

and stored according to established procedures.
C Where uncertainties accumulate, the method of

calculation of the uncertainty must be specified.
C Gages, measuring equipment, and test equipment

will be used, calibrated, and stored in conditions
that ensure the stability of the equipment.
C Documented procedures are required for the

qualifications and training of personnel that make
measurement or test determinations.
CSSBB 2014

Measurement Methods
Transfer Tools
Transfer tools (such as spring calipers) have no reading

scale. Jaws on these instruments measure the length,
width, or depth in question by positive contact. The
dimension measurement is then transferred to another
measurement scale for direct reading.
Attribute Gages
Attribute gages are fixed gages which typically are used

to make a go/no-go decision. Examples of attribute
instruments are master gages, plug gages, contour
gages, thread gages, limit length gages, assembly
gages, etc. Attribute data indicates only whether a
product is good or bad. Attribute gages are quick and
easy to use but provide minimal information.
CSSBB 2014

Measurement Methods (Continued)

Variable Gages
Variable measuring instruments provide a physical

measured dimension. Examples of variable instruments
are rulers, vernier calipers, micrometers. Variable data
is often useful for process capability determination and
may be monitored via control charts.
Reference/Measuring Surfaces
A reference surface is the surface of a measuring tool

that is fixed. The measuring surface is movable.
CSSBB 2014

Instrument Selection
The terms measuring tool, instrument, and gage are
often used interchangeably in this text. The description
of all available instruments is beyond the scope of this
Primer. The reader is advised to seek other sources for
a more in-depth treatment of specific instruments. The
causes of measurement error are discussed later in this
Section.
Obviously, the appropriate gage should be used for the

required measurement. Some gage accuracies and
applications follow.
CSSBB 2014

Instrument Selection (Continued)

Gage Accuracy Application
Adjustable snap Usually accurate within Measures diameters on a production
gages 10% of the tolerance. basis where an exact measurement
is needed.
Air gages Accuracy depends upon Used to measure the diameter of a
t h e g a g e d e s i g n . bore or hole. However, other
Measurements of less applications are possible.
than 0. 000050" are
possible.
Automatic sorting Accurate within 0.0001". Used to sort parts by dimension.
gages
Combination Accurate within one Used to make angular checks.
square degree.
Coordinate Accuracy depends upon Can be used to measure a variety of
measuring the part. Axis accuracies characteristics, such as contour,
machines are within 35 millionths taper, radii, roundness, squareness,
and T.I.R. within etc.
0.000005".
Dial bore gages Accurate within 0.0001" Used to measure bore diameters,
using great care. tapers, or out-of-roundness.
Dial indicator Accuracy depends upon Measures a variety of features such
the type of indicator. as: flatness, diameter,
Some measure within concentricity, taper, height, etc.
0.0001".
Electronic Accurate from 0.00001" to Used where the allowable tolerance
comparator 0.000001". is 0.0001" or less.
CSSBB 2014


Fixed snap gages No set accuracy. Normally used to determine if
diameters are within specification.
Flush pin gages Accuracy of about 0.002". Used for high volume single purpose
applications.
Gage blocks Accuracy of the gage Gage blocks are best adapted for
block depends upon the precision machining and as a
grade. Normally the comparison master.
accuracy is 0.000008" or
better.
Height verniers M e c h a n i c a l m o d e l s Used to check dimensional
measure to 0.0001". tolerances on a surface plate.
Some digital models
attain 0.00005".
Internal and No exact reading. Will Used for measuring inside and
external thread discriminate to a given outside pitch thread diameters.
gages specification limit.
Micrometer Mechanical accuracy is Used for checking large hole
(inside) about 0.001". Some diameters.
digital models are
accurate to 0.00005".
Micrometer Mechanical accuracy is Normally used to check diameter or
(outside) about 0.001". Some thickness. Special models can
d i g i t a l m o d e l s a r e check thread diameters.
Optical The accuracy can be Measures difficult contours and part
comparator within 0.0002". configurations.
Optical flat Depending on operator Used only for very precise tool room
skill, accurate to a few work. Best used for checking
millionths of an inch. flatness.
Plug gages Accuracy very good for Checking the diameter of drilled or
checking the largest or reamed holes. Will not check for out
smallest hole diameter. of roundness.
CSSBB 2014


Precision straight Visual 0.10". With a feeler Used to check flatness, waviness or
edge gage 0.003". squareness of a face to a reference
plane.
Radius & Accuracy is no better than Used to check small radii, and
template gages 0.015". contours.
Ring gages Will only discriminate Best application is to approximate a
against diameters larger mating part in assembly. Will not
or smaller than the print check for out of roundness.
specification.
Split sphere & No better than 0.0005" Used for measuring small hole
telescope using a micrometer diameters.
graduated in 0.0001".
Steel ruler or No better than 0.015". Used to measure heights, depths,
scale diameters, etc.
Surface plates Flatness expected to be Used to measure the overall flatness
no better than 0.0005" of an object.
between any 2 points.
Tapered parallels U s i n g a n a c c u r a t e Used to measure bore sizes in low
micrometer, the accuracy volume applications.
is about 0.0005".
Tool maker's flat Accuracy is no better than Used with a surface plate and gage
0.0005" depending upon blocks to measure height.
the instrument used to
measure the height.
Vernier calipers About 0.001". Some Used to check diameters and
d i g i t a l m o d e l s a r e thickness.
Vernier depth About 0.001". Some Used to check depths.
gage digital models are
CSSBB 2014

Attribute Screens
Attribute screens are screening tests performed on a
sample with the results falling into one of two
categories, such as acceptable or not acceptable.
Screen tests are conducted on either the entire
population of items or on a significantly large proportion
of the population. Screen tests are nondestructive.
Screening programs have the following characteristics:
C A clearly defined purpose

C High sensitivity to the attribute being measured
C High specificity to the attribute being measured
C Benefits of the program outweigh the costs
C Measured attributes identify major problems
C Results lead to useful actions
Common applications of screening are reliability

assessments and medical screening.
CSSBB 2014

Tool Care
Measuring instruments are typically expensive and
should be treated with care to preserve their accuracy
and longevity. Some instruments require storage in a
customized case or controlled environment when not in
use. Even sturdy hand tools are susceptible to wear and
damage. Hardened steel tools require a light film of oil
to prevent rusting. Care must be taken in the
application of oil since dust particles will cause buildup
on the gage's functional surfaces. Measuring tools must
be calibrated on a scheduled basis as well as after any
suspected damage.
CSSBB 2014

Gage Blocks
Gage blocks are used in almost every shop
manufacturing a product requiring mechanical
inspection. They are used to set a length dimension for
a transfer measurement, and for calibration of a number
of other tools.
There are three basic gage block forms - rectangular,

square and round. The rectangular and square varieties
are in much wider use. Generally, gage blocks are made
from high carbon or chromium alloyed steel. Tungsten
carbide, chromium carbide, and fused quartz are also
used. All gage blocks are manufactured with tight
tolerances on flatness, parallelism, and surface
smoothness.
CSSBB 2014

Gage Blocks (Continued)
Federal Accuracy Grades Accuracy

in Length*
New Designation Old Designation
0.5 AAA ± 0.000001
1 AA ± 0.000002
2 A+ + 0.000004
- 0.000002
3 A&B + 0.000008
- 0.000004
Master blocks are grade 0.5 or 1

Inspection blocks are grade 1 or 2
Working blocks are grade 3
* Applies to gage blocks up to 1". The accuracy

tolerance then increases as the gage block size
increases.
CSSBB 2014

Gage Blocks (Continued)

Gage blocks should always be handled on the non-
polished sides. Blocks should be cleaned prior to
stacking with filtered kerosene, benzene or carbon
tetrachloride. A soft clean cloth or chamois should be
used. A light residual oil film must remain on blocks for
wringing purposes.
Block stacks are assembled by a wringing process

which attaches the blocks by a combination of
molecular attraction and the adhesive effect of a very
thin oil film. Air between the block boundaries is
squeezed out. The sequence is shown below.
Hold Crosswise Swivel the Pieces Slip into Position Finished Stack
CSSBB 2014

Gage Block Sets

Individual gage blocks may be purchased up to 20" in
size. Naturally, the length tolerance of the gage blocks
increases as the size increases. Typical gage block sets
vary from 8 to 81 pieces based upon the needed
application. The contents of a typical 81 piece set are:
Ten-thousandth blocks (9): 0.1001, 0.1002, ..., 0.1009

One-thousandth blocks (49): 0.101, 0.102, ........., 0.149
Fifty-thousandth blocks (19): 0.050, 0.100, ........., 0.950
One inch blocks (4): 1.000, 2.000, 3.000, 4.000
For the purpose of stack protection, some gage

manufacturers provide wear blocks that are either 0.050"
or 0.100" in thickness.
CSSBB 2014

Calipers
Calipers are used to measure length. The length can be
an inside dimension, outside dimension, height, or
depth. Some calipers are used for only one of these
lengths, while other calipers can be used to measure all
four types of lengths. Calipers are generally one of four
types:
C Spring calipers C Dial calipers

C Vernier calipers C Digital calipers
Spring Calipers
Spring calipers are transfer tools that perform a rough
measurement of wide, awkward or difficult to reach part
locations. These tools usually provide a measurement
accuracy of approximately 1/16 of an inch.
CSSBB 2014

Vernier Calipers
Vernier calipers use a vernier scale to indicate the
measurement of length. Length, depth and height are
variations of the length measurement capability they
provide. Resolution of vernier calipers is often 0.001
inch. Vernier calipers have been replaced with dial or
digital calipers in many applications.
The Vernier Scale

Vernier scales are used on a variety of measuring
instruments such as height gages, depth gages, inside
or outside vernier calipers and gear tooth verniers.
Except for the digital varieties, readings are made
between a vernier plate and beam scales. A vernier
scale reading example is shown in the Primer.
CSSBB 2014

Dial Calipers
Dial calipers function in the same way as vernier
calipers, however the measurement is indicated by a
combination of a scale reading to the nearest 0.1 of an
inch and a dial indicating the resolution to 0.001 of an
inch. The dial hand typically makes one revolution for
each 0.1 of an inch of travel of the caliper jaws.
Digital Calipers
Digital calipers use a digital display instead of the dial
and scale found in dial calipers. Most digital calipers
have the ability to be read in either inches or millimeters,
and the zero point can be set at any point along the
travel.
CSSBB 2014

Optical Comparators
A comparator is a device for comparing a part to a form
that represents the desired part contour or dimension.
The relationship of the form with the part indicates
acceptability.
Surface Plates
To make a precise dimensional measurement, there
must be a reference plane or starting point. Since a
perfectly flat reference plane does not exist, a surface
plate is commonly used.
Surface plates are customarily used with accessories

like: a toolmaker's flat, angles, parallels, V blocks and
cylindrical gage block stacks. Dimensional
measurements are taken from the plate up since the
plate is the reference surface. Surface plates must
possess the following important characteristics:
C Sufficient strength and rigidity

C Sufficient accuracy for the measurements
CSSBB 2014

Micrometers
Micrometers are commonly used hand-held measuring
devices. Micrometers may be purchased with frame
sizes from 0.5" to 48". Normally, the spindle gap and
design permits a 1" reading span. Most common “mics”
have an accuracy of 0.001 of an inch. With the addition
of a vernier scale, an accuracy of 0.0001 of an inch can
be obtained. Improvements in micrometers have led to
“supermicrometers” which are able to make linear
measurements to one millionth of an inch.
Ring Gages
Ring gages are used to check external cylindrical
dimensions, and may also be used to check tapered,
straight, or threaded dimensions. A pair of rings with
hardened bushings are generally used. One bushing
has a hole of the minimum tolerance.
CSSBB 2014

Plug Gages
Plug gages are generally go/no-go gages, and are used
to check internal dimensions. The average plug gage is
a hardened and precision ground cylinder about an inch
long. The go/no-go set is usually held in a hexagonal
holder with the go plug on one end and the no-go plug
on the other end.
Dial Indicators
Dial indicators are mechanical instruments for
measuring distance variations. Most dial indicators
amplify a contact point reading by use of an internal
gear train mechanism. An illustration is shown in the
Primer.
CSSBB 2014

Pneumatic Gages
There are two general types of pneumatic amplification
gages in use. One type is actuated by varying air
pressure and the other by varying air velocity at
constant pressure. Depending upon the amplification
and the scale, measurements can be read to millionths
of an inch.
Interferometry
The greatest possible accuracy and precision are
achieved by using light waves as a basis for
measurement. A measurement is accomplished by the
interaction of light waves that are 180E out of phase.
This phenomenon is known as interference.
CSSBB 2014

Laser Designed Gaging

The use of lasers have been prevalent when the intent of
inspection is a very accurate non-contact measurement.
The laser beam is transmitted from one side of the gage
to a receiver on the opposite side of the gage.
Measurement takes place when the beam is broken by
an object and the receiver denotes the dimension of the
interference to the laser beam.
Coordinate Measuring Machines (CMM)

Coordinate measuring machines are used to verify
workpiece dimensions using computer controlled
measurements which are taken on three mutually
perpendicular axes. Workpieces are placed on a surface
plate and a probe is maneuvered to various contact
points to send an electronic signal back to the computer
that is recording the measurements.
CSSBB 2014

Non-Destructive Testing and Evaluation

Non-destructive testing (NDT) and non-destructive
evaluation (NDE) techniques evaluate material
properties without impairing the future usefulness of the
items being tested. Today, there is a large range of NDT
methods available, including ultrasonic, radiography,
fluoroscopy, microwave, magnetic particle, liquid
penetrant, eddy current, and holography. The
advantages of NDT techniques include the use of
automation, 100% product testing and the guarantee of
internal soundness.
Visual Inspection
One of the most frequent inspection operations is the
visual examination of products, parts, and materials.
The color, texture, and appearance of a product gives
valuable information if inspected by an alert observer.
CSSBB 2014

Ultrasonic Testing
The application of high frequency vibration to the
testing of materials is a widely used and important non-
destructive testing method. Ultrasonic waves are
generated in a transducer and transmitted through a
material which may contain a defect. A portion of the
waves will strike any defect present and be reflected or
“echoed” back to a receiving unit, which converts them
into a “spike” or “blip” on a screen.
Ultrasonic inspection has also been used in the

measurement of dimensional thickness.
CSSBB 2014

Magnetic Particle Testing

Magnetic particle inspection is a non-destructive
method of detecting the presence of many types of
defects or voids in ferromagnetic metals or alloys. This
technique can be used to detect both surface and
subsurface defects in any material capable of being
magnetized.
The first step in magnetic particle testing is to magnetize

a part with a high amperage, low voltage electric
current. Then fine, steel particles are applied to the
surface of the test part. These particles will align
themselves with the magnetic field and concentrate at
places where magnetic flux lines enter or leave the part.
The test part is examined for concentrations of magnetic
particles which indicate that discontinuities are present.
CSSBB 2014

Liquid Penetrant Testing

Liquid penetrant inspection is a rapid method for
detecting open surface defects in both ferrous and
nonferrous materials. It may be effectively used on
nonporous metallic and nonmetallic materials. The size
of dye molecules used in fluorescent penetrant
inspection are so small that there may be no surface
cracks too small for penetration.
The factors that contribute to the success of liquid

penetrant inspection are the ability of a penetrant to
carry a dye into a surface defect and the ability of a
developer to contrast that defect by capillary attraction.
CSSBB 2014

Eddy Current Testing

Eddy currents involve the directional flow of electrons
under the influence of an electromagnetic field.
Nondestructive testing applications require the
interaction of eddy currents with a test object. This is
achieved by:
C Measuring the flow of eddy currents in a material

having virtually identical conductivity
characteristics as the test piece
C Comparing the eddy current flow in the test piece

(which may have defects) with that of the standard
CSSBB 2014

Radiography
Many internal characteristics of materials can be
photographed and inspected by the radiographic
process. Radiography is based on the fact that gamma
and X-rays will pass through materials at different levels
and rates. Therefore, either X-rays or gamma rays can
be directed through a test object onto a photographic
film and the internal characteristics of the part can be
reproduced and analyzed.
CSSBB 2014

Neutron Radiography
Neutron radiography is a fairly recent radiographic
technique that has useful and unique applications.
Neutrons are uncharged and move through materials
unaffected by density. The test object is placed in a
neutron beam in front of an image detector.
Related Techniques
There have been new developments in the radiographic
field of non-destructive testing, several common recent
applications include fluoroscopy, gamma radiography,
televised X-ray (TVX), microwave testing, and
holographic inspection.
CSSBB 2014

Force Measurement Techniques

A brief description of common force measurement tests
is listed below.
Tensile Test
The tensile test is performed by applying a uniaxial load

to a test bar and gradually increasing the load until it
breaks. The load is then measured against the
elongation using an extensometer. The tensile data may
be analyzed using a stress-strain curve.
Shear Test
Shear strength is the ability to resist a “sliding past”

type of action when parallel, but slightly off-axis, forces
are applied. Shear can be applied in either tension or
compression.
CSSBB 2014

Force Measurement Techniques (Cont’d)

Compression Test
Compression is the result of forces pushing toward

each other. The compression test is run much like the
tensile test. The specimen is placed in a testing
machine, a load is applied and the deformation is
recorded. A compressive stress-strain curve can be
drawn from the data.
Fatigue Test
Fatigue strength is the ability of material to take

repeated loading. There are several types of fatigue
testing machines. In all of them, the number of cycles
are counted until a failure occurs and the stress used to
cause the failure is determined.
CSSBB 2014

Titration
Titration is a method of analysis that allows a
determination of the precise endpoint of a reaction and
therefore the precise quantity of reactant in the titration
flask. A buret is used to deliver the second reactant to
the flask and an indicator or pH Meter is used to detect
the endpoint of the reaction. Titrations are used in
chemical analysis.
CSSBB 2014

Hardness Measurement
Hardness testing (which measures the resistance of any
material against penetration) is performed by creating
an indentation on the surface of a material with a hard
ball, a diamond pyramid, or cone and then measuring
the depth of penetration. Hardness testing is often
categorized as a non-destructive test since the
indentation is small and may not affect the future
usefulness of the material. The major hardness tests
are discussed in the Primer and summarized in the
following table.
CSSBB 2014

Hardness Measurement (Continued)

Type Technique Penetrator Loading Scale
HBW
Area of 10 mm 500, 1500 or
Brinell or
Penetration Ball 3000 kg
HBS
Presence
File File Manual HRC
of Scratch
Area of Pyramidal
Knoop 25 to 3600 g HK
Penetration Diamond
Presence 10 Units
Mohs Manual
of Scratch Stones Mohs
Diamond HRB,
Depth of Point or 60-100- HRC,
Rockwell
Penetration 1/16 or 1/8 150 kg and
Ball others
Diamond
15N,
Point
Rockwell Depth of 15, 30 or 30T,
or
Superficial Penetration 45 kg 45X,
1/16 or 1/8
etc.
Ball
Height of 40 Grain Units
Shore Gravity
Bounce Weight Shore
Vibration Vibrating
Sonodur N.A. BHN
Frequency Rod
Area of Pyramidal 25 g to
Vickers HV
Penetration Diamond 120 kg
CSSBB 2014

Torque Measurement
Torque measurement is required when the product is
held together by nuts and bolts. The wrong torque can
result in the assembly failing due to a number of
problems. Torque is described as a force producing
rotation about an axis. The formula for torque is:
Torque = Force x Distance
For example, a force of 2 pounds applied at a distance

of 3 feet equals 6 lbf-ft.
Impact Test
Impact strength is a material's ability to withstand
shock. Tests such as Charpy and Izod use notched
samples which are struck with a blow from a calibrated
pendulum. The major difference between the two are
the way the bar is anchored and the speed in which the
pendulum strikes the bar.
CSSBB 2014

The Steel Rule

The steel rule is a widely used factory measuring tool for
direct length measurement. Steel rules and tapes are
available in different degrees of accuracy and are
typically graduated on both edges. See the drawing
below.
The fine divisions on a steel rule (one thirty-seconds on

the one above) establish its discrimination which are
typically 1/32, 1/64, or 1/100 of an inch. Obviously,
measurements of 0.010" or finer should be performed
with other tools (such as a digital caliper).
CSSBB 2014
VI. MEASURE - DATA

QUESTIONS
6.3. What is the best definition of takt time?
a. It is a calculated time element that equals customer demand

b. It is the speed at which parts must be manufactured in order to
satisfy demand
C. It is the heartbeat of any lean system
d. It is the application of kaizen to continuous flow manufacturing
6.6. Which of the following is most important when calibrating a piece

of equipment?
a. The calibration sticker

b. The maintenance history card
c. The standard used
d. The calibration interval
6.8. Random selection of a sample:
a. Theoretically means that each item in the lot had an equal chance
to be selected
b. Ensures that the sample average will equal the population average
c. Means that a table of random numbers was used to dictate the
selection
d. Is a meaningless theoretical requirement
Answers: 6.3. a, 6.6 c, 6.8. a

CSSBB 2014
VI. MEASURE - DATA

QUESTIONS
6.11. Which of the following enterprise measurements could be applied

to suppliers?
a. Sales dollars compared to marketing dollars

b. Percentage of early deliveries
c. Number of projects behind schedule
d. Average cycle times
6.12. Calibration intervals should be adjusted when:
a. No defective product is reported as acceptable due to

measurement errors
b. Few instruments are scrapped during calibration
c. The results of previous calibrations reflect few out of tolerance
conditions during calibration
d. A particular characteristic on the gage is consistently found to be
out of tolerance
6.13. Value stream mapping means:
a. Flow charting techniques

b. An identification of inputs, tasks, and outputs
c. A pictorial view that identifies process steps
d. A graphical flow charting technique that shows material and
information flows
Answers: 6.11. b, 6.12 d, 6.13. d

CSSBB 2014
VI. MEASURE - DATA

QUESTIONS
6.19. If the throughput rate for an operation is 3,600 units per hour, what
is the cycle time in seconds?
a. 1 second
b. 2 seconds
c. 30 seconds
d. 60 seconds
6.20. The measurement and recording of the length of a product in

centimeters would reflect the application of which measurement
scale?
a. Nominal
b. Ordinal
c. Interval
d. Ratio
6.22. One use of recording check sheets is:
a. Automated charting of variable data

b. Collecting tally counts of attribute data
c. Identification of process variables
d. To create a process map
Answers: 6.19. a, 6.20 d, 6.22. b

CSSBB 2014
VI. MEASURE - DATA

QUESTIONS
6.24. Little's law is widely used as a determination of:
a. Value added times

b. Process flow times
c. Productive maintenance requirements
d. Nonconforming material control
6.26. Typically, spaghetti diagrams focus on:
a. Machine locations
b. Employee activities
c. Information and material flow
d. Takt time for all key products
6.27. What set of basic quality tools would be most applicable for a work
team to use when there is a desire to follow procedures and work
instructions more closely?
a. Pareto and affinity diagrams

b. Data sheets and histograms
c. Checklists and flow charts
d. Fishbone diagrams and control charts
Answers: 6.24. b, 6.26. c, 6.27. c

© QUALITY COUNCIL OF INDIANA VII-1 (571)
CSSBB 2014
VII. MEASURE - STATISTICS
DO NOT PUT YOUR FAITH IN

WHAT STATISTICS SAY UNTIL
YOU HAVE CAREFULLY
CONSIDERED WHAT THEY DO
NOT SAY.
WILLIAM W. WATT
CSSBB 2014
VII. MEASURE - STATISTICS V.D.1

BASIC STATISTICS/BASIC TERMS
Measure - Statistics
Measure - Statistics is described in the following

topic areas:
C Basic statistics
C Probability
C Process capability
Basic Statistics
Basic statistics is presented in the following topic areas:
C Basic terms
C Central limit theorem
C Descriptive statistics
C Graphical methods
C Statistical conclusions
CSSBB 2014

BASIC STATISTICS/BASIC TERMS
Basic Terms
Continuous A distribution containing infinite
Distribution: (variable) data points that may be
displayed on a continuous
measurement scale. Examples: normal,
uniform, exponential, and Weibull
distributions.
Discrete A distribution resulting from countable
Distribution: (attribute) data that has a finite number
of possible values. Examples:
binomial, Poisson, and hypergeometric
distributions.
Parameter: The true numeric population value,
often unknown, estimated by a statistic.
Population: All possible observations of similar
items from which a sample is drawn.
Statistic: A numerical data value taken from a
sample that may be used to make an
inference about a population.
CSSBB 2014

BASIC STATISTICS/CENTRAL LIMIT THEOREM
Central Limit Theorem

If a random variable, X, has mean µ, and finite variance
σ2, as n increases, X̄ approaches a normal distribution
 2x
with mean µ and variance  . Where, 
2
x2
2
x2
= and n is
n
the number of observations on which each mean is
based.
Normal Distribution
of Sample Means
F(x)
Distribution of
Individuals
The sample means X̄i will be more normally distributed

around μ than individual readings Xj. The distribution of
sample means approaches normal regardless of the
shape of the parent population. This is why X̄ - R
control charts work!
CSSBB 2014

Central Limit Theorem (Continued)

The spread in sample means X̄i is less than Xj with the
standard deviation of X̄i equal to the standard deviation
of the population (individuals) divided by the square
root of the sample size. SX̄ is referred to as the standard
error of the mean:
x sx
x  Which is estimated by sx 
n n
Assume the following are weight variation results: X̄ =

20 grams and σ = 0.124 grams. Estimate σX̄ for a sample
size of 4:
sx 0.124
Solution: sx = = = 0.062 grams
n 4
CSSBB 2014

Central Limit Theorem (Continued)

The significance of the central limit theorem on control
charts is that the distribution of sample means
approaches a normal distribution.
Population Distribution Population Distribution Population Distribution Population Distribution
n=2 n=2
n=2
n=2
n=4 n=4
n=4 n=4
n = 25
n = 25 n = 25
n = 25
Sampling Distribution of X Sampling Distribution of X Sampling Distribution of X Sampling Distribution of X
A variety of population distributions approach normality

for the sampling distribution of X ¯ as n increases.
CSSBB 2014

BASIC STATISTICS/DESCRIPTIVE STATISTICS
Descriptive Statistics
Descriptive statistics include measures of central
tendency, measures of dispersion, probability density
function, frequency distributions, and cumulative
distribution functions.
Measures of Central Tendency

Measures of central tendency represent different ways
of characterizing the central value of a collection of data.
Three of these measures will be addressed here: mean,
mode, and median.
The Mean (X-bar, X )
The mean is the total of all data values divided by the

number of data points.
X=
X
n
CSSBB 2014

Measures of Central Tendency (Cont’d)

The Mode
The mode is the most frequently occurring number in a

data set. It is possible for groups of data to have more
than one mode.
The Median (Midpoint)
The median is the middle value when the data is

arranged in ascending or descending order. For an
even set of data, the median is the average of the middle
two values.
CSSBB 2014

Measures of Central Tendency (Cont’d)

For a Normal Distribution
MEAN = MEDIAN = MODE
For a Skewed Distribution
MODE
MEDIAN
MEAN
CSSBB 2014

Measures of Dispersion
Other than central tendency, the other important
parameter to describe a set of data is spread or
dispersion. The three main measures of dispersion are
range, variance, and standard deviation.
Range (R)
The range of a set of data is the difference between the
largest and smallest values.
Variance (σ2, s2)

The variance, σ2 or s2, is equal to the sum of the squared
deviations from the mean, divided by the sample size.
The formula for variance is:
 X - 
2
Population,  2 =
N
  X - X
2
2
Sample, s =
n
The variance is equal to the standard deviation squared.

CSSBB 2014

Standard Deviation (σ, s)

The standard deviation is the square root of the
variance.
 X - 
2
Population,  =
N
  X - X
2
Sample, s =
n-1
N is used for a population and n - 1 for a sample to

remove potential bias in relatively small samples (less
than 30).
Coefficient of Variation (COV)

The coefficient of variation equals the standard
deviation divided by the mean and is expressed as a
percentage.
s σ
COV =  100%  or COV =  100% 
X μ
CSSBB 2014

Standard Deviation Determinations

The following methods for determining standard
deviation are discussed in the Primer.
C Classical method (manual)

C Shortcut method
C Calculator method
C Probability paper
C Control chart method
CSSBB 2014

Probability Density Function

The probability density function, f(x), describes the
behavior of a random variable. Typically, the probability
density function is viewed as the “shape” of the
distribution. It is normally a grouped frequency
distribution. Consider the histogram for the length of a
product shown below with an overlaid normal model.
The smooth curve above represents the population. The

differences between the sample data represented by the
histogram and the population data represented by the
smooth curve are due to sampling error or an incorrect
model.
CSSBB 2014

Probability Density Function (Continued)

The area below the probability density function to the
left of a given value, x, is equal to the probability of the
random variable represented on the x-axis being less
than the given value x. Since the probability density
function represents the entire sample space, the area
under the probability density function must equal one.
Stating these two requirements mathematically for

continuous distributions with f(x) > 0:

 f  x  dx = 1
-
For discrete distributions for all values of n with f(x) > 0:

n
 f x = 1
0
CSSBB 2014

Probability Density Function (Continued)

The figure below demonstrates how the probability
density function is used to compute probabilities. The
area of the shaded region represents the probability of
a single product drawn randomly from the population
having a length less than 185. This probability is 15.9%
and can be determined by using the standard normal
table (z table) discussed later in this Section.
0.030
0.025
Probability Density
0.020
0.015
0.010
0.005
0.000
155
159
163
167
171
175
179
183
187
191
195
199
203
207
211
215
219
223
227
231
235
239
243
Length
CSSBB 2014

Cumulative Distribution Function

The cumulative distribution function, F(x), denotes the
area beneath the probability density function to the left
of x. This is demonstrated below.
0.030
P r o b a b ility D e n s ity
0.025
0.020
0.015
0.010
0.005
0.000
155 159 163 167 171 175 179 183 187 191 195 199 203 207 211 215 219 223 227 231 235 239 243
Length
1.000
C u m u la tiv e D is tr ib u tio n F u n c tio n
0.800
0.600
0.400
0.200
0.000
155 159 163 167 171 175 179 183 187 191 195 199 203 207 211 215 219 223 227 231 235 239 243
Length
The area of the shaded region of the probability density

function in the figure above is 0.2525 which corresponds
to the cumulative distribution function at x = 190.
CSSBB 2014

BASIC STATISTICS/GRAPHICAL METHODS
Graphical Methods
Graphical methods include boxplots, stem and leaf
plots, scatter diagrams, run charts, histograms, and
normal probability plots.
Boxplots
The boxplot is a five number summary of the data. The
data median is a line dividing the box. The upper and
lower quartiles of the data define the ends of the box.
The minimum and maximum data points are drawn as
points at the end of lines (whiskers) extending from the
box.
.67
.65
9
8
Measurement Quality
.63
7
.61
6
.59
5
.57
4
.55 3
.53 2
.51 1
.49
1 2 3 4 5 6
CSSBB 2014

Stem and Leaf Plots

The stem and leaf diagram is a convenient, manual
method for plotting data sets. The diagram consists of
grouping the data by class intervals, as stems, and the
smaller data increments as leaves.
14
12
10
8
Frequency
6
4
2
0
41# 43# 45# 47# 49# 51# 53#
Strength
5
2
5 3 8
6 9 8 7 0 9
1 6 8 8 5 9 1
5 1 4 6 4 4 9 6 6
8 6 2 1 2 4 0 8 6 4 4
Leaf 2 4 8 2 4 5 0 6 8 3 0 2
Stem 0 1 2 3 4 5 6 7 8 9 0 1 2
4 4 4 4 4 4 4 4 4 4 5 5 5
CSSBB 2014

Scatter Diagrams
A scatter diagram is a graphic display of many XY
coordinate data points which represent the relationship
between two different variables. It is also referred to as
a correlation chart. Scatter diagrams can help
determine if a relationship exists and how to control the
effect of the relationship on the process.
In most cases, there is an independent variable and a

dependent variable. Traditionally, the dependent
variable is represented by the vertical axis and the
independent variable is represented by the horizontal
axis.
The dependent variable can be controlled if the

relationship is understood. Correlation originates from
the following:
C A cause-effect relationship
C A relationship between one cause and another
cause
C A relationship between one cause and two or more
other causes
CSSBB 2014

Scatter Diagrams Continued)
Low-positive High-positive No correlation
High-negative Non-linear Non-linear

Relationship Relationship
Not all scatter diagrams reveal a linear relationship.

CSSBB 2014

Sample Correlation Coefficient

A sample correlation coefficient “r” can be calculated to
determine the degree of association between two
variables.
n
X
i=1
i -X  Y - Y 
i
r=
 n 2 n 2
 i
i = 1
X - X 
  i
 i = 1
Y -Y 


The following interpretations apply:
r = –1.0 strong negative when X increases, Y decreases

r = –0.5 slight negative when X increases, Y generally
decreases
r= 0 no correlation the two variables are independent
r = +0.5 slight positive when X increases, Y generally
increases
r = +1.0 strong positive when X increases, Y increases
CSSBB 2014

Concluding Correlation Comments

C A correlation analysis seeks to uncover
relationships. Common sense must be liberally
applied. There are such things as nonsense
correlations.
C The line of “best fit” can be obtained by calculating

a “regression line.” However, the line can be
“eyeballed.” Draw a straight line through the points
attempting to have approximately one-half above
and one-half below.
C Scatter diagrams should always be analyzed prior

to making decisions in correlation statistics.
CSSBB 2014

Run (Trend) Charts

Data can be presented in either summary (static) or time
sequence (dynamic) fashion. For many business
activities, trend charts will show patterns that indicate if
a process is running normally or whether desirable or
undesirable changes are occurring.
Normal convention has time increasing across the page

and the measurement increasing up the page.
UPWARD TREND DOWNWARD TREND PROCESS SHIFT
100 100 100
80 80 80
60 60 60
40 40 40
20 20 20
0 0 0
1 5 10 15 20 1 5 10 15 20 1 5 10 15 20
UNUSUAL VALUES CYCLES

100 100
INCREASING VARIABILITY
100
80 80
80
60 60
60
40 40
40
20 20
20
0 0
0
1 5 10 15 20 1 5 10 15 20
1 5 10 15 20
CSSBB 2014

Histograms
Histograms have the following characteristics:
C Frequency column graphs that display a static

picture of process behavior. Histograms require a
minimum of 50-100 data points.
C A histogram is characterized by the number of data

points that fall within a given bar or interval. This is
commonly referred to as “frequency.”
C A stable process is frequently characterized by a

histogram exhibiting unimodal or bell-shaped
curves. A stable process is predictable.
C An unstable process is often characterized by a

histogram that does not exhibit a bell-shaped curve.
Obviously distribution shapes like exponential,
lognormal, beta, Weibull, Poisson, binomial,
hypergeometric, etc. exist as stable processes.
C When the bell curve is the approximate distribution

shape, variation around the bell curve is chance or
natural variation. Other variation is due to special
or assignable causes.
CSSBB 2014

Histogram Construction
FREQUENCY TALLY SPECIFICATION LIMITS 0.50 - 0.60
USL
LSL
.50
.51
.52
.53
.54
.55
.56
.57
.58
.59
.60
.61
.62
.63
.64
.65
CSSBB 2014

Histogram Examples
Column Graph Bar Graph Normal

Histogram
Histogram with Bimodal Histogram

Special Causes (May Also be Polymodal)
LSL USL
Negatively Skewed Truncated Histogram

(After 100% Inspection)
CSSBB 2014

BASIC STATISTICS/STATISTICAL CONCLUSIONS
Drawing Valid Statistical Conclusions

Analytical (Inferential) Studies
The objective of statistical inference is to draw

conclusions about population characteristics based on
the information contained in a sample. The steps
involved in statistical inference are:
C Define the problem objective precisely
C Decide if the problem will be evaluated by a one-tail

or two-tail test
C Formulate a null and an alternate hypothesis
C Select a test distribution and a critical value of the

test statistic reflecting the degree of uncertainty
that can be tolerated (the alpha, α, risk)
C Calculate a test statistic value from the sample
C Make an inference about the population by

comparing the calculated value to the critical value.
This step determines if the null hypothesis is to be
rejected.
CSSBB 2014

Drawing Valid Conclusions (Continued)

Null Hypothesis and Alternate Hypothesis
The null hypothesis is the hypothesis to be tested. The

null hypothesis directly stems from the problem
statement and is denoted as H0. The alternate
hypothesis must include all possibilities which are not
included in the null hypothesis and is designated H1.
Examples of null and alternate hypothesis:
Null hypothesis: Ho: Ya = Yb H o: A <

_B
Alternate hypothesis: H1: Ya =/ Yb H1: A > B
A null hypothesis can only be rejected, or fail to be

rejected, it cannot be accepted because of a lack of
evidence to reject it.
Test Statistic
In order to test a null hypothesis, a test calculation must

be made from sample information. This calculated value
is compared to an appropriate critical value. A decision
can then be made to reject or not reject the null
hypothesis.
CSSBB 2014


Types of Errors
When formulating a conclusion regarding a population

based on observations from a small sample, two types
of errors are possible:
C Type I error: This error results when the null

hypothesis is rejected when it is, in fact, true.
C Type II error: This error results when the null

hypothesis is not rejected when it should be
rejected.
The degree of risk (α) is normally chosen by the

concerned parties (α is normally taken as 5%) in arriving
at the critical value of the test statistic.
CSSBB 2014


Enumerative and Analytical Studies
Enumerative data is data that can be counted. Useful

tools for tests of hypothesis conducted on enumerative
data are the chi square, binomial, and Poisson
distributions. Deming defined a contrast between
enumeration and analysis:
Enumerative study: A study in which action will be

taken on the universe.
Analytical study: A study in which action will be taken

on a process to improve performance in the future.
Numerical descriptive measures create a mental picture

of a set of data. The measures calculated from a sample
are called statistics. When these measures describe a
population, they are called parameters.
Measures Statistics Parameters

Mean X̄ μ
Standard Deviation s σ
CSSBB 2014
VII. MEASURE - STATISTICS V.E.1

PROBABILITY/BASIC CONCEPTS
Probability
Probability is presented in the following topic areas:
C Basic concepts
C Commonly used distributions
C Other distributions
Most quality theories use statistics to make inferences

about a population based on information contained in
samples. The mechanism one uses to make these
inferences is probability.
Conditions for Probability

The probability of any event, E, lies between 0 and 1.
The sum of the probabilities of all possible events in a
sample space, S, = 1.
CSSBB 2014

Simple Events
An event that cannot be decomposed is a simple event,
E. The set of all sample points for an experiment is
called the sample space, S.
If an experiment is repeated a large number of times, N,

and the event, E, is observed nE times, the probability of
E is approximately:
nE
PE 
N
The probability of observing 3 on the toss of a single die

is:
1
PE3 =
6
What is the probability of getting 1, 2, 3, 4, 5, or 6 by

throwing a die?
PET = P  E1  + P  E2  + P  E3  + P  E4  + P  E5  + P  E6 
1 1 1 1 1 1
PET = + + + + + =1
6 6 6 6 6 6
CSSBB 2014

Compound Events
Compound events are formed by a composition of two
or more events. The two most important probability
theorems are additive and multiplicative. For the
following discussion, EA = A and EB = B.
I. Composition. Consists of two possibilities -- a union

or intersection.
A. Union of A and B
If A and B are two events in a sample space, S, the

union of A and B (A c B) contains all sample points
in event A, B, or both.
B. Intersection of A and B
If A and B are two events in a sample space, S, the

intersection of A and B (A 1 B) is composed of all
sample points that are in both A and B.
CSSBB 2014

Compound Events (Continued)

II. Event Relationships. There are three relationships
in finding the probability of an event:
complementary, conditional, and mutually
exclusive.
A. Complement of an Event
The complement of event A is all sample points in

the sample space, S, but not in A. The complement
of A is 1-PA.
B. Conditional Probabilities
The conditional probability of event A occurring,

given that event B has occurred is:
P  A  B
P  A|B  = if P  B   0
P B
Two events A and B are said to be independent if either:
P(A|B) = P(A) or P(B|A) = P(B)
Otherwise, the events are said to be dependent.

CSSBB 2014

Compound Events (Continued)

C. Mutually Exclusive Events
If event A contains no sample points in common

with event B, then they are said to be mutually
exclusive.
D. Testing for Event Relationships
Are events A and B mutually exclusive, complementary,

independent, or dependent? If events A and B contain
two sample points in common, they are not mutually
exclusive. If event B does not contain all points in S that
are not in event A, then they are not complementary.
CSSBB 2014

The Additive Law

If the two events are not mutually exclusive:
1. P (A c B) = P(A) + P(B) - P (A 1 B)
Note that P (A c B) is shown in many texts as P (A + B)

and is read as the probability of A or B.
If one owns two cars and the probability of each car

starting on a cold morning is 0.7, what is the probability
of getting to work?
P (A c B) = 0.7 + 0.7 - (0.7 x 0.7)

= 1.4 - 0.49
= 0.91 = 91%
CSSBB 2014

The Additive Law (Continued)

If the two events are mutually exclusive, the law reduces
to:
2. P (A c B) = P(A) + P(B) also P (A + B) = P(A) + P(B)
If the probability of finding a black sock in a dark room

is 0.4 and the probability of finding a blue sock is 0.3,
what is the chance of finding a blue or black sock?
P (A c B) = 0.4 + 0.3 = 0.7 = 70%
Note: The problem statements center around the word

“or.” Will car A or B start? Will one get a black or blue
sock?
CSSBB 2014

The Multiplicative Law

If events A and B are dependent, the probability of event
A influences the probability of event B.
For any two events, A and B, such that P(B) =/ 0:

P  A  B
1. P  A|B  = and P  A  B  = P  A|B  P  B 
P B
Note in some texts P (A 1 B) is shown as P(A C B) and is

read as the probability of A and B. P(B|A) is read as the
probability of B given that A has occurred.
If a shipment of 100 TV sets contains 30 defective units

and two samples are obtained, what is probability of
finding both defective? (Event A is the first sample and
the sample space is reduced, and event B is the second
sample.)
30 29 870
P  A  B = x = = 0.088
100 99 9900
P  A  B  = 8.8%
CSSBB 2014

The Multiplicative Law (Continued)

If events A and B are independent:
2. P (A 1 B) = P(A) X P(B)
One relay in an electric circuit has a probability of

working equal to 0.9. Another relay in series has a
chance of 0.8. What's the probability that the circuit will
work?
P (A 1 B) = 0.9 X 0.8 = 0.72

P (A 1 B) = 72%
Note: The problem statements center around the word

“and.” Will TV A and B work? Will relay A and B
operate?
CSSBB 2014

PROBABILITY/COMMON DISTRIBUTIONS
Commonly Used Distributions
Commonly used distributions include the following:
C Normal C Chi square

C Binomial C Student’s t
C Poisson C F distribution
Normal Distribution
The normal distribution has numerous applications. It
is useful when it is equally likely that readings will fall
above or below the average.
The normal probability density function is:
2
1 x -  
1 -  
f x = e 2  
,-<x<
 2
Where μ is the mean and σ is the standard deviation.

CSSBB 2014

Normal Distribution (Continued)

The normal probability density function is not skewed,
as shown in below.
0.4
Probability Density
0.3
0.2
0.1
0.0
-3 -2 -1 0 1 2 3
X
The density function shown above is the standard

normal probability density function. The standard
normal probability density function has a mean of 0 and
a standard deviation of 1.
CSSBB 2014


The normal probability density function cannot be
integrated implicitly. Because of this, a transformation
to the standard normal distribution is made, and the
normal cumulative distribution function or reliability
function is read from a table. The standard normal table
is shown in the Appendix. If x is a normal random
variable, it can be transformed to standard normal using
the expression:
x-
Z=

Other Z transformations include:
X-X X- X - Std X - Std

Z= = = =
σ σX  X
CSSBB 2014


A battery is produced with an average voltage of 60 and
a standard deviation of 4 volts. If 9 batteries are
selected at random, what is the probability that the total
voltage of the 9 batteries is greater than 530? The
expected total voltage for nine batteries is 540. The
expected standard deviation of the voltage of the total of
nine batteries is:
S2TOTAL = 9 x  4  = 144
2
STOTAL = 12
Transforming to 530 - 540

Z= = - 0.833
standard normal: 12
From the standard normal table, the area to the right of

z is 0.7976.
What is the probability that the average voltage of the 9

batteries is less than 62? The expected value is 60. The
standard deviation is :
σ 4 62 - 60
SX = = = 1.333 Thus, z = = 1.5
n 9 1.333
The area to the left of z is 1 - 0.0668 = 0.9332

CSSBB 2014

Binomial Distribution
The binomial distribution applies when the population is
large (N > 50) and the sample size is small compared to
the population. The binomial is best applied when the
sample size is less than 10% of N (n < 0.1N). Binomial
sampling is with replacement. It is most appropriate to
use when the proportion defective is equal to or greater
than 0.1. The binomial is an approximation to the
hypergeometric. The normal distribution approximates
the binomial when np > 5.
n!
P  r  = Cnr pr  1 - p  pr  1 - p 
n-r n-r
=
r!  n - r  !
p = 0.1
p = 0.3
p = 0.5
P(r)
r
The binomial mean = μ = np
The binomial sigma = σ = np  1 - p 
CSSBB 2014

Poisson Distribution
The Poisson distribution can be used to model discrete
data. The Poisson distribution can be an approximation
to the binomial when p is equal to or less than 0.1, and
the sample size n is fairly large (generally, n > 16) by
using np as the mean of the Poisson distribution.
 np  e -np
r
μr e -μ
P r  = =
r! r!
p = 0.1 p = 0.3
p = 0.5
P(r)
μ = np = average = the population mean

The Poisson average = μ = np = c*
The Poisson standard deviation = μ = np = c
CSSBB 2014

Chi Square Distribution

The chi square distribution is formed by summing the
squares of standard normal random variables. For
example, if z is a standard normal random variable, then:
y  z12  z22  z23  ...  zn2
The chi square probability density function is:
x   /2 - 1 e - x /2
f(x) =  /2 ,x  0
2   / 2
where ν is the degrees of freedom, and Γ(x) is the

gamma function. The chi square probability density
function is shown below.
0.40
0.30
= 2
Probability Density
0.20 = 1
=5
0.10  = 10
0.00
0 5 10 15 20
X
CSSBB 2014

F Distribution
If X is a chi square random variable with ν1 degrees of
freedom, and Y is a chi square random variable with ν2
degrees of freedom, and if X and Y are independent,
then:
X / 1
F
Y / 2
is an F distribution with ν1 and ν2 degrees of freedom.
The F distribution is used extensively to test for equality

of variances from two normal populations.
The F probability density function is:
  ν  ν   ν  ν1 /2   
 Γ    
1 2 1
ν1 /2  1
  2 ν
 2   x 
f(x)   ,x  0
 ν1   ν 2      ν1  ν2  /2 
 Γ   Γ     1  ν1x  
  2   2    ν2   
   
CSSBB 2014

F Distribution (Continued)
The F probability density function is shown below
1.00 v1=1, v2=1
v1=1, v2=10
0.80
Probability Density
v1=15, v2=15
0.60
0.40
0.20
v1=10, v2=1
0.00
0 0.5 1 1.5 2 2.5 3
X
The F distribution function is given in the Appendix.
Most texts only give one tail, and require the other tail to
be computed using the expression:
1
F , n1, n2 =
F1- , n2, n1
Given that with ν1 = 8 and ν2 = 10, F0.05 is 3.07, find the

value of F0.95 with ν1 = 10 and ν2 = 8.
1 1
F0.95,10,8 = = = 0.326
F0.05,8,10 3.07
CSSBB 2014

Student’s t Distribution
If z is a standard normal random variable, and O2 is a chi
square random variable with ν degrees of freedom, then
a random variable with a t distribution is:
z
t=
2

The t distribution is used in place of the normal

distribution when the standard deviation is unknown.
The t distribution compensates for the error in the
estimated standard deviation. If the sample size is large,
n>100, the error in the estimated standard deviation is
small, and the t distribution is approximately normal.
The t probability density function is:
  + 1 -   + 1
 
 2  x  2
2
f x = 1 +  for -  < x < 
  
  
2
CSSBB 2014

Student’s t Distribution (Continued)

The t probability density function is shown below.
0.40
Standard =10
Normal
Probability Density
0.30
=3
0.20
=1
0.10
0.00
-4 -2 0 2 4
X
The mean and variance of the t distribution are:

=0 2 = ,  3
-2
From a random sample of n items, the probability that:
X-
t=
s/ n
falls between any two specified values is equal to the

area under the t probability density function.
CSSBB 2014

PROBABILITY/OTHER DISTRIBUTIONS
Other Distributions
Other less commonly used distributions include the

following:
C Hypergeometric
C Bivariate
C Exponential
C Weibull
C Lognormal
CSSBB 2014

Hypergeometric Distribution
The hypergeometric distribution is used to model
discrete data. The hypergeometric distribution applies
when the population size, N, is small compared to the
sample size (n >0.1N). Sampling is done without
replacement. The hypergeometric distribution is a
complex combination calculation and is used when the
defined occurrences are known or can be calculated.
The number of successes, r, in the sample follows the

hypergeometric function:
Crd CnN -- rd
P r  =
CNn
The hypergeometric distribution is similar to the

binomial distribution. The binomial distribution requires
the probability of success to be the same for all trials,
while the hypergeometric distribution does not.
CSSBB 2014

Hypergeometric Distribution (Continued)

From a group of 20 products, 10 are selected at random
for testing. What is the probability that the 10 selected
contain the 5 best units?
N = 20, n = 10, d = 5, (N-d) = 15 and r = 5
C55 C15  n! 
P r  = 5
 note that C n
= 
C1020

r
r!  n - r  ! 
 5!   15! 
  
 5!10!   5!10!   15!   10!10! 
P r  = =    = 0.0163 = 1.63%
 20!   5!10!   20! 
 
 10!10! 
The mean and the variance of the hypergeometric

distribution are:
nm  nm   m  N - n 
= 2 =   1 -  
N  N  N  N - 1
Note that m = d in this version of the equation.

CSSBB 2014

Choosing A Discrete Distribution
Start
Modeling a rate with Yes

no upper bound for the Poisson
number of successes?
No
Yes Probability of success Yes

Binomial Fixed number of trials?
same on all trials?
No
No
Hypergeometric
Probability of success
No Yes
Negative the same on all trials
Geometric
Binomial and number of
successes = 1?
CSSBB 2014

Bivariate Normal Distribution

The joint distribution of two variables is called a
bivariate distribution. Bivariate distributions may be
discrete or continuous.
The graphical representation of a bivariate distribution

is a three dimensional plot, with the x and y-axis
representing the independent variables and the z-axis
representing the frequency for discrete data or the
probability for continuous data.
A special case of the bivariate distribution is the

bivariate normal distribution shown below:
CSSBB 2014

Exponential Distribution
The exponential distribution is used to model items with
a constant failure rate. The exponential distribution is
closely related to the Poisson distribution. The
exponential probability density function is:
1 - x
f  x  = e =  e - x , x  0

Where: λ is the failure rate and θ is the mean. From the

equation above, it can be seen that λ = 1/θ. The
exponential density function is shown below.
Probability Density
The variance of the exponential distribution is equal to

the mean squared.
1 1
 2 = 2 = hence  =  =
2 
The exponential distribution is characterized by its

hazard function which is constant. Because of this, the
exponential distribution exhibits a lack of memory.
CSSBB 2014

Lognormal Distribution
If a data set is known to follow a lognormal distribution,
transforming the data by taking a logarithm yields a data
set that is approximately normally distributed.
The most common transformation is made by taking the

natural logarithm, but any base logarithm also yields an
approximate normal distribution.
When random variables are summed, as the sample size

increases, the distribution of the sum becomes a normal
distribution, regardless of the distribution of the
individuals. Since lognormal random variables are
transformed to normal random variables by taking the
logarithm, when random variables are multiplied, as the
sample size increases, the distribution of the product
becomes a lognormal distribution regardless of the
distribution of the individuals.
CSSBB 2014

Lognormal Distribution (Continued)

The standard lognormal probability density function is:
2
1  ln x -  
1 -  
f x = e 2  
,x>0
x  2
The lognormal distribution takes on several shapes

depending on the value of the shape parameter. The
lognormal distribution is skewed right, and the
skewness increases as the value of σ increases. This is
shown below.
=2  = 0.25
Probability Density
=1
 = 0.5
X
CSSBB 2014

Weibull Distribution
The Weibull distribution is one of the most widely used
distributions in reliability and statistical applications. It
is commonly used to model time to fail, time to repair,
and material strength. There are two common versions
of the Weibull distribution, the two parameter Weibull
and the three parameter Weibull. The difference is the
three parameter Weibull distribution has a location
parameter when there is some non-zero time to first
failure.
The three parameter Weibull probability density function

is:
 -1 
x -  x - 
f x =   exp -   , for x  
     
β is the shape parameter

θ is the scale parameter
δ is the location parameter
CSSBB 2014

Weibull Distribution (Continued)

The shape parameter is what gives the Weibull
distribution its flexibility. By changing the value of the
shape parameter, the Weibull distribution can model a
wide variety of data. If β = 1 the Weibull distribution is
identical to the exponential distribution, if β = 2, the
Weibull distribution is identical to the Rayleigh
distribution; if β is between 3 and 4, the Weibull
distribution approximates the normal distribution.
The Weibull distribution approximates the lognormal

distribution for several values of β. The effect of the
shape parameter is shown below.
0.025
=6
0.020
Probability Density
 = 0.8
0.015  = 3.6
=2
0.010
=1
0.005
0.000 0 50 100 150 200

X
CSSBB 2014


The scale parameter determines the range of the
distribution as shown below.
=1
0.020  = 50
 = 2.5
 = 50
Probability Density
0.015
0.010  = 2.5
 = 100
0.005
=1
 = 100
0.000 0 50 100 150 200

X
The location parameter is used to define a failure-free

zone. Generally, the location parameter, δ, is assumed
to be zero. The effect of the location parameter is
shown below.
=0
 = 30
0.015
Probability Density
0.010
0.005
0.000 0 50 100 150

X
CSSBB 2014


The mean and variance of the Weibull distribution are
computed using the gamma distribution. The mean of
the Weibull distribution is equal to the characteristic life
if the shape parameter is equal to one.
The mean of the Weibull distribution is:

 1
 =   1 + 
 
The variance of the Weibull distribution is:
  2  1 
 2 = 2    1 +  -  2  1 +  
     
The variance of the Weibull distribution decreases as

the value of the shape parameter increases.
CSSBB 2014
VII. MEASURE - STATISTICS V.F.3

PROCESS CAPABILITY/CAPABILITY STUDIES
Process Capability
Process Capability is presented in the following

topic areas:
C Capability studies
C Capability indices
C Performance indices
C Short-term vs. long-term
C Non-normal data
C Attributes data
C Performance metrics
CSSBB 2014

Process Capability Studies

The determination of process capability requires a
predictable pattern of statistically stable behavior (most
frequently a bell-shaped curve) where the chance
causes of variation are compared to the engineering
specifications. A capable process is a process whose
spread on the bell-shaped curve is narrower than the
tolerance range or specification limits.
LSL USL
MIN MAX
It is often necessary to compare the process variation

with the engineering or specification tolerances to judge
the suitability of the process. A process capability
study includes three steps:
C Planning for data collection

C Collecting data
C Plotting and analyzing the results
CSSBB 2014

Process Capability Studies (Continued)

The objective of process quality control is to establish
a state of control over the process and then maintain
that state of control through time. When the natural
process limits are compared with the specification
range, any of the following possible courses of action
may result:
C Do nothing
C Change the specifications
C Center the process
C Reduce variability
C Accept the losses

CSSBB 2014

Process Capability Studies (Continued)

Other capability applications:
C Setting up a variables control chart

C Evaluating new equipment
C Reviewing tolerances based on process variability
C Assigning more capable equipment to tougher jobs
C Performing routine process performance audits
C Determining the effects of adjustments
CSSBB 2014

Identifying Characteristics
The identification of characteristics to be measured in a
process capability study should meet the following
requirements:
C The characteristic should be indicative of a key

factor in the quality of the product or process
C It should be possible to adjust the value of the

characteristic
C The operating conditions that affect the measured

characteristic should be defined and controlled
Selecting one, or possibly two, key dimensions provides

a manageable method of evaluating the process
capability. The characteristic selected may also be
determined by the history of the part and the parameter
that has been the most difficult to control.
Customer purchase order requirements or industry

standards may also determine the characteristics that
are required to be measured.
CSSBB 2014

Identifying Specifications/Tolerances
The process specifications or tolerances are determined
either by customer requirements, industry standards, or
the organization’s engineering department.
Developing Sampling Plans

The appropriate sampling plan for conducting process
capability studies depends upon the purpose and
whether there are customer or standards requirements
for the study.
If the process is currently running and is in control,

control chart data may be used to calculate the process
capability indices. If the process fits a normal
distribution and is in statistical control, then the
standard deviation can be estimated from:
R
R 
d2
For new processes a pilot run may be used to estimate

the process capability. A design of experiments can be
used to determine the optimum values of the process
variables which yield the lowest process variation.
CSSBB 2014

Verifying Stability and Normality

If only common causes of variation are present in a
process, then the output of the process forms a
distribution that is stable over time and is predictable.
If special causes of variation are present, the process
output is not stable over time.
The Figure below depicts an unstable process with both

process average and variation out-of-control. The
process may also be unstable if either the process
average or variation is out-of-control.
CSSBB 2014

Verifying Stability and Normality (Cont’d)

The validity of the normality assumption may be tested
using the chi square hypothesis test. To perform this
test, the data is partitioned into data ranges. The
number of data points in each range is then compared
with the number predicted from a normal distribution.
Continuous data may be tested using a variety of

goodness-of-fit tests.
CSSBB 2014

The Normal Distribution

When all special causes of variation are eliminated,
many variable data processes, when sampled and
plotted, produce a bell-shaped distribution. If the base
of the histogram is divided into six (6) equal lengths
(three on each side of the average), the amount of data
in each interval exhibits the following percentages:
68.27%
95.45%
:-3F :-2F :-1F : :+1F :+2F :+3F
99.73%
CSSBB 2014

The Z Value
The area outside of specification for a normal curve can
be determined by a Z value.
X - LSL USL - X
ZLOWER  ZUPPER 
S S
The Z transformation formula is:
X-
Z=

This transformation will convert the original values to

the number of standard deviations away from the mean.
The result allows one to use a single standard normal
table to describe areas under the curve (probability of
occurrence).
CSSBB 2014

Z Value (Continued)
There are several ways to display the normal
(standardized) distribution:
1. As a number under the curve, up to the Z value.
0 1.0
P(Z = -  to 1) = 0.8413
2. As a number beyond the Z value.
0 1.0
P(Z =1 to + ) = 0.1587
CSSBB 2014

Z Value (Continued)
3. As a number under the curve, and at a distance from
the mean.
P(Z = 0 to 1) = 0.3413
The standard normal table in this Primer uses the

second method of calculating the probability of
occurrence.
CSSBB 2014

Z Value Example
Tenth grade students weights follow a normal
distribution with a mean μ = 150 lb and a standard
deviation of 20 lb. What is the probability of a student
weighing less than 100 lb?
X- μ
μ = 150 Z=
σ
x = 100 100 - 150 50
Z= =- = -2.5
σ = 20 20 20
Since the normal table has values about the mean, a Z

value of - 2.5 can be treated as 2.5.
P(Z = -  to -2.5) = 0.0062. That is, 0.62% of the students

will weigh less than 100 lb.
CSSBB 2014

PROCESS CAPABILITY / CAPABILITY INDICES
Capability Index Failure Rates

There is a direct link between the calculated Cp (and Pp
values) with the standard normal (Z value) table. A Cp of
1.0 is the loss suffered at a Z value of 3.0 (doubled, since
the table is one sided). Refer to the Table below.
Z
Cp ppm
value
0.33 1.00 317,311
0.67 2.00 45,500
1.00 3.00 2,700
1.10 3.30 967
1.20 3.60 318
1.30 3.90 96
1.33 4.00 63
1.40 4.20 27
1.50 4.50 6.8
1.60 4.80 1.6
1.67 5.00 0.57
1.80 5.40 0.067
2.00 6.00 0.002
CSSBB 2014

Index Failure Rates (Continued)

In the prior Table, ppm equals parts per million of
nonconformance (or failure) when the process:
C Is centered on X
C Has a two-tailed specification
C Is normally distributed
C Has no significant shifts in average or dispersion
When the Cp, Cpk, Pp, and Ppk values are 1.0 or less, Z
values and the standard normal table can be used to
determine failure rates. With the drive for increasingly
dependable products, there is a need for failure rates in
the Cp range of 1.5 to 2.0.
CSSBB 2014

Process Capability Indices

To determine process capability, an estimation of sigma
is necessary:
R
R =
d2
σR is an estimate of process capability sigma and comes

from a control chart.
The capability index is defined as:
 USL - LSL 
CP =
6R
As a rule of thumb:
CP > 1.33 Capable

CP = 1.00 to 1.33 Capable with tight control
CP < 1.00 Incapable
CSSBB 2014

Process Capability Indices (Continued)

The capability ratio is defined as:
6 R
CR =
 USL - LSL 
As a rule of thumb:
CR < 0.75 Capable

CR = 0.75 to 1.00 Capable with tight control
CR > 1.00 Incapable
Note, this rule of thumb logic is somewhat out of step

with the six sigma assumption of a ±1.5 sigma shift. The
above formulas only apply if the process is centered,
stays centered within the specifications, and CP = Cpk.
Cpk is the ratio giving the smallest answer between:
USL - X X - LSL
Cpk = or
3 R 3 R
CSSBB 2014
VII. MEASURE - STATISTICS V.F.1, 2

Process Capability Exercise

For a process with X = 12, σR = 2 an USL = 16 and LSL
= 4, determine Cp and Cpk min:
USL - LSL 16 - 4 12
Cp = = = =1
6σR 6  2 12
USL - X 16 - 12 4
Cpk upper = = = = 0.667
3σR 3 2 6
X - LSL 12 - 4 8
Cpk lower = = = = 1.333
3σR 3  2 6
Cpk min = Cpk upper = 0.667
CSSBB 2014
VII. MEASURE - STATISTICS V.F.1, 2

Cpm Index
The Cpm index is defined as:
USL - LSL
Cpm =
 - T
2
6 + 2
Where: USL = upper specification limit

LSL = lower specification limit
μ = process mean
T = target value
σ = process standard deviation
Cpm is based on the Taguchi index, which places more

emphasis on process centering on the target.
CSSBB 2014

PROCESS CAPABILITY / PERFORMANCE INDICES
Process Performance Indices

To determine process performance, an estimation of
sigma is necessary:
  X - X
2
i =
 n - 1
σi is a measure of total data sigma and generally comes

from a calculator or computer.
The performance index is defined as:
 USL - LSL 
PP =
6i
The performance ratio is defined as:
6 i
PR =
 USL - LSL 
Ppk is the ratio giving the smallest answer between:
Ppk =
 USL - X  or
 X - LSL 
3 i 3 i
CSSBB 2014

PROCESS CAPABILITY/SHORT & LONG TERM CAPABILITY
Short-Term and Long-Term Capability

When a process capability is determined using one
operator on one shift, with one piece of equipment the
process variation is relatively small. Control limits
based on a short-term process evaluation are closer
together than control limits based on the long-term
process.
A modified X̄ and R chart can be used for short runs,

based on an initial 3 to 10 pieces, using a calculated
value is compared with a critical value. Inflated D4 and
A2 values are used to establish control limits. Control
limits are recalculated after additional groups are run.
The X and MR chart can be used for small runs, with a

limited amount of data. The X represents individual data
values and the MR is the moving range, a measure of
piece-to-piece variability.
Process capability or Cpk values determined from either

of these methods must be considered preliminary
information. As the number of data points increases,
the calculated process capability will approach the true
capability.
CSSBB 2014

PROCESS CAPABILITY/NON-NORMAL DATA
Process Capability for Non-Normal Data

Data does not always fit a normal distribution. One
strategy is to make non-normal data resemble normal
data by using a transformation.
A family of power transformations for positive data

values are attributed to G.E.P Box and D.R. Cox. The
Box-Cox power transformations are given by:
x  =
 x - 1 
for   0

x    = ln  x  for  = 0
Given data observations x1, x2, ...xn, select the power λ
that maximizes the logarithm of the likelihood function:

  
2
n  n xi    -x    n
 +   -1 ln  x 
f  x,   = - ln  
2  i=1  i
n i=1
 
Where the arithmetic mean of the transformed data is:
1 n
x    =  xi   
ni=1
CSSBB 2014

Capability for Non-normal Data (Cont’d)

Assume that a process capability study has been
conducted. Some 30 data points from a non-normal
distribution are shown below. An investigator can
check the data for normality using techniques such as
the dot plot, histogram, and normal probability plot.
1.46 19.45 23.43 92.35 104.86
118.59 282.58 311.17 341.88 374.81
410.06 676.52 731.16 789.05 850.31
915.06 983.45 1055.60 1131.65 1384.58
1477.63 1575.30 1677.72 1785.06 2541.17
2687.39 2839.82 4304.67 4521.22 6857.50
A histogram displaying the above non-normal data

indicates a distribution that is skewed to the right.
10
Frequency
0
0 1000 2000 3000 4000 5000 6000 7000
Data
CSSBB 2014


A probability plot can also be used to display the non-
normal data. Since this is a non-normal distribution, a
traditional process capability index is meaningless.
Refer to the Figure below.
.999
.99
.95
P ro b a b il it y
.80
.50
.20
.05
.01
.001
0 1000 2000 3000 4000 5000 6000 7000
If the data history is known to follow a Poisson

distribution, then a square root transformation is a
possibility. Some typical data transformations include:
C Log transformation (log x)

C Square root or power transformation (x y)
C Exponential (e x)
C Reciprocal (1/x)
CSSBB 2014


In order to find the right transformation, some
exploratory data analysis may be required. Among the
useful power transformation techniques is the Box-Cox
procedure.
One can also use Excel or Minitab to handle the data

calculations and to draw the normal probability plot. In
this example, a power transform of 0.337 is indicated.
CSSBB 2014


A probability plot of the newly transformed data will
show a near normal distribution. See the Figure below.
.999
.99
.95
Probability
.80
.50
.20
.05
.01
.001
5 10 15 20
Average: 9.72959 Anderson-Darling Normality Test

StDev: 4.43272 A-Squared: 0.143
N: 30 P-value: 0.967
Now, a process capability index can be determined for

the data. However, the investigator must remember to
also transform the specifications.
CSSBB 2014

PROCESS CAPABILITY/ATTRIBUTE DATA
Process Capability for Attribute Data

The control chart represents the process capability,
once special causes have been identified and removed
from the process. For attribute charts, capability is
defined as the average proportion or rate of
nonconforming product.
C For p charts, the process capability is the process

average nonconforming, p The proportion
conforming to specification, 1 - p , may be used.
C For np charts, the process capability is the process

average nonconforming, p .
C For c charts, the process capability is the average

number of nonconformities, c .
C For u charts, the process capability is the average

number of nonconformities per reporting unit, u .
The average proportion of nonconformities may be

reported on a defects per million opportunities scale by
multiplying p times 1,000,000.
CSSBB 2014

PROCESS CAPABILITY/PERFORMANCE METRICS
Process Performance Metrics

There are a large number of six sigma measurements.
The authors of this Primer have presented only those
that are widely used:
Widely Used Symbols
C Defects = D C Opportunities (for a defect) = O

C Units = U C Yield = Y
Defect Relationships
C Total opportunities: TO = TOP = U x O
D
C Defects per unit: DPU = also = -ln(Y) See yield
U
C Defects per normalized unit: = -ln(Ynorm ) See yield
DPO D
C Defects per unit opportunity: DPO = =
O UxO
C Defects per million opportunities: DPMO = DPO x 106

CSSBB 2014

Process Performance Metrics (Cont’d)

A matrix chart indicates the following information for
100 production units. Determine DPU:
Defects/unit 0 1 2 3 4 5
Number of units 70 20 5 4 0 1
D 1 20  + 2  5  + 3  4  + 5  1 47
DPU = = = = 0.47
U 100 100
One would expect to find an average of 0.47 defects per

unit.
Assume that each unit in the previous example had 6

opportunities for a defect (i.e. characteristics A, B, C, D,
E, and F). Determine DPO and DPMO.
DPU 0.47
DPO = = = 0.078333
O 6
DPMO = DPO x 106 = 78,333

CSSBB 2014


Yield Relationships
The Poisson equation is normally used to model defect

occurrences. If there is a historic defect per unit (DPU)
level for a process, the probability that an item contains
X flaws (Px) is described mathematically by the equation:
e-DPU DPUx
P x  =
X!
Where: X is an integer greater or equal to 0

DPU is greater than 0
Note that 0! (zero factorial) = 1 by definition.
If one is interested in the probability of having a defect

free unit (as most of us are), then X = 0 in the Poisson
formula and the math is simplified:
P 0  = e-DPU
CSSBB 2014


The following common yield formulas follow:
Yield or first pass yield: Y = FPY = e -DPU
Defects per unit: DPU = - ln (Y)

n
Rolled throughput yield: Yrt = RTY = 
i=1
Yi
Normalized yield: Ynorm = n

RTY  Where n = # steps 
Total defects per unit: TDPU = -ln (Yrt)
Assume that a process has a DPU of 0.47. Determine

the yield.
- DPU
Y=e = e- 0.47 = 0.625 = 62.5%
For a process with a first pass yield of 0.625 determine

the DPU.
DPU = - ln  Y  = - ln 0.625 = 0.47
CSSBB 2014


Yield Relationships (Continued)
A process consists of 4 sequential steps: 1, 2, 3, and 4.

The yield of each step is as follows: Y1 = 99%, Y2 =98%,
Y3 = 97%, Y4 = 96%. Determine the rolled throughput
yield and the total defects per unit.
n
Yrt =  Y =  0.99  0.98  0.97  0.96  = 0.90345 = 90.345%
i=1
i
TDPU = - ln  RTY  = - ln 0.90345 = 0.1015

CSSBB 2014

Rolled Throughput Yield (RTY)

RTY is the cumulative calculation of yield or defects
through multiple processes. Use the following steps.
C Calculate the yield for each step and the RTY

C The RTY for a process will be the baseline metric
C Revisit the project scope
C Yield differences suggest improvement areas
A RTY calculation and analysis for 5 process steps.
Weld 1 Weld 2 Fab 1 Fab 2 Assembly
yield: yield: 92% yield:

yield: 90% yield: 65%
86% 87%
RTY = 0.90 x 0.86 x 0.92 x 0.87 x 0.65 = 0.403
The process RTY is only 40.3%. A very significant drop

in yield occurs in the assembly process with only 65%
yield. This indicates that the assembly process could
warrant an initial improvement project.
CSSBB 2014

Sigma Relationships
Probability of a defect = P(d)
P(d) = 1-Y or 1 - FPY
also P(d) = 1 - YRT (for a series of operations)
P(d) can be looked up in a Z table (using the table in

reverse to determine Z).
The first pass yield for a single operation is 95%. What

is the probability of a defect and what is the Z value?
P(d) = 1 - 0.95 = 0.05
Using the Z table for 0.05 approximates 1.645 sigma.
The Z value in the above example is called Z long-term

or Z equivalent.
CSSBB 2014

Sigma Relationships (Continued)

Z short-term is defined as: ZST = ZLT + 1.5 shift
If Z long-term = 1.645, what is Z short-term?
ZST = ZLT + 1.5 = 1.645 + 1.5 = 3.145
The 6 sigma quality level (with the 1.5 sigma shift) can
be approximated by:
6 Sigma Quaity Level = 0.8406 + 29.37 - 2.221 x ln  ppm 
If a process were producing 80 defectives/million, what

would be the 6 sigma quality level?
6σ = 0.8406 + 29.37 + 2.221 x ln  80 

= 0.8406 + 29.37 + 2.221  4.3820 
= 0.8406 + 4.4314 = 5.272  about 5.3 
This answer can be also looked up in Appendix Table II,

which (by interpolation) appears to be about 5.3.
CSSBB 2014

QUESTIONS
7.2. Which of the following distributions requires the use of the natural
logarithmic base for probability calculations?
a. Hypergeometric
b. Student’s t
c. Binomial
d. Weibull
7.7. The reported Cpk, for a process with an average of 28, a spread of
10 units, and upper and lower specification limits of 35 and 15
respectively, would be:
a. 1.6
b. 1.4
c. 1.8
d. 0.714
7.10. A process consists of three sequential steps with the following

yields:
Y1 = 99.8, Y2 = 97.4, Y3 = 96.4
Determine the total defects per unit.
a. 0.063
b. 0.065
c. 0.067
d. 0.069
Answers: 7.2. d, 7.7. b, 7.10. b

CSSBB 2014

QUESTIONS
7.12. The major difference between process capability indices and

process performance indices is:
a. The specification spread must be known for process capability

indices
b. How the standard deviation term is determined for the two indices
c. The target value must be known for process performance indices
d. The required failure rate limit must be known for process capability
indices
7.16. If we were to compare short-term capability with long-term

capability for the same process, one should not be surprised to
find:
a. The long-term capability improves

b. The Cp is better short-term
c. The results are very comparable
d. The average drifts but the variation stays
7.17. Determine the coefficient of variation for the last 500 pilot plant
test runs of high temperature film having a mean of 900° Kelvin
with a standard deviation of 54°.
a. 6%
b. 16.7%
c. 0.6%
d. 31%
Answers: 7.12. b, 7.16. b, 7.17. a

CSSBB 2014

QUESTIONS
7.19. One thousand units of product were examined to detect 5 possible

undesirable characteristics. A total of 80 defects were found. How
many defects would be expected in a million opportunities?
a. 16,000
b. 26,666
c. 61,458
d. 80,000
7.27. A capability study, conducted during a pilot run of 100 units,

indicated that the Cpk upper value to be 1.8 while the Cpk lower
value was 0.90. The customer requires a Cpk minimum value of
1.25. What action should be taken?
a. Center the process

b. Reduce variability
C. Renegotiate the customer specification
d. 100% inspect until new equipment is available
7.28. For two events, A and B, which one of the following is a true
probability statement?
a. P(A or B) = P(A) + P(B) if A and B are independent

b. P(A or B) = P(A) + P(B) if A and B are mutually exclusive
c. P(A and B) = P(A) x P(B) if A and B are mutually exclusive
d. P(A or B) = P(A) x P(B) if A and B are independent
Answers: 7.19. a, 7.27. a, 7.28. b

CSSBB 2014

QUESTIONS
7.29. What graphical data method can show the value of all individual
readings?
a. A stem and leaf plot

b. A grouped probability density function
c. A typical histogram
d. A complex boxplot
7.31. There are five fabricating operations that can be performed in any
sequence. Four of the operations yield 98% recovery and one
yields 80% recovery. What is the preferred production sequence,
assuming that the 80% operation cannot be immediately improved?
a. Place the 80% operation first

b. Place the 80% operation in the middle
c. Place the 80% operation last
d. It doesn’t matter; the RTY is the same
7.36. The probability of a train arriving on time and leaving on time is

0.8. The probability of the same train arriving on time is 0.84. The
probability of this train leaving on time is 0.86. Given the train
arrived on time, what is the probability it will leave on time?
a. 0.93
b. 0.84
c. 0.88
d. 0.95
Answers: 7.29. a, 7.31. a, 7.36. d

© QUALITY COUNCIL OF INDIANA VIII-1 (672)
CSSBB 2014
VIII. ANALYZE
MOST PEOPLE WOULD RATHER LIVE

WITH A PROBLEM THEY CAN'T
SOLVE, THAN ACCEPT A SOLUTION
THEY CAN'T UNDERSTAND.
Woolsey and Swanson

CSSBB 2014
VIII. ANALYZE VI.A.2

MEASURING RELATIONSHIPS/REGRESSION
Analyze
Analyze is presented in the following topic areas:
C Measuring and modeling relationships

C Hypothesis testing
C Failure mode and effects analysis (FMEA)
C Additional analysis methods
Measuring and Modeling Relationships
Measuring and modeling relationships between

variables is reviewed in the following topic areas:
C Regression
C Correlation coefficient
C Multivariate tools
Note that the authors have presented regression ahead

of the correlation coefficient.
CSSBB 2014

Regression
Simple linear regression and multiple linear regression
will be discussed here. Note that non-linear regression
models will not be tested on the CSSBB exam, and will
not be described here.
Simple Linear Regression

Consider the problem of predicting the test results (y)
for students based upon an input variable (x), the
amount of preparation time using the data presented
below.
Student Study Time Test Results (%)

1 60 hrs 67%
2 40 hrs 61%
3 50 hrs 73%
4 65 hrs 80%
5 35 hrs 60%
6 40 hrs 55%
7 50 hrs 62%
8 30 hrs 50%
9 45 hrs 61%
10 55 hrs 70%
CSSBB 2014

Simple Linear Regression Model (Cont’d)

An initial approach to the analysis of the data is to plot
the points on a graph known as a scatter diagram, as
shown below. Observe that y appears to increase as x
increases. One method of obtaining a prediction
equation relating y to x is to place a ruler on the graph
and move it about until it seems to pass through the
majority of the points, thus providing what is regarded
as the “best fit” line.
81
74
67
60
53
30 35 40 45 50 55 60 65
Study Time (Hours), X
CSSBB 2014

Simple Linear Regression Model (Cont’d)

The mathematical equation of a straight line is:
y = β0 + β1x
Where β0 is the y intercept when x = 0 and β1 is the slope
of the line. There will also be a random error which is
the difference between an observed value of y and the
mean value of y for a given value of x. One assumes a
normal probability distribution. The concept is
illustrated below:
x1 x2 x3
The model for an observed value of y is:
 Mean value of y for 
y=   +  random error 
 a given value of x 
y =  0 + 1x + 
CSSBB 2014

The Method of Least Squares

The statistical procedure of finding the “best-fitting”
straight line is, in many respects, a formalization of the
procedure used when one fits a line by eye. The
objective is to minimize the deviations of the points from
the prospective line. If one denotes the predicted value
of y obtained from the fitted line as y  the prediction
equation is:
 = β  β x
y 0 1
Where: β and β represent estimates of the true β and

0 1 0
β1, as shown below.
81
Test Results, Y
74
 = β  β x
y
67 0 1
60
53
30 35 40 45 50 55 60 65
Study Time (Hours), X
CSSBB 2014

Method of Least Squares (Continued)

The best fit criterion of goodness known as the principle
of least squares is employed:
Choose, as the best fitting line, the line that

minimizes the sum of squares of the deviations of
the observed values of y from those predicted.
Expressed mathematically, minimize the sum of squared

errors given by:
 
n 2
SSE = 
yi - y i
i=1
Substituting for y^i one obtains the following expression:
 
n 2
Sum of squared errors = SSE =   yi -  0 +  1xi 
i=1
 
CSSBB 2014

Method of Least Squares (Continued)

The least square estimator of β0 and β1 are calculated as
follows:
2
 n 
n   xi 
S x2 =  xi2 -  i = 1 
i=1 n
 n  n 
n   xi    yi 
Sxy =  xi yi -  i = 1   i = 1 
i=1 n
Sxy
 1 = and  0 = y -  1 x
S x2
Once  and  have been computed, substitute their

0 1
values into the equation of a line to obtain the least
squares prediction equation, or regression line.
 is:
As noted earlier, the prediction equation for y
ŷ = ˆ 0 + ˆ 1 X
 and  represent estimates of the true 
Where:  0 1 0
 0 and  1 .
CSSBB 2014

Least Squares Example

Obtain the least squares line for the data below:
2 2
i xi yi xi xiyi yi
1 60 67 3,600 4,020 4,489
2 40 61 1,600 2,440 3,721
3 50 73 2,500 3,650 5,329
4 65 80 4,225 5,200 6,400
5 35 60 1,225 2,100 3,600
6 40 55 1,600 2,200 3,025
7 50 62 2,500 3,100 3,844
8 30 50 900 1,500 2,500
9 45 61 2,025 2,745 3,721
10 55 70 3,025 3,850 4,900
Sum 470 639 23,200 30,805 41,529
2
 n 
n   xi   470 
2
S x2 =  xi -
2  i=1  = 23,200 - = 1,110
i=1 n 10
 n  n 
n   xi    yi   470  639 
Sxy =  xi yi -  i = 1   i = 1  = 30,805 - = 772
i=1 n 10
470 639
x= = 47 y= = 63.9
10 10
CSSBB 2014

Least Squares Example (Continued)

Example continued:
S xy 772
 1 = = = 0.6955
S x2 1,110
 0 = y -  1 x = 63.9 - (0.6955)(47) = 31.2115
ŷ = 31.2115 + 0.6955 x
One may predict y for a given value of x by substitution

into the prediction equation. For example, if 60 hours of
study time is allocated, the predicted test score would
be:
ŷ = 31.2115 +  0.6955  60 
ŷ = 72.9415 = 73%
CSSBB 2014

Hints on Regression Analysis

C Be careful of rounding errors. Normally, the
calculations should carry a minimum of six
significant figures in computing sums of squares of
deviations. The prior example consisted of
convenient whole numbers which does not occur
often.
C Always plot the data points and graph the least

squares line. If the line does not provide a
reasonable fit to the data points, there may be a
calculation error.
C Projecting a regression line outside of the test area

can be risky. The above equation suggests, without
study, a student would make 31% on the test. The
odds favor 25% if answer a is selected for all
questions. The equation also suggests that with
100 hours of study the student should attain 100%
on the examination - which is highly unlikely.
CSSBB 2014

Calculating S2e , an Estimator of  2

The first step toward acquiring a boundary on a
prediction error requires that one estimates  2 . It is
reasonable to use SSE (sum of squares for error) based
on (n - 2) degrees of freedom, one for each variable (x
and y).
SSE
ˆ 2 = ˆ 2 is sometimes shown as s2e
n-2

CSSBB 2014

The Slope β1 of a Line

The Primer shows examples of how to draw inferences
and determine confidence interval estimates for the
slope β1 of a line.
CSSBB 2014

Multiple Linear Regression

Multiple linear regression is an extension of the
methodology for linear regression to more than one
independent variable. By including more than one
independent variable, a higher proportion of the
variation in y may be explained. A few models and
definitions are presented in the Primer.
A First-Order Linear Model

y =  0 + 1x1 +  2 x 2 + ... + k xk + 
A Second-Order Linear Model
y =  0 + 1x1 +  2 x 2 +  3 x1x 2 +  4 x12 +  5 x 22 + 

SSE
s2 =
n - (k + 1)
Just like r2 (the linear coefficient of determination) R2

(the multiple coefficient of determination) take values in
the interval:
0 < R2 < 1
CSSBB 2014

MEASURING RELATIONSHIPS/CORRELATION COEFFICIENT
Correlation Coefficient (r)

The sample linear correlation coefficient, r, measures
the strength of the linear relationship between the paired
x and y values in a sample. r is a sample statistic. For
a sample, the Pearson product moment coefficient of
correlation, rx,y is given by:
n
Sxy x i - x  yi - y 
rx,y = = i=1
S x 2 S y2 n n
x - x y - y
2 2
i i
i=1 i=1
Note that -1 < r < +1
Using the study time and test score data reviewed

earlier, determine the correlation coefficient.
Given: Sxy = 772 S x2 = 1,110 S y2 = 696.9

Sxy 772
Solution: rx,y = = = 0.878
S x 2 S y2  1,110  696.9 
CSSBB 2014

Correlation Coefficient (Continued)

Note that the numerator used in calculating r is identical
to the numerator of the formula for the slope β1. Thus,
the coefficient of correlation r will assume exactly the
same sign as β1 and will equal zero when β1 = 0.
C A positive value for r implies that the line slopes

upward to the right.
C A negative value for r implies that the line slopes

downward to the right.
C Note that r = 0 implies no linear correlation, not

simply “no correlation.” A pronounced curvilinear
pattern may exist.
When r = 1 or r = -1, all points fall on a straight line;

when r = 0, they are scattered and give no evidence of a
linear relationship. Any other value of r suggests the
degree to which the points tend to be linearly related.
CSSBB 2014

Coefficient of Determination (R2)

The coefficient of determination is R2. The square of
the linear correlation coefficient is r2. It can be shown
that: R2 = r2
 Sxy 
2
S y2 - SSE SSE
R2 = r 2 = =1- =
S y2 S y2 S x 2 S y2
The coefficient of determination is the proportion of the

explained variation divided by the total variation, when
a linear regression is performed. r 2 lies in the interval of
0 < r2 < 1. r2 will equal +1 only when all the points fall
exactly on the fitted line. That is, when SSE equals zero.
Using the data from Example, determine the coefficient

of determination.
S 
2
2 xy  772 2
r = = = 0.771
S x2 S y2  1,110  696.9 
or r 2 =  0.878  = 0.771
2
One can say that 77% of the variation in test scores can
be explained by variation in study hours.
CSSBB 2014

Correlation Example
25
24
d1
23
d2
D1
22
d3 D4 d4
21 D2
MPG d5
D7
20 Average
20 MPG
D5
D3
19 D8 d7
D6 d6
18
d8
17 D9
16 d9
2000 3000 4000

Car Weight
Correlation Plot of Car Weight and MPG

n
SST = D
i=1
2
i = D12 + D22 + ... + D29
n
and SSE = d
i=1
2
i = d12 + d22 + ... + d29
SSE SST - SSE

r2 = 1 - or
SST SST
CSSBB 2014

Correlation Versus Causation

In the previous example, there is strong evidence of a
correlation between car weight and gas milage. The
student should be aware that a number of other factors
(carburetor type, car design, air conditioning, passenger
weights, speed, etc.) could also be important. The most
important cause may be a different or a collinear
variable. For example, car and passenger weight may
be collinear. There can also be such a thing as a
nonsensical correlation, i.e. it rains after my car is
washed.
CSSBB 2014

MEASURING RELATIONSHIPS/MULTIVARIATE TOOLS
Multivariate Tools
Univariate analysis was the subject matter of earlier
portions of this Section. In univariate statistics there are
one or more independent variables (X1, X2), and only
one dependent variable (Y). Multivariate analysis is
concerned with two or more dependent variables, Y1,
Y2, being simultaneously considered for multiple
independent variables, X1, X2, etc.
Recent advances in computer software and hardware

have made it possible to solve more problems using
multivariate analysis. This coverage of multivariate
analysis can only be considered an introduction to the
subject.
Multivariate analysis has found wide usage in the social

sciences, psychology, and educational fields. This
element will highlight the following multivariate
concepts or techniques:
C Factor analysis
C Discriminant function analysis
C Cluster analysis
C Canonical correlation analysis
C Multivariate analysis of variance
CSSBB 2014

Factor Analysis
Factor analysis is a data reduction technique to identify
factors that explain variation. It is very similar to the
principal components analysis technique. That is, factor
analysis attempts to simplify complex sets of data,
reducing many factors to a smaller set. However, there
is some subjective judgment involved in describing the
factors in this method of analysis. The output variables
are linearly related to the input factors.
The variables under investigation should be measurable,

have a range of measurements, and be symmetrically
distributed. There should be four or more input factors
for each dependent variable. Factor analysis undergoes
two stages: factor extraction and factor rotation. The
first analysis will distinguish the major factors for
further study (extraction). The second stage will rotate
the factors, to make them more meaningful.
There is a Primer example.

CSSBB 2014

Discriminant Analysis
If one has a sample with known groups, discriminant
analysis can be used to classify the observations or
attributes into two or more groups. Discriminant
analysis can be used as either a predictive or a
descriptive tool. The decisions could involve medical
care, college success attributes, car loan credit
worthiness, or economic development issues.
Discriminant analysis can be used as a follow-up to the

use of MANOVA (the last area of coverage in this
Section element). Linear combinations of predictors or
groups are provided by the researcher. The possible
number of linear combinations (discriminant functions)
for a study would be the smaller of the number of
groups -1, or the number of variables.
Some assumptions in discriminant analysis are: the

variables are multivariately normally distributed, the
population variances and covariances among the
dependent variables are the same, and the samples
within the variables are randomly obtained and exhibit
independence of scores from the other samples.
Minitab provides two forms of analysis: linear and
quadratic discriminant analysis. An example is
described in the Primer.
CSSBB 2014

Cluster Analysis
Cluster analysis is used to determine groupings or
classifications for a set of data. A variety of rules or
algorithms have been developed to assist in group
formations. The natural groupings should have
observations classified so that similar types are placed
together. A file on attributes of high achieving students
could be grouped or classified by IQ, parental support,
school system, study habits, and available resources.
Cluster analysis is used as a data reduction method in

an attempt to make sense of large amounts of data from
surveys, questionnaires, polls, scores, etc.
The economic development example in the previous

discussion is used in the Primer to validate groupings.
The two types of groups are new economy and not new
economy. A graphic output from the analysis is called
a classification tree or dendogram. It is a graphic line
graph linking variables and groups at various stages.
CSSBB 2014

Cluster Analysis Example

The dendogram below provides a visual confirmation
that San Jose is distinctive and of a higher ranking than
the other communities.
CSSBB 2014

Canonical Correlation Analysis

Canonical analysis tests the hypothesis that effects can
have multiple causes and causes can have multiple
effects. This technique was developed by Hotelling in
1935, but was not widely used for over 50 years. The
emergence of personal computers and statistical
software has led to its fairly recent adoption.
Canonical correlation analysis is a form of multiple

regression to find the correlation between two sets of
linear combinations. Each set may contain several
related variables. The relating of one set of independent
variables to one set of dependent variables will form
linear combinations. The largest correlation values for
sets are used in the analysis. The pairings of linear
combinations are called canonical variates, and the
correlations are called canonical correlations
The canonical correlation coefficient, rc, is similar to the

Pearson product-moment correlation coefficient. The
rule of thumb is to have values above 0.30. Linear
combinations can be determined from linear matrix
algebra or statistical software.
CSSBB 2014

Canonical Correlation Analysis (Cont’d)

A Primer Table illustrates the correlation of sets of
independent variables to sets of dependent variables.
An survey can be conducted to see if there is a
correlation between the characteristics of a quality
engineer to the listed job skills of a quality engineer.
There may be a set of variables that are strongly
correlated and canonical correlation can be used.
Hotelling’s T2 test is used on more than 2 variables at a

time. The student t test can also be used to compare 2
samples at a time, but if it is used to compare 5 samples,
2 at a time, the probability of obtaining a type one error
is increased. That is, finding a significant difference
when the two samples are the same. If a 5% error is
used, the probability of obtaining such an error is 1 -
0.95p. Where p is the number of samples. Hotelling’s T2
test is the and recommended test method.
CSSBB 2014

MANOVA (Multiple Analysis of Variance)

MANOVA tests whether mean differences among groups
of a combination of Ys are significant or not. The
concept of various treatment levels and associated
factors is still valid. The data should be normality
distributed, have homogeneity of the covariance
matrices, and have independence of observations.
In ANOVA, a sum of squares is used for the treatments

and for the error term. In MANOVA the terms become
matrices of the “sum of squares and cross-products”
SSPCP. ANOVAs used multiple times across the
dependent variables could result in inflated alpha errors.
The MANOVA method is used to reduce the alpha risk
by having only one test.
CSSBB 2014

MANOVA (Continued)
The table below describes ANOVA and MANOVA
differences.
ANOVA MANOVA Description

Between sums of SSCP (B) Sum of the square and cross product
squares SSbetween matrix between groups
Within sums of SSCP (W) Sum of the square and cross product
squares SSwithin matrix within groups
Grand mean SStotal SSCP (T) Sum of the square and cross product
matrix total
F test used for Wilks’ lambda W
= (division of matrices)
significance λ T
testing The variance not explained by the
independent variable. (1 - λ)2 = η2 (eta-
square) is equivalent to R2 in regression.
Lawley -
Hotelling trace
These three other statistical tests
Roy’s largest provide similar results. Minitab displays
root all four tests.
Pillai - Bartlett
trace
There is a MANOVA example in the Primer.

CSSBB 2014
VIII. ANALYZE VI.B.1

HYPOTHESIS TESTING/TERMINOLOGY
Hypothesis Testing
Hypothesis Testing is reviewed in the following topic

areas:
C Terminology C ANOVA
C Significance C Goodness-of-fit
C Sample size C Contingency tables
C Estimates C Nonparametric tests
C Major Tests
Terminology
Commonly used hypothesis test terms are presented
below.
Test Statistic
In order to test a null hypothesis, a test calculation must
be made from sample information. This calculated value
is called a test statistic and is compared to an
appropriate critical value. A decision can then be made
to reject or not reject the null hypothesis.
CSSBB 2014

Null Hypothesis
This is the hypothesis to be tested. The null hypothesis
directly stems from the problem statement and is
denoted as H0. Example:
C If a claim is made that the average of process A is

greater than the average of process B, the null
hypothesis (one-tail) would state that process A <_
process B. This is written as Ho: A <_ B.
A null hypothesis can only be rejected, or fail to be

rejected, it cannot be accepted because of a lack of
evidence to reject it. When rejecting the null hypothesis,
the alternate hypothesis must be accepted.
CSSBB 2014

Types of Errors
When formulating a conclusion regarding a population
based on observations from a small sample, two types
of errors are possible:
C Type I error: This error occurs when the null

hypothesis is rejected when it is, in fact, true. The
probability of making a type I error is called α
(alpha) and is commonly referred to as the
producer’s risk (in sampling). Examples are:
incoming products are good but called bad; a
process change is thought to be different when, in
fact, there is no difference.
C Type II error: This error occurs when the null

hypothesis is not rejected when it should be
rejected. This error is called the consumer’s risk (in
sampling) and is denoted by the symbol β (beta).
Examples are: incoming products are bad, but
called good; an adverse process change has
occurred but is thought to be no different.
The degree of risk (α) is normally chosen by the

concerned parties (α is normally taken as 5%) in arriving
at the critical value of the test statistic.
CSSBB 2014

Types of Errors (Continued)

The types of errors are shown below:
Null Hypothesis
True False
Fail to p=1-α p=β

Reject Correct Type II Error
The H0 Decision
Decision Made
Reject p=α p = 1- β
H0 Type I Correct
Error Decision
CSSBB 2014

One-Tail Test vs. Two-Tail Test

Any test of hypothesis has a risk associated with it and
one is generally concerned with the α risk (a type I error
which rejects the null hypothesis when it is true).
The level of this α risk determines the level of

confidence (1 - α) that one has in the conclusion. This
risk factor is used to determine the critical value of the
test statistic which is compared to a calculated value.
CSSBB 2014

One-Tail Test
If a null hypothesis is established to test whether a
sample value is smaller or larger than a population
value, then the entire α risk is placed on one end of a
distribution curve. This constitutes a one-tail test.
C A study was conducted to determine if the mean

battery life produced by a new method is greater
than the present battery life of 35 hours. In this
case, the entire α risk will be placed on the right tail
of the existing life distribution curve.
H0: new < or = to present H1: new > present
ENTIRE " = 5%
0
:0 = 35 HOURS
Determine if the true mean is within the α region.
CSSBB 2014

Two-Tail Test
If a null hypothesis is established to test whether a
population shift has occurred, in either direction, then a
two-tail test is required. The allowable α error is
generally divided into two equal parts. Example:
C An economist must determine if unemployment

levels have changed significantly over the past
year.
C A study is made to determine if the salary levels of

company A differ significantly from those of
company B.
H0: levels are = H1: levels are …
 = 0.025  = 0.025
2 2
-1.96 0 +1.96
0
Determine if the true mean is within either the upper or

lower α critical regions.
CSSBB 2014

HYPOTHESIS TESTING/SIGNIFICANCE
Practical vs. Statistical Significance

On occasion, an issue of practical versus statistical
significance may arise. That is, some hypothesis or
claim is found to be statistically significant, but may not
be worth the effort or expense to implement.
This could occur if a large sample was tested to a

certain value, such as a diet that results in a net loss of
0.5 pounds for 10,000 people. The result is statistically
significant, but a diet losing 0.5 pounds per person
would not have any practical significance.
Huck indicates that issues of practical significance will

often occur if the sample size is not adequate. A power
analysis may be needed to aid in the decision-making
process.
CSSBB 2014

Power of Test H0: μ = μ0

Consider a null hypothesis that a population is believed
to have mean μo= 70.0 and σX̄ = 0.80. The 95%
confidence limits are 70 ±(1.96)(0.8) = 71.57 and 68.43.
_
One accepts the hypothesis μ = 70 if Xs are between
these limits. The alpha risk is that sample means will
exceed those limits.
One can ask “what if” questions such as, “What if μ

shifts to 71, would it be detected?” There is a risk that
the null hypothesis would be accepted even if the shift
occurred. This risk is termed β.
The value of β is large if μ is close to μ0 and small if μ is

very different from μ0. This indicates that slight
differences from the hypothesis will be difficult to detect
and large differences will be easier to detect.
CSSBB 2014

Power of Test H0: μ = μ0 (Continued)

The normal distribution curves below show the null and
alternative hypotheses. If the process shifts from 70 to
71, there is a 76% probability that it would not be
detected.
Distribution, ==7070
Normal Distribution,
0.45
0.4
0.35
0.3
0.25
LCL UCL
0.2
0.15
.025 .025
0.1
0.05
0
67 68 69 70 71 72 73
X
Normal Distribution, : = 71
0.45
0.4
$
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
68 69 70 71 72 73 74
X
Illustration of Beta (β) Risk

CSSBB 2014

Power of Test H0: μ = μ0 (Continued)

To construct a power curve, 1 - β is plotted against
alternative values of μ. The power curve for the process
under discussion is shown below. A shift in a mean
away from the null increases the probability of
detection. In general, as alpha increases, beta
decreases and the power of 1 - β increases.
1
0.9
0.8
0.7
0.6
1 - β
0.5
0.4
0.3
0.2
0.1
0
67 68 69 70 71 72 73
μ
Power Curve, (1 - β) vs μ
1 - β = Probability of rejecting the null hypothesis given

that the null hypothesis is false.
CSSBB 2014

HYPOTHESIS TESTING/SAMPLE SIZE
Sample Size
The ideal procedure is to determine the α and β error
desired and then to calculate the sample size necessary
to obtain the desired decision confidence. The sample
size (n) needed for hypothesis testing depends on:
C The desired type I (α) and type II (β) risk
C The minimum value to be detected between the

population means (μ - μ0)
C The variation in the characteristic being measured

(S or σ)
Variable data sample size is illustrated by the following

example:
CSSBB 2014

Sample Size (Continued)

Assume in a pilot process one wishes to determine
whether an operational adjustment will alter the process
hourly mean yield by as much as 4 tons per hour. What
is the minimum sample size which, at the 95%
confidence level (Z=1.96), would confirm the
significance of a mean shift greater than 4 tons per
hour? Historic information suggests that the standard
deviation of the hourly output is 20 tons. The general
sample size equation for variable data (normal
distribution) is:
 1.96   20 
2 2
Z2 2
n= = = 96
 4
2
E2
Obtain 96 pilot hourly yield values and determine the

hourly average. If this mean deviates by more than 4
tons from the previous hourly average, a significant
change at the 95% confidence level has occurred. If the
sample mean deviates by less than 4 tons/hr, the
observable mean shift can be explained by chance
cause.
CSSBB 2014

Sample Size (Continued)

For binomial data, use the following formula:
Z 2  p  1 - p 
n=
 p 
2
Where,
Z = The appropriate Z value

∆p = The desired proportion interval
p̄ = Proportion rate
n = Sample size
CSSBB 2014

HYPOTHESIS TESTING/ESTIMATES
Estimators
In analyzing sample values to arrive at population
probabilities, two major estimators are used: point
estimation and interval estimation.
Consider the following tensile strength readings from 4

piano wire segments: 28.7, 27.9, 29.2 and 26.5 psi.
Based on this data, the following expressions are true:
C Point estimation: If a single estimate value is

desired (i.e., the sample average), then a point
estimate represented by X̄ can be obtained.
ΣX i
X=
n
28.7 + 27.9 + 29.2 + 26.5
=
4
= 28.08 and  s = 1.1786 
28.08 psi is the point estimate for the population

mean.
CSSBB 2014

Estimators (Continued)
C Interval Estimate or CI (Confidence Interval): From

previous sample data one can calculate the interval
within which the population mean is predicted to
fall. Confidence intervals are always estimated for
population parameters and, in general, are derived
from the mean and standard deviation of sample
data. For small samples, a critical value from the t
distribution is required and for 95% confidence, t =
3.182 for n-1 degrees of freedom. The CI equation
and interval would be:
S  1.1786 
X ± 3.182 = 28.08 ± 3.182  
n  2 
= 26.205 and 29.955
If the population sigma is known (say σ = 2 psi), the
Z distribution is used. The CI equation and interval
would be:
σ 2
X ± 1.96 = 28.08 ± 1.96   = 26.12 and 30.04
n 2
A confidence interval is a two-tail event and requires
critical values based on an alpha/2 risk in each tail.
Note that other confidence interval formulas exist.

CSSBB 2014

Confidence Intervals for the Mean

Continuous Data - Large Samples
Use the normal distribution to calculate the confidence

interval for the mean.

X  Z
2 n
The average of 100 samples is 18 with a population

standard deviation of 6. Calculate the 95% confidence
interval for the population mean.
6
 = 18  1.96 = 18  1.176
100
16.82    19.18
CSSBB 2014

Confidence Intervals for the Mean (Cont’d)

Continuous Data - Small Samples
If a relatively small sample is used (<30) then the t

distribution must be used.
s
X  t
2 n
Use the same values as in the prior example except that

the sample size is 25.
 6
 = 18  2.064 = 18  2.48
25
15.52    20.48
CSSBB 2014

Confidence Intervals for Variation

The confidence intervals for the mean were symmetrical
about the average. This is not true for the variance,
since it is based on the chi square distribution. The
formula is:
 n - 1 s2  n - 1 s2
  2

 2
 2 
,n-1 1- ,n-1
2 2
The sample variance for a set of 25 samples was found

to be 36. Calculate the 90% confidence interval for the
population variance.
 24  36   24  36 
 2 
36.42 13.85
23.72   2  62.38
CSSBB 2014

Confidence Intervals for Proportion

For large sample sizes, with n(p) and n(1-p) greater than
or equal to 4 or 5, the normal distribution can be used to
calculate a confidence interval for proportion. The
following formula is used:
p 1 - p 
p ± Zα
2
n
If 16 defectives were found in a sample size of 200 units,

calculate the 90% confidence interval for the proportion.
 0.08  0.92 
0.08 ± 1.645 = 0.08 ± 0.032
200
0.048  p  0.112
CSSBB 2014

HYPOTHESIS TESTING/MAJOR TESTS
Hypotheses Tests
X vs  
Large Means Z= x-
samples / n
x1 - x2
 x - x = 1 2
Z=
2 2
x - x2
X1 vs X 2 1 n +
2
1
Normal 1 n2
S 12 vs  2 ( - 1) s2
X = (n
2
Variances
2
S2
S12 vs S 22 F= 1
S 22
Small
t = X-

samples X vs  s/ n
(n1 - 1)s12 + (n2 - 1)s

) 22
Means 21 = 22 S2 =
n1 + n2 - 2
X1 vs X 2
S X - X2 = S 1 + 1
1
n 1 n2
x -x
t= 1 2
df = n1 + n2 - 2 S x1 - x2
A = S12 / n 1
12  22 B = S22 / n 2
S X1- X2 = A + B
Welch-Satterthwaite
Approximation
(A + B) 2
df =
A2 + B2
n1 - 1 n2 - 1
CSSBB 2014

Extreme Value Test
Normal Single Outlier Outlier ratio

samples test table
Attribute Distribution Hypotheses

p-p
p vs.  p = p (1 - p) Z=
p
n
Binomial
p1 - p2
p1 vs. p2 1 1 Z=
p1 - p2 = p(1 - p ) + p -p
n1 n2 1 2
n1 p1 + n2 p2
p=
n1 + n2
c - c
c vs.  = c Z=
c
Poisson
c = no. of defects
c1 vs. c2
k = no. samples
2 2
c1 c2 k1 + k2
2 = + - ( c1 + c2)
k1 k2 c1 + c 2
CSSBB 2014

Z Test
When the population follows a normal distribution and
the population standard deviation, σX, is known, then the
hypothesis tests for comparing a population mean, μ,
with a fixed value, μ0, are given by the following:
H0: μ = μ0 H0: μ <

_ μ0 H0: μ >
_ μ0
H1: μ =/ μ0 H1: μ > μ0 H1: μ < μ0
The null hypothesis is denoted by H0 and the alternative

hypothesis is denoted by H1. The test statistic is:
X - 0 X - 0
Z= =
X X / n
...where the sample average is X̄, the number of samples

is n and the standard deviation of the mean is σX̄. Note,
if n > 30, that the sample standard deviation, s, is often
used as an estimate of the population standard
deviation, σX. The test statistic, Z, is compared with a
critical value Zα or Zα/2 which is based on a significance
level, α, for a one-tailed test or α/2 for a two-tailed test.
CSSBB 2014

Z Test (Continued)
The average vial height from an injection molding
process has been 5.00" with a standard deviation of
0.12". An experiment is conducted using new material
which yielded the following vial heights: 5.10", 4.90",
4.92", 4.87", 5.09", 4.89", 4.95", and 4.88".
Can one state with 95% confidence that the new material
is producing shorter vials. The Z test applies. The null
and alternative hypotheses are:
H0: μ >
_ 5.00" H1: μ < 5.00"
The sample average is X̄ = 4.95" with n = 8 and the

population standard deviation is σX = 0.12". The test
statistic is:
X - 0 4.95 - 5.00 - 0.05
Z= = = = - 1.18
X / n 0.12 / 8 0.042
Since the H1 sign is <, it is a left, one-tailed test. The

level of significance, α = 1 - 0.95 = 0.05. The critical
value in a Z table, = -1.645. The test statistic, -1.18, does
not fall in the reject region. The null hypothesis cannot
be rejected. There is insufficient evidence to conclude
that the vials are shorter.
CSSBB 2014

Student’s t Test
The student’s t distribution applies to samples drawn
from a normally distributed population. It is used for
making inferences about a population mean when the
population variance, σ2, is unknown and the sample
size, n, is small usually < 30. The test statistic formula
is:
X - 0
t=
s/ n
The null and alternative hypotheses are the same as

were given for the Z test.
The test statistic, t, is compared with a critical value, tα

or tα/2, which is based on a significance level, α, for a
one-tailed test or α/2 for a two-tailed test, and the
number of degrees of freedom, d.f. The degrees of
freedom is determined by the number of samples, n, and
is simply:
d.f. = n - 1
CSSBB 2014

Student’s t Test (Continued)

A chemical process has been 880 tons (μ = 880 tons). A
new process has been evaluated for 25 days with a yield
of 900 tons (X̄) and sample standard deviation, s = 20
tons. Can one say with 95% confidence that the process
has changed? The null and alternative hypotheses are:
H0: μ = 880 tons H1: μ =/ 880 tons
The test statistic calculation is:
X - 0 900 - 880 20
t= = = =5
s/ n 20 / 25 4
Since the H1 sign is =/, it is a two-tailed test and with a

95% confidence, the level of significance, α = 0.05.
Since it is a two-tail test, α/2 is used to determine the
critical values. The degrees of freedom, d.f. = n - 1 = 24.
t0.025 = -2.064 and t0.975 = 2.064. Since the test statistic, 5,
falls in the right-hand reject (or critical) region, the null
hypothesis is rejected. One concludes with 95%
confidence that the process has changed.
This technique was developed by W.S. Gosset and

published in 1908 under the pen name “Student.”
CSSBB 2014

t Distribution Table
d.f. t0.100 t0.050* t0.025** t0.010 t0.005

1 3.078 6.314 12.706 31.821 63.657
2 1.886 2.920 4.303 6.965 9.925
3 1.638 2.353 3.182 4.541 5.841
4 1.533 2.132 2.776 3.747 4.604
5 1.476 2.015 2.571 3.365 4.032
6 1.440 1.943 2.447 3.143 3.707
7 1.415 1.895 2.365 2.998 3.499
8 1.397 1.860 2.306 2.896 3.355
9 1.383 1.833 2.262 2.821 3.250
10 1.372 1.812 2.228 2.764 3.169
11 1.363 1.796 2.201 2.718 3.106
12 1.356 1.782 2.179 2.681 3.055
13 1.350 1.771 2.160 2.650 3.012
14 1.345 1.761 2.145 2.624 2.977
15 1.341 1.753 2.131 2.602 2.947
16 1.337 1.746 2.120 2.583 2.921
17 1.333 1.740 2.110 2.567 2.898
18 1.330 1.734 2.101 2.552 2.878
19 1.328 1.729 2.093 2.539 2.861
20 1.325 1.725 2.086 2.528 2.845
21 1.323 1.721 2.080 2.518 2.831
22 1.321 1.717 2.074 2.508 2.819
23 1.319 1.714 2.069 2.500 2.807
24 1.318 1.711 2.064 2.492 2.797
25 1.316 1.708 2.060 2.485 2.787
26 1.315 1.706 2.056 2.479 2.779
27 1.314 1.703 2.052 2.473 2.771
28 1.313 1.701 2.048 2.467 2.763
29 1.311 1.699 2.045 2.462 2.756
inf. 1.282 1.645 1.960 2.326 2.576
* One-tail 5% α risk ** Two-tail 5% α risk
CSSBB 2014

Chi Square (O2) Test

When the population follows a normal distribution, the
hypothesis tests for comparing a population variance,
σ2X, with a fixed value, σ20, are given by the following:
H0: σ2X = σ20 H0: σ2X <

_ σ20 H0: σ2X >
_ σ20
H1: σ2X =/ σ20 H1: σ2X > σ20 H1: σ2X < σ20
The test statistic is given by:
 n - 1 s2
 =
2
 2x
The test statistic, O2, is compared with a critical value

O2α or O2α/2 which is based on a significance level, α, for
a one-tailed test or α/2 for a two-tailed test. The degrees
of freedom is:
d.f. = n - 1
If the H1 sign is =/, it is a two-tailed test. If the H1 sign is

>, it is a right, one-tailed test, and if the H1 sign is <, it is
a left, one-tailed test.
CSSBB 2014

Chi Square (O2) Test (Continued)
Left tail Right tail
f(x2) f(x2)
0 O21-α 0 O2α
Case I, Population Variance is Known.

Will new material show a four sigma tensile variation
less than or equal to 60 psi 95% confidence? This
translates to a sigma of 15 psi. An eight sample test
yielded a standard deviation of 8 psi.
 15   15 
2 2
H0 : 12  H1: 12 <
This is a left tail test with n - 1 = 7. The critical value for
95% confidence is 2.17. The calculated statistic is:
 n - 1 s2  7  8 
2
2 = = = 1.99
  15 
2 2
Since 1.99 is less than 2.17, the null hypothesis must be

rejected. There is decreased variation in the new steel
alloy tensile strength.
CSSBB 2014


DF O20.99 O20.95 O20.90 O20.10 O20.05 O20.01
1 0.00016 0.0039 0.0158 2.71 3.84 6.63
2 0.0201 0.1026 0.2107 4.61 5.99 9.21
3 0.115 0.352 0.584 6.25 7.81 11.34
4 0.297 0.711 1.064 7.78 9.49 13.28
5 0.554 1.15 1.61 9.24 11.07 15.09
6 0.872 1.64 2.20 10.64 12.59 16.81
7 1.24 2.17 2.83 12.02 14.07 18.48
8 1.65 2.73 3.49 13.36 15.51 20.09
9 2.09 3.33 4.17 14.68 16.92 21.67
10 2.56 3.94 4.87 15.99 18.31 23.21
11 3.05 4.57 5.58 17.28 19.68 24.73
12 3.57 5.23 6.30 18.55 21.03 26.22
13 4.11 5.89 7.04 19.81 22.36 27.69
14 4.66 6.57 7.79 21.06 23.68 29.14
15 5.23 7.26 8.55 22.31 25.00 30.58
16 5.81 7.96 9.31 23.54 26.30 32.00
18 7.01 9.39 10.86 25.99 28.87 34.81
20 8.26 10.85 12.44 28.41 31.41 37.57
24 10.86 13.85 15.66 33.20 36.42 42.98
30 14.95 18.49 20.60 40.26 43.77 50.89
40 22.16 26.51 29.05 51.81 55.76 63.69
60 37.48 43.19 46.46 74.40 79.08 88.38
120 86.92 95.70 100.62 140.23 146.57 158.95
Chi Square Critical Values
The table addresses both the left and right tails of O2.
95% of the area under the O2 distribution lies to the right
of O20.95.
CSSBB 2014


Chi square Case II. Comparing observed and expected
Frequencies of test outcomes. (attribute data). This
application of chi square is called the contingency table
or row and column analysis.
The chi-square statistic using observed frequencies (O),

expected frequencies (E),and degrees of freedom (R-
1)(C-1) with the entire level of significance, α, in the
one-tail, right side, of the distribution.
O - E
2
 = 2
 E
An airport authority wanted to evaluate the ability of

three X-ray inspectors to detect transistor radios in
luggage.
Inspectors Treatment Totals
1 2 3
Radios detected 27 25 22 74
Radios undetected 3 5 8 16
Sample total 30 30 30 90
Is there any significant difference in the abilities of the

inspectors? (95% confidence)
CSSBB 2014


H0: p1 = p2 = p3 (No difference among the inspectors)
H1: p1 =/ p2 =/ p3 (at least one porportion is different)
DF= (rows - 1)(columns - 1) = (2-1)(3-1) = 2
The critical value of O2 for DF = 2 and α = 0.05 in the one-
tail, right side of the distribution, is 5.99
The expected values for the test are shown below:
Treatment
1 2 3 Totals
Radios detected 24.67 24.67 24.67 74
Radios undetected 5.33 5.33 5.33 16
Sample total 30 30 30 90
O - E
2
2 =  E
 2.33   0.33   2.67   2.33   0.33   2.67 
2 2 2 2 2 2
2 = + + + + +
24.67 24.67 24.67 5.33 5.33 5.33
 2 = 0.220 + 0.004 + 0.289 + 1.019 + 0.020 + 1.338
 2 = 2.89
Since the calculated value of O2 is less than the

previously calculated critical value of 5.99 and this is a
right tail test, the null hypothesis cannot be rejected.
There is insufficient evidence to say with 95%
confidence that the abilities of the inspectors differ.
CSSBB 2014

p Test
When testing a claim about a population proportion,
with a fixed number of independent trials having
constant probabilities, a p test can be used. When np <
5 or n(1-p) < 5, the binomial distribution is used to test
hypotheses relating to proportion.
The hypothesis tests for comparing a sample

proportion, p, with a fixed value, p0, are given by the
following:
H0: p = p0 H0: p <

_ p0 H0: p >
_ p0
H1: p =/ p0 H1: p > p0 H1: p < p0
The null hypothesis is denoted by H0 and the alternative

hypothesis is denoted by H1. The test statistic is given
by:
x - np0
Z=
np0  1 - p0 
CSSBB 2014

Paired-Comparison Hypotheses Tests

2 Mean, Equal Variance t Test
Tests the difference between 2 sample means (X̄1 vs X̄2)

when σ1 and σ2 are unknown but considered equal.
H0: μ1 = μ2 H1: μ1 =/ μ2
sp = Pooled standard deviation
 n1 - 1 s12 +  n2 - 1 s22
sp =
n1 + n2 - 2
X1 - X 2
t= DF = n1 + n2 - 2
1 1
sp +
n1 n2
CSSBB 2014

Paired-Comparison Hypotheses (Cont’d)

2 Mean, Unequal Variance t Test
Tests the difference between 2 sample means (X̄1 vs X̄2)

when σ1 and σ2 are unknown, but are not considered
equal.
H0: μ1 = μ2 H1: μ1 =/ μ2
X1 - X 2
t=
s12 s22
+
n1 n2
1
DF = 2 2
 s12   s22 
 n   n 
 2 1 2  2 2 2
 s1 s2   s1 s2 
n + +
 1 n2  n
 1 n2 
+
n1 - 1 n2 - 1
CSSBB 2014

Paired-Comparison Hypotheses (Cont’d)

Paired t Test
Tests the difference between 2 sample means. Data is

taken in pairs with the difference calculated for each
pair.
H0: μ1 = μ2 H1: μ1 =/ μ2
d
t=
sd n
DF = n - 1. A paired t test is usually a two tail test.

d = average of differences of pairs of data.
The paired method (dependent samples), compared to

treating the data as two independent samples, will often
show a more significant difference because the standard
deviation of the d’s (Sd) includes no sample-to-sample
variation. This comparatively more frequent
significance occurs despite the fact that “n - 1”
represents fewer degrees of freedom than “n1 + n2 -2.”
In general, the paired t test is a more sensitive test than

the comparison of two independent samples.
CSSBB 2014

F Test
If independent random samples are drawn from two
normal populations with equal variances, the ratio of
(s1)2/(s2)2 creates a sampling distribution known as the
F distribution. The hypothesis tests for comparing a
population variance, σ21, with another population
variance, σ22, are given by the following:
H0: σ21 = σ22 H0: σ21 <

_ σ22 H0: σ21 >
_ σ22
H1: σ21 =/ σ22 H1: σ21 > σ22 H1: σ21 < σ22
The number of degrees of freedom associated with s12

and s22 are represented by ν1 and ν2 respectively. ν1 is
the DF in the numerator.
The F statistic is the ratio of two sample variances (two

chi square distributions) and is given by the formula:
S12
F= 2
S2
Where s12 and s22 are sample variances.
It is customary to designate the larger sample variance

as s12 and place it in the numerator.
CSSBB 2014

F Test (Continued)
There are numerous F Table formats based on α values
of 0.10, 0.05, 0.025, 0.01, etc. Listed below is a partial F
Table for α = 0.05.
ν 1
1 2 3 4 5 6 7 8 9 10
ν 2
1 161.4 199.5 215.7 224.6 230.2 234.0 236.8 238.9 240.5 241.9
2 18.51 19.00 19.16 19.25 19.30 19.33 19.35 19.37 19.38 19.40
3 10.13 9.55 9.28 9.12 9.01 8.94 8.89 8.85 8.81 8.79
4 7.71 6.94 6.59 6.39 6.26 6.16 6.09 6.04 6.00 5.96
5 6.61 5.79 5.41 5.19 5.05 4.95 4.88 4.82 4.77 4.74
6 5.99 5.14 4.76 4.53 4.39 4.28 4.21 4.15 4.10 4.06
7 5.59 4.74 4.35 4.12 3.97 3.87 3.79 3.73 3.68 3.64
8 5.32 4.46 4.07 3.84 3.69 3.58 3.50 3.44 3.39 3.35
9 5.12 4.26 3.86 3.63 3.48 3.37 3.29 3.23 3.18 3.14
10 4.96 4.10 3.71 3.48 3.33 3.22 3.14 3.07 3.02 2.98
The above critical values for F may be used for a one-

tailed test (α=0.05, 95% confidence) or a two-tailed test
(α/2=0.05, 90% confidence).
CSSBB 2014

F Test (Continued)
f(F)
f()
F Distribution for Two Normal Samples
Example: s1 = 900 psi, n1 = 9, s2 = 300 psi, n2 = 7.

At a 95% confidence level, can one conclude there is
now less variation?
H0: σ12 <
_ σ22 H1: σ12 > σ22 and DF1 = 8 DF2 = 6
Use a one tail test with the entire α risk in the right tail.
The critical value of F is 4.15. Since the calculated F
value is in the critical region, the null hypothesis is
rejected. There is sufficient evidence to indicate a
reduced variance.
CSSBB 2014

Summary of Inference Tests

TYPE TEST STATISTIC DF APPLICATION
X-
Z= Single sample mean. Standard
Z 
N.A. deviation of population is known.
n
X-
t= Single sample mean. Standard
t Test S
n-1 deviation of population unknown.
n
2 sample means. Variances are
2 Mean X1 - X 2 unknown, but considered equal.
Equal t=
1 1 n1 + n2 - 2
Variance *Sp + n1 - 1 S12 +  n2 - 1 S22
t Test n1 n2 *Sp =
n1 + n2 - 2
2 Mean X1 - X 2 2 sample means. Variances are

Unequal t= unknown, but considered
Variance S12 S2 * unequal. DF is determined from
+ 2
t Test n1 n2 the Welch-Satterthwaite
d 2 sample means. Data is taken in

Paired t= pairs. A different d is calculated
sd / n n-1
t Test for each pair.
O2  n - 1 S2 Tests sample variance against

2 = n-1
σ2 Known  2 known variance.
 O - E 2 Compares observed and expected

O2 2
 =  E
(r-1)(c-1)
frequencies of test outcomes.
S12 n1 - 1 Tests if two sample variances are

F=
F S22 n2 - 1 equal.
CSSBB 2014

HYPOTHESIS TESTING/ANOVA
Analysis of Variance Introduction

In many investigations, it is necessary to compare three
or more population means simultaneously. There are
underlying assumptions in this analysis of variance of
means: the variance is the same for all factor
treatments or levels, the individual measurements within
each treatment are normally distributed and the error
term is considered a normally and independently
distributed random effect.
With analysis of variance, the variations in response

measurement are partitioned into components that
reflect the effects of one or more independent variables.
The variability of a set of measurements is proportional
to the sum of squares of deviations used to calculate the
variance:
G(X-X̄)2
Analysis of variance partitions the sum of squares of

deviations of individual measurements from the grand
mean (called the total sum of squares) into parts: the
sum of squares of treatment means plus a remainder
which is termed the experimental or random error.
CSSBB 2014

A Comparison of Three or More Means

An analysis of variance to detect a difference in three or
more population means first requires obtaining some
summary statistics for calculating variance of a set of
data as shown below:
Where:
GX2 is called the crude sum of squares
(GX)2/N is the CM (correction for the mean), or CF

(correction factor)
GXGX2 - (GX)2/N is termed SS (total sum of squares, or

corrected SS).
ΣX 2 -  ΣX  /N
2
= σ2  variance 
N-1
Total Sum of Squares
=
Total DF  degrees of freedom 
CSSBB 2014

One-Way ANOVA
In the one-way ANOVA, the total variation in the data has
two parts: the variation among treatment means and the
variation within treatments. The equations are given in
the Primer.
To test the null hypothesis:
HO: μ1 = μ2 = ... = μt H1: At least one mean different
MST
F When F > Fα , reject HO
MSE
As an example of a comparison of three means,

consider a single factor experiment: The following
coded results were obtained from a single factor
randomized experiment, in which the outputs of three
machines were compared. Determine if there is a
significant difference in the results (α=0.05).
Machines Data Sum n Avg TCM = (Sum)2/n GX2

A 5, 7, 6, 7, 6 31 5 6.2 192.2 195
B 2, 0, 1, -2, 2 3 5 0.6 1.8 13
C 1, 0, -2, -3, 0 -4 5 -0.8 3.2 14
Total 30 15 197.2 222
CSSBB 2014

One-Way ANOVA (Continued)

ΣX = 30 N = 15 Total DF = N - 1 = 15 - 1 = 14
GM = ΣX/N = 30/15 = 2.0 ΣX2 = 222
 ΣX   30 
2 2
CM = = = 60
N 15
Total SS = ΣX2 - CM = 222 - 60 = 162
Σ(TCM) = 197.2
SST = Σ(TCM) - CM = 197.2 - 60 = 137.2
SSE = Total SS - SST = 162 - 137.2 = 24.8
The completed ANOVA table is:

Source
SS DF Mean Square F Fα,ν1,ν2
(of variation)
Machines 137.2 2 68.6 33.2 F0.05,2,12 = 3.89
Error 24.8 12 2.067
Total 162 14 σe = 2.07 = 1.44
Since the computed value of F (33.2) exceeds the critical

value of F, the null hypothesis is rejected.
CSSBB 2014

Two-Way ANOVA
It will be seen that the two-way analysis procedure is an
extension of the patterns described in the one-way
analysis. Recall that a one-way ANOVA has two
components of variance: Treatments and experimental
error (may be referred to as columns and error or rows
and error).
In the two-way ANOVA there are three components of

variance: Factor A treatments, Factor B treatments, and
experimental error (may be referred to as columns,
rows, and error).
An example of a two Factor two-way ANOVA with two

instructors and three materials is given in the Primer.
CSSBB 2014

Components of Variance
The analysis of variance can be extended with a
determination of the COV (components of variance).
The COV table uses the MS (mean square), F, and
F(alpha) columns from the previous ANOVA TABLE and
adds columns for EMS (expected mean square),
variance, adjusted variance and percent contribution to
design data variation. The model for the ANOVA is:
Xijk = μ + Mi + Ij + MIij + εijk
The model states that any measurement (X) represents

the combined effect of the population mean (μ), the
different materials (M), the different instructors (I), the
materials/instructor interaction (MI), and the
experimental error (ε).
Where: i represents materials at 3 levels, j represents

instructors at 2 levels, k represents 3 replications per
cell.
CSSBB 2014

ANOVA for A x B Factorial Experiment

In a factorial experiment involving factor A at a levels
and factor B at b levels, the total sum of squares can be
partitioned into:
Total SS = SS(A) + SS(B) + SS(AB) + SSE
ANOVA Table for an A x B Factorial Experiment
Source DF SS MS
Factor A (a-1) SS(A) SS(A)/(a-1)

Factor B (b-1) SS(B) SS(B)/(b-1)
Interaction AB (a-1)(b-1) SS(AB) SS(AB)/(a-1)(b-1)
Error (N-ab) SSE SSE/(N-ab)
Total (N-1) Total SS

CSSBB 2014

ANOVA for a Randomized Block Design

The randomized block design implies the presence of
two independent variables, blocks and treatments.
The total sum of squares of the response measurements

can be partitioned into three parts, the sum of the
squares for the blocks, treatments, and error. The
analysis of a randomized block design is of less
complexity than an A x B factorial experiment.
ANOVA Table for a Randomized Block Design
Source DF SS MS
Blocks b-1 SSB MSB=SSB/(b-1)

Treatments t-1 SST MST=SST/(t-1)
Error (b-1)(t-1) SSE MSE=SSE/(b-1)(t-1)
Total bt-1 Total SS

CSSBB 2014

HYPOTHESIS TESTING/GOODNESS-OF-FIT TESTS
Goodness-of-Fit Tests
GOF (goodness-of-fit) tests are part of a class of
procedures that are structured in cells. In each cell
there is an observed frequency, (Fo). One can calculate
the expected or theoretical frequency, (Fe). Chi square
(O2) is then summed across all cells according to the
formula:
 Fo - Fe 
2
2 =  Fe
The calculated chi square is then compared to the chi

square critical value for the following appropriate
degrees of freedom:
GOF Distribution (DF)

Normal No. of cells - 3
Poisson No. of cells - 2
Binomial No. of cells - 2
Uniform No. of cells - 1
The Primer provides examples for the following:

C Uniform Distribution (GOF)
C Normal Distribution (GOF)
C Poisson Distribution (GOF)
C Binomial Distribution (GOF)
CSSBB 2014

HYPOTHESIS TESTING / CONTINGENCY TABLES
Contingency Tables
A two-way classification table (rows and columns)
containing original frequencies can be analyzed to
determine whether the two variables (classifications) are
independent or have significant association. That the
chi square procedure will test whether there is
dependency between the two classifications. In
addition, a contingency coefficient (correlation) can be
calculated. If the chi square test shows a significant
dependency, the contingency coefficient shows the
strength of the correlation.
A measure of the difference found between observed

and expected frequencies is supplied by the statistic chi
square, O2, where:
 On - En   O1 - E1   O2 - E2   Ok - Ek 
2 2 2 2
k
2 = 
n=1 En
=
E1
+
E2
++
Ek
If O2 = 0, the observed and theoretical frequencies agree

exactly. If O2 > 0, they do not agree exactly. The larger
the value of O2, the greater the discrepancy between
observed and theoretical frequencies.
CSSBB 2014

Contingency Tables (Continued)

Example: Each of 15 cells makes a contribution to chi
square (O2). For a selected (illustrative) cell, the
contribution is:
 O - E  15 - 12.5 
2 2
= = 0.48
E 12.5
Chi Square over all cells equals 66.22.
Assume alpha to be 0.01.
d.f. = (rows - 1) x (columns - 1) = (5 - 1) x (3 - 1) = 8
The critical chi square:  2 =  20.01, 8 = 20.09
The calculated chi square is larger than critical chi

square, therefore, one rejects the null hypothesis of
hospital equality of results.
CSSBB 2014

Coefficient of Contingency (C)

The degree of relationship, association or dependence
of the classifications in a contingency table is given by:
2
C=
2 + N
Where N equals the grand frequency total.
The contingency coefficient for the previous example is:
2 66.22
C= = = 0.38
2 + N 66.22 + 393
The maximum coefficient of contingency is:

k-1 3-1
Max C = = = 0.816
k 3
Correlation of Attributes
For (k = r = c) tables, the correlation coefficient, φ, is
defined as:
2
=
N  k - 1
Where 0 <
_φ<
_1
CSSBB 2014

HYPOTHESIS TESTING/NONPARAMETRIC TESTS
Parametric vs. Nonparametric Tests

Parametric implies that a distribution is assumed for the
population, commonly the normal distribution.
Nonparametric implies that there is no assumption of a
specific distribution for the population.
An advantage of a parametric test is that the probability

of rejecting H0, when it is false, is higher than the power
of a corresponding nonparametric test with equal
sample sizes.
An advantage of nonparametric tests is that the test

results are more robust against violation of the
assumptions.
Nonparametric Techniques
Nonparametric techniques of hypothesis testing are
applicable for many quality engineering problems and
projects. The nonparametric tests are often called
“distribution-free” since they make no assumption
regarding the population distribution. Nonparametric
tests may be applied ranking tests in which data is not
specific in any continuous sense, but are simply ranks.
CSSBB 2014

Nonparametric Techniques (Continued)

Three powerful nonparametric techniques will be
described with examples:
C The Kruskal Wallis One Way ANOVA
C The Mann Whitney U Test
C Wilcoxon-Mann-Whitney Rank Sum Test

CSSBB 2014

Kruskal-Wallis ANOVA by Ranks

This is a test of independent samples. The
measurements may be continuous data, but the
underlying distribution is either unknown, or known to
be non-normal.
Example results
Plant A Plant B Plant C
Rank Sum 74.5 54.5 102.0
n 8 6 7
(RankSum)2/n 693.781 495.042 1486.286
G = (RankSum)2/n = 693.781 + 495.042 + 1486.286

= 2675.109 N = 8 + 6 + 7 = 21
The significance statistic is H. H is distributed as chi

square. Tie values are included in the calculation of chi
square. Let t = number of tied values in each tied set
Then T = t3 - t for that set.
CSSBB 2014

Kruskal-Wallis (Continued)
 12 
 G - 3  N+1 
N  N + 1 
H=  
 J 
 1- 
  N3 - N 
 12 
  2675.109  - 3  21+1 
 21 22 
H=   = 3.76
 690 
 1- 
  213
- 21 
Let k = number of sample sets. DF = k - 1 = 3 - 1 = 2.

Let α = 0.05. Critical chi square =  20.05, 2 = 5.99
H is less than critical chi square. Therefore, the null

hypothesis of equality of population medians cannot be
rejected.
The Primer contains examples for:
C Mann-Whitney U Tests
C Wilcoxon -Mann-Whitney Rank Sum Test
CSSBB 2014

Nonparametric Test Summary

C One-sample sign performs a test of the median and
calculates the corresponding point estimate and
confidence interval. This test is a nonparametric
alternative to the one-sample Z and one-sample t
tests.
C One-sample Wilcoxon performs a signed rank test

of the median and calculates the corresponding
point estimate and confidence interval. This test is
a nonparametric alternative to the one-sample Z and
one-sample t tests.
C Mann-Whitney performs a hypothesis test of the

equality of two population medians and calculates
the corresponding point estimate and confidence
interval. This test is a nonparametric alternative to
the two-sample t test.
CSSBB 2014

Nonparametric Test Summary (Cont’d)

C Kruskal-Wallis performs a hypothesis test of the
equality of population medians for a one-way design
(two or more populations). This test is a
nonparametric alternative to the one-way analysis
of variance.
C Mood’s median test performs a hypothesis test of

the equality of population medians in a one-way
design. It provides a nonparametric alternative to
the usual one-way analysis of variance.
The Kruskal-Wallis test is more powerful (the confidence

interval is narrower, on average) than Mood’s median
test for analyzing data from many distributions,
including data from the normal distribution, but is less
robust against outliers.
CSSBB 2014

Nonparametric Test Summary (Cont’d)

Test Name Data Type Test Application
2
Kruskal-Wallis Test Measurement or O Data is ranked or converted to ranks for a 1-way
Count analysis of variance.
Kendall Coefficient Ranked Data O2 Determines degree of association among
of Concordance classifications based on ranked scores.
Spearman Rank Ranked Data Formula A measure of association (rs) which requires
Correlation both variables be measured on at least an
Coefficient ordinal scale.
Kendall Rank Ranked Data Formula Same as Spearman, except it calculates r, which
Correlation O2 also permits determining partial correlation
Coefficient coefficient.
Contingency Count Data O2 A measure of association between two
Coefficient classifications.
Mann-Whitney U Ranked Data Tables Determines if two independent groups are from
Test same population. An alternative if t test
assumptions cannot be met.
Wilcoxon-Mann- Ranked Data Tables Same as above. Slightly simpler to use. An
Whitney Rank Sum alternative if Z and t test assumptions cannot be
Test met. It is equivalent to the Mann-Whitney U test.
Levene’s Test Converts Data to t Verifies homogeneity of variances across k
Squares of samples. An alternative procedure if F and t
Deviations assumptions are not met.
Mood’s Median Sample Medians O2 Determines equality of sample medians by
Test scoring sample medians relative to population
median.
Kolmogorov- Sample Values Table Goodness-of-fit between observed scores and
Smirnov 1-sample specified theoretical distribution.
Kolmogorov- Sample Values Table Determines whether two independent samples
Smirnov 2-sample come from the same distribution.
McNemar Test Classification or O2 Determines significance of change in before and
Ranks after designs.
Walsh Test Response Scores α% of 2N Randomization test for matched pairs.
Fisher Exact Classification or Table or Determines whether two groups differ in the
Probability Ranks alpha proportions within two classifications.
Cochran Q Test Classification or O2 Determines whether 3 or more matched set of
Ranks proportions differ significantly.
Friedman Test Ranks O2 2-way analysis of variance for k matched
samples.
Runs Test Symbol Table Determines whether a sequence of samples are
Sequence randomly distributed.
CSSBB 2014
VIII. ANALYZE VI.C

FAILURE MODE AND EFFECTS ANALYSIS
Failure Mode and Effects Analysis

A FMEA provides a systematic technique to analyze a
system, subsystem, or item for all potential or possible
failure modes. This method then places a probability
that the failure mode will actually occur and what the
effect of this failure is on the rest of the system. If
criticality of failure is considered, the technique is called
FMECA.
A FMEA is a detailed analysis of a system down to the

component level. Once all items are classified as to the
1) failure mode, 2) effect of the failure, and 3) probability
that failure will occur, they are rated as to their severity
via an index called a RPN (risk priority number).
It is common to work from the highest RPN value down.

CSSBB 2014
VIII. ANALYZE VI.C

FMEA Process Steps

1. FMEA number: This is a log controlled number.
2. The part number or appropriate description.
3. The design responsibility
4. The person responsible for FMEA preparation
5. The date the FMEA was prepared or revised
6. The subsystem or component part number
7. The component function
8. The potential failure mode
9. The potential effect of failure
10. The potential cause of failure
11. What are the controls to prevent recurrence?

CSSBB 2014
VIII. ANALYZE VI.C

Risk Assessment and RPN

The next major action is to weigh the associated risks.
12. P is the probability this failure mode will occur.

Values for this index generally index from 1 to 10
13. S is the severity of the effect of the failure on the

rest of the system if the failure occurs. These
values are often indexed from 1 to 10.
14. D is a measure of the effectiveness of the current

controls (in place) to identify the potential weakness
or failure prior to release to production. This index
may also range from 1 to 10.
15. RPN. The risk priority number is the product of the

indices from the previous three columns.
RPN = (P)(S)(D)
16. Actions are then based upon what items either have
the highest RPN or have major safety issues.
17. There is a column for actions to be taken to reduce

the risk, including responsibilities
CSSBB 2014
VIII. ANALYZE VI.C

Example FMEA
Part No./Name: 37XT11-Lock Mech. P = Probability FMEA No. 43
Project: Re-design S = Seriousness Final Design Deadline: July 15, 2014
Other Departments: Shop Service, etc. D = Likelihood Prepared By: RCD
Subsystem Name: Quill Clamping Mechanisms RPN = Risk Priority Reviewed By: BLW
Suppliers Involved: Wilton and others Number FMEA Date: 6-15-12 Rev.
Design Responsibility: Bob Dovich
PART FUNCTION POTENTIAL POTENTIAL CURRENT RISK RECOMMENDED ACTION(S) REVISED RISK RESPONSIBLE
NUMBER FAILURE EFFECT(S) POTENTIAL CONTROLS ASSESSMENT CORRECTIVE TAKEN ASSESSMENT DEPT OR
NAME MODE(S) OF FAILURE CAUSE(S) OF ACTION(S) INDIVIDUAL
FAILURE
P S D RPN P S D RPN
WILTON CLAMP LEAK HOUSE- ACCEPT DISCUSS WITH

POWER KEEPING WEAR SUPPLIER'S 2 4 3 24 SUPPLIER
LOCK INFO
LOSES MACHINING SELECTED PERFORM LOAD

CLAMPING PARTS INADEQUATE ENG. TESTS
2 4 4 32
FORCE OVERSIZE SIZE POWER STANDARD
(SHIFTING) LOCK
MATERIALS & NONE

WORKMAN- STD. Q.C. 1 4 2 8
SHIP
REVIEW NEED FOR

OVER SYSTEM TO
NONE 2 4 2 16
PRESSURE PREVENT OVER-
PRESSURIZATION
REVIEW PRESSURE
PUMP ENG.
1 4 2 8 DELIVERED IN FIELD
SIZING STANDARD
AND ACTUAL NEED
CSSBB 2014
VIII. ANALYZE VI.C

Risk Assessment
Risk assessment is the combination of the probability of
an event or failure and the consequence(s) of that event
or failure to a system’s operators, users, or its
environment. The analysis of risk of failure normally
utilizes two measures of failure:
C Severity of failure
C Probability of failure
The severity of failure is generally defined by the hazard

severity categories from MIL-STD-1629.
Classification Description
I Catastrophic A failure that may cause death or mission loss.
II Critical A failure that may cause severe injury or major system
damage.
III Marginal A failure that may cause minor injury or degradation in
mission performance.
IV Minor A failure that does not cause injury or system damage
but may result in system failure and unscheduled
maintenance.
CSSBB 2014
VIII. ANALYZE VI.C

Risk Assessment (Continued)

An example, for commercial applications, is a severity
index based on a scale of 1 to 10.
Rank Criteria
1 Unreasonable to expect that the minor nature
of this failure will degrade system
performance.
2-3 Minor nature of failure will cause slight
annoyance to the customer.
4-6 Moderate failure will cause customer
dissatisfaction. Customer will notice some
system performance deterioration.
7-8 High degree of customer dissatisfaction and
inoperability of the system. Does not involve
safety or noncompliance to regulations.
9 - 10 Very high severity ranking in terms of safety-
related failures and nonconformance to
regulations and standards.
CSSBB 2014
VIII. ANALYZE VI.C


The hazard classification or severity index is generated
for each component or subsystem. This classification
is based on the expected results of the failure of the
component or subsystem. The probability of failure may
also be ranked. A common ranking of failure
probabilities is shown below.
Failure Level Description Probability

A High likelihood >10-1
B Probable occurrence 10-1 to 10-2
C Occasionally occurs 10-2 to 10-3
D Remote probability 10-3 to 10-6
E Highly unlikely <10-6
CSSBB 2014
VIII. ANALYZE VI.C


A number of systems are used to combine the
probability of failure and the hazard category. These
systems are based on accepting a degree of risk of
occurrence with respect to the severity of the hazard.
The Table below shows one type of risk assessment.
Hazard Category Allowable Failure Probability Level

I Catastrophic E (Unlikely)
II Critical E (Unlikely)
III Marginal D (Remote)
IV Minor C (Occasional)
Frequently, catastrophic failure modes have additional

safety measures, such as redundant components, or
frequent inspection during service, etc.
No hazard category has an allowable failure probability

of “frequent” or “probable.”
CSSBB 2014
VIII. ANALYZE VI.C

Failure Mechanisms Versus Modes

The failure mode is the actual symptom of the failure.
Failure modes are the result of failure mechanisms.
Failure mechanisms are the individual, or multiple

reasons that cause the failure mode. For instance,
failure mechanisms might be corrosion, or
contamination.
CSSBB 2014
VIII. ANALYZE VI.C

Types of FMEAs
There are four types of FMEAs:
C Design FMEAs
C Process FMEAs
C System FMEAs
C Functional FMEAs
The following is an abbreviated description of the types:
Design FMEAs are performed on the product or service

(system) at the design level. Design FMEAs are used
before products are released to manufacturing.
Process FMEAs are performed on the manufacturing

processes. They are conducted through the planning
phase as an aid during production.
System FMEAs comprise part level FMEAs. All of the

part level FMEAs will tie together to form the system.
Functional FMEAs focus on the performance of the

intended part or device.
CSSBB 2014
VIII. ANALYZE VI.D.1

ADDITIONAL ANALYSIS METHODS/GAP ANALYSIS
Additional Analysis Methods
Additional Analysis Methods are reviewed in the

following topic areas:
C Gap analysis
C Root cause analysis
C Waste analysis
Gap Analysis
Gap analysis is an assessment tool to compare an
organization’s current performance to a desired or
potential performance. The objective of a gap analysis
is to identify the difference between what is and what
should be.
What should be (the desired result)?
This is the gap needing attention
What is (the current result)?

CSSBB 2014

Gap Analysis (Continued)

Gap analysis provides an organization with the mandate
to direct resources to the appropriate areas and thereby
achieve a desirable objective. Examples of such
objectives are:
C Stay in business
C Retain or improve market share
C Improve employee moral
C Match or exceed the best external benchmarks
C Match or exceed the current competition
C Reduce cycle times
C Improve customer satisfaction
C Lower internal costs
C Achieve ISO 9001 certification
C Increase productivity
C Improve quality levels
Any form of gap analysis requires a minimum of three

categories of information:
C Where are we now?

C Where do we want to be?
C How are we going to measure the results?
CSSBB 2014

Scenario Planning
When developing strategic plans, management may be
prone to groupthink or overconfidence. In order to
overcome potential errors in decision-making, best and
worst case scenarios should be considered. Once the
different scenarios are visualized, appropriate
contingency plans can be developed.
Scenario planning is a strategic process in which a

relatively small number of futuristic stories are
developed. These scenarios depict how an
organization’s future might unfold, and then proceeds to
develop the best approaches for addressing these
events.
CSSBB 2014

Scenario Planning (Continued)

All aspects of scenario planning are difficult to formalize
because many essential elements are often unknown.
Scenario planning recognizes that there can certainly be

subjectivity in the interpretation of facts or data. The
scenario planning technique is a group process which
encourages an open knowledge exchange surrounding
the central issues important to an organization. The
goal is to develop a number of potential responses to
the perceived corporate risks.
CSSBB 2014


The scenario planning process follows a sequence
similar to that described below:
C Select people who will contribute a wide range of

perspectives
C Develop a list of perceived social, technical, or

economic shifts
C Cluster these views into related patterns
C Develop a list of the best views (priorities)
C Develop a rough picture of the future based on

these priorities
C Determine how scenarios affect the organization
C Determine potential courses of action to take
C Identify early warning signs of scenario arrival
C Monitor, evaluate, and review the scenarios

CSSBB 2014


Often a list of 6 - 10 perceived threats or opportunities
are condensed into 2 or 3 developed scenarios.
There are some traps to avoid in scenario planning:
C Not using an experienced facilitator
C Considering scenarios to be forecasts
C Making scenarios too simplistic
C Limiting the global impact on scenarios
C Failing to include an executive team in the process
C Treating scenarios as only an informational activity
C Limiting the imaginative stimulus into the scenario

design
C Failing to develop scenarios for key business

impact areas
CSSBB 2014

Hoshin Planning
Hoshin planning is an execution tool, that is used both
to organize and deploy strategic plans.
Hoshin planning translates the company’s vision into

dramatic measurable results and strategic
breakthroughs. The focus of hoshin planning is to
identify the vital few breakthrough achievements.
Hoshin planning has six basic objectives:
C Align the organizational goals

C Focus on the vital few strategic gaps
C Work with others to close the gaps
C Specify the methods to achieve the objectives
C Make the linkage among local plans visible
C Continuously improve the planning process
CSSBB 2014

Other Key Analysis Techniques

A number of other widely used techniques can also be
employed for gap analysis. In this Primer, there is
additional coverage of the following applicable topics:
C Benchmarking
C SWOT analysis
C PEST analysis
CSSBB 2014

Organizational Assessments
An organizational assessment typically consists of a
functional analysis, which includes data collection.
Examples of data collection techniques are:
C Face-to-face interviews
C Surveys of an appropriate sample
C Focus group inputs
C Observations from plant walk throughs
C Data gathered from industry sources
An assessment and analysis typically divides an

organization into key functional areas. A summary of
the important areas includes:
C Leadership
C Business practices
C Financial analysis
C Marketing
C Quality management
C Research & design
CSSBB 2014

Organizational Assessments (Cont’d)

C Manufacturing
C Safety & health
C Information technology
C Production engineering
C Production planning & scheduling
C Plant facilities
C Human resources
Obviously, variations in the above assessment areas

exist. The results of an assessment may be analyzed
and presented to management. For effective use of the
assessment, clear action plans must be developed and
implemented.
Typically, an assessor gathers the data and summarizes

it into major categories for review by the management
group. Thus, management will have ownership of the
solutions, and be committed to implement the solutions
and nurture them to success.
CSSBB 2014

Organizational Metrics
Organizational performance goals and corresponding
measurements are often established in the areas of:
C Profit
C Cycle times
C Resources
C Marketplace responses
A company should develop metrics for each major

performance goal. A unit and a method of measurement
must be defined. For the above performance goals,
possible metrics include:
Profit
C Stockholder value
C Capital investment
C Personnel costs
C Competitor comparisons
C Return on investment
C Sales dollars
Profit may be short-term or long-term.

CSSBB 2014

Organizational Metrics (Continued)

Cycle Times
C Existing cycle times

C Internal benchmarks
C External benchmarks
C Reductions in cycle times
Resources
C Number of improvement projects

C Return on improvement projects
C Process capability studies
C Variation reductions
C Costs of poor quality compared to some base
C Percent defects compared to some base
Marketplace Responses
C Market survey of customers

C Analysis of returns
C New product development
C Customer retention
C Customer losses
C Courtesy and facilities ratings
CSSBB 2014

Organizational Metrics (Continued)

Metrics can be and must be developed in order to
measure achievement of the organizational goals.
The device used to collect the data must be accurate.

Additionally, the questions of when, where, and how will
impact the accuracy and precision of data. According to
Juran, the development of any measurement system
should take into account the following factors:
C The metric should have a standard meaning

C The data should help the decision-making process
C It should provide worthwhile information
C It should be easy to install
C If worthwhile, it should be used elsewhere
In many applications, the appropriate metrics are based

on customer, supplier, or internal appraisal feedback.
CSSBB 2014

ADDITIONAL ANALYSIS METHODS/ROOT CAUSE ANALYSIS
Root Cause Analysis

An individual or team is often given the responsibility of
determining the root cause of a deficiency and
correcting it. The proposed action may take several
steps. Refer to the Table below:
Situation Immediate Intermediate Action Root Cause Action

Action
The dam Plug it Patch the dam Find out what caused the leak
leaks so it does not happen again.
Then rebuild the dam.
Parts are 100% Put an oversize Analyze the process and take
oversized inspection kickout device in action to eliminate the
line production of oversized parts.
CSSBB 2014

Root Cause Analysis (Continued)

To help locate the system’s true problem, a variety of
problem solving tools are available. Listed below are
commonly used techniques.
Subjective Tools Analytical Tools
Ask why 5 times Data collection

Brainstorming Data analysis
Process flow analysis Pareto analysis
Plan-do-check-act Regression analysis
Systematic methods Check sheets
Nominal technique Data matrix analysis
Operator observation Capability analysis
Fishbone diagrams Variation partitioning
Consensus techniques Subgrouping of data
Six thinking hats Simple trials
Use of teams DOE
FMEA Analytical tests
FTA Control charting
CSSBB 2014

Root Cause Analysis (Continued)

When permanent corrective action is proposed,
management must determine if:
C The root cause analysis has identified the full extent

of the problem
C The corrective action is satisfactory to eliminate or

prevent recurrence
C The corrective action is realistic and maintainable

CSSBB 2014

5 Whys
The 5 whys approach to root cause analysis is
described as asking the question “Why?” five times.
The following is an example of 5 whys. Symptom:
Customer shipments were late.
Why? We ran out of parts because the die stamping

press broke down.
Why? The press had not received scheduled

maintenance for three months.
Why? The maintenance department staff had been

reduced to six people from eight.
Why? The maintenance department was over budget.

The general manager eliminated all overtime
and fired 2 people.
Why? The company was not reaching profit goals so

the CEO had issued an order to avoid all
unnecessary spending. The root cause was the
CEO was worried about getting fired.
There is nothing magical about 5 whys. The root cause

may be found after 3 or 4 whys.
CSSBB 2014

5 Ws and H
The 5 Ws and H approach to root cause analysis is
summarized as asking the questions who?, what?,
when?, where?, why?, and how? Responses to these
questions can be organized with each element depicted
as a branch on a cause and effect (fishbone) diagram.
In some references, this same basic method is simply

referred to as the 5 Ws (who, what, where, when, and
why).
CSSBB 2014

Cause and Effect (Fishbone) Diagrams

Cause and effect diagrams are effective team based
tools to determine the potential root causes of a
problem.
A cause and effect diagram:
C Breaks problems down into bite-size pieces

C Displays many possible causes in a graphic manner
C Is also called a 4-M or Ishikawa diagram
C Shows how various causes interact
C Follows brainstorming rules when generating ideas
A fishbone session is divided into three parts:

brainstorming, prioritizing, and development of an
action plan. A team identifies the problem statement
and brainstorm the categories in a fishbone diagram. To
prioritize causes, polling is often used. The three most
probable causes may be identified for the development
of an action plan.
CSSBB 2014

Cause and Effect Diagrams (Continued)

Generally, the 4-M (manpower, material, method,
machine) version of the fishbone diagram will suffice.
Occasionally, an expanded version must be used. In a
laboratory environment, measurement is a key issue.
When discussing the brown grass in the lawn,
environment is important. A 5-M and E schematic is
shown below.
Machine Material Measurement
Problem
Statement
Method Manpower Environment

CSSBB 2014

Pareto Diagrams
Pareto diagrams are very specialized forms of column
graphs. They are used to prioritize problems (or
opportunities) so that the major problems (or
opportunities) can be identified.
The word “Pareto” comes from Vilfredo Pareto (1848-

1923). Pareto, an economist, made extensive studies
about the unequal distribution of wealth and formulated
mathematical models to quantify this maldistribution.
Dr. Joe M. Juran, world renowned leader in the quality

field, was preparing the Quality Control Handbook in the
late 1940s. He needed a short name to apply to the
phenomenon of the “vital few” and the “trivial many.”
He depicted some cumulative curves in this manuscript
with “Pareto’s principle of unequal distribution...”
CSSBB 2014

Pareto Diagrams (Continued)

Briefly stated, the Pareto principle suggests (in most
situations) that a few problem categories (approximately
20%) will present the most opportunity for improvement
(approximately 80%).
Pareto diagrams are used to:
C Analyze a problem from a new perspective

C Focus attention on problems in priority order
C Compare data changes during different time periods
C Permit the construction of a cumulative line
CSSBB 2014

Typical Pareto Diagram

Book defects are shown in Pareto form below:
100
300
Number of Binding Problems
75
200 Cumulative Line
Percent
50
100
25
0
A B C D E F G H I J K L M N
Problem Categories
The “all others” category is placed last. Cumulative

lines are convenient for answering such questions as,
“What defect classes constitute 70% of all defects?”
“First things first” is the thought behind the Pareto

diagram. There is an example of weighted Pareto
analysis in the Primer.
CSSBB 2014

Ford 8D Problem Solving

Ford Motor Company has developed a method of root
cause analysis for use in group problem solving. It is a
staged problem solving methodology. The 8D process
has the following steps:
D1. Establish the team

D2. Describe the problem
D3. Develop a containment action
D4. Identify the root cause
D5. Develop alternative solutions
D6. Implement a permanent corrective action
D7. Prevent recurrence
D8. Recognize team and individual contributions
CSSBB 2014

Fault Tree Analysis

Fault tree analysis (FTA) is a systematic, deductive
methodology for defining a single, specific, undesirable
event, and determining all possible reasons (failures)
that could cause that event to occur.
FTA is extremely useful during the initial product design

phase as an evaluation tool for driving preliminary
design modifications. Other potential uses of FTA
include:
C Functional analysis of highly complex systems

C Evaluation of subsystem events on the top event
C Evaluation of safety requirements
C Evaluation of system reliability
C Identification of design defects and safety hazards
C Evaluation of potential corrective actions
C Maintenance and troubleshooting simplification
C Logical elimination of causes of failure
CSSBB 2014

Fault Tree Analysis (Continued)

FTA is preferred over FMEA when:
C Safety of personnel is paramount

C A small number of top events can be identified
C A functional profile is of critical importance
C There is a high potential for failure
C The primary concern is a quantified risk evaluation
C Product functionality is highly complex
C The product is not repairable once initiated
FMEA is preferred over FTA when:
C Top events cannot be explicitly defined

C Multiple potentially successful profiles are feasible
C The identification of all failure modes is important
C Product functionally has little outside intervention
CSSBB 2014

FTA Symbols
Fault tree analysis uses the concepts of logic gates to
determine the overall reliability of a system. There are
numerous FTA symbols. These are broken down into
two main categories; event symbols and gate symbols.
Only a few examples are described in the Primer:
Event Symbols
or Top event: Contains a description of

a system-level fault or undesired
event.
Basic event: Usually the lowest level

of event fault that one wishes to
study.
Fault event: Contains a description

of a lower-level fault. It can receive
inputs from or provide outputs to a
logic gate.
CSSBB 2014

FTA Symbols (Continued)

Logic (Gate) Symbols
“and” gate (series)

The output event occurs only if all the
input events occur simultaneously.
“or” gate (parallel)

The output event occurs if any one of
the input events occur.
CSSBB 2014

Fault Tree Analysis Example

Fault tree analysis begins at the system level, and
assumes that a failure will occur. For instance, consider
the hypothetical probabilities for a home computer
failing to work as shown below:
COMPUTER FAILS
TO WORK
OR
0.0001 HARD
CPU KEYBOARD MONITOR
CALCULATED DRIVE
0.001 0.015 0.020 0.015

AND
CD ROM CD ROM
DRIVE D DRIVE E
0.010 0.010
CSSBB 2014

FTA Example (Continued)

The fault tree indicates that the computer will fail to
operate if CD ROM drive D and drive E fail, or the hard
drive fails, or the CPU fails, or the keyboard fails, or the
monitor fails. Since “and” gates multiply and “or” gates
add, the probability of the home computer not working
can be computed.
μsystem = 1 - (1 - 0.0001)(1 - 0.001)(1 - 0.015)(1 - 0.020)(1 - 0.015)
μsystem = 1 - (0.9999)(0.999)(0.985)(0.98)(0.985)
μsystem = 1 - 0.9498 = 0.0502
The probability of a failure is 5.02%
As a root cause analysis tool, this fault tree indicates

that the keyboard should receive priority attention (0.20)
followed by the CPU (0.015) and monitor (0.015). Other
Improvements can wait.
CSSBB 2014

ADDITIONAL ANALYSIS METHODS/WASTE ANALYSIS
Waste Analysis
Non-value added activities are classified as muda. This
term describes the waste that exists in a process. Imai
provides a list of seven waste categories that have been
widely used:
C Overproduction
C Inventory
C Repair/rejects
C Motion
C Transport
C Processing
C Waiting
CSSBB 2014

Overproduction
The muda of overproduction is producing too much at
a particular point in time. It is characterized by:
C Producing more than is needed by the next process

C Producing earlier than needed by the next process
C Producing faster than is needed by the next process
In the just-in-time environment, producing too early is as

bad as producing too late. Having the product too early,
too late, or in quantities that are too great will result in
undesirable consequences, such as:
C Extra space used at the customer’s plant

C Extra space used at the organization’s plant
C Extra raw materials in use
C Extra utilities used
C Extra transportation for all parties
C Extra scheduling costs
CSSBB 2014

Excess Inventory
Parts, raw materials, work-in-process, inventory,
supplies, and finished goods are all forms of inventory.
Excess inventory is considered muda since it does not
add value to the product. Inventory will:
C Require space in the shop

C Require transportation
C Require forklifts
C Require conveyor systems
C Require additional labor
C Require interest on material costs
Inventory sitting around in various process stages can

be adversely affected in the following ways:
C It may gather dust

C It can deteriorate
C It can become obsolete
C It may get wet
C It can experience handling damage
CSSBB 2014

Repair/ Rejects
The repair or rework of defective parts involves a
second attempt at producing a good item. Scrapping
the whole part is a definite waste of resources. Having
rejects on a continuous flow line defeats the purpose of
continuous flow. Line operators and maintenance will
be used to correct problems, putting the takt time off
course. Rework may require nonconforming product
forms to be filled out generating additional muda.
Various design changes can be muda also. Design

changes are often considered rework or an extra
development effort. Both of these activities will create
the need for additional labor.
CSSBB 2014

Motion
Extra unneeded human motions are wasteful. Operators
should not have to walk excessively, lift heavy loads,
bend awkwardly, reach too far, repeat motions, etc. The
layout of the workplace should be designed to take
advantage of proper ergonomics. Each work station
should be analyzed for motion requirements.
Ergonomics can eliminate factors that can cause

injuries and lost production. Some guidelines for
providing sound ergonomic principles in the workplace
include:
C Emphasize safety at all times

C Fit the employee to the job
C Change the workplace to fit the employee
C Maintain neutral body positions
C Redesign tool handles to reduce stress and injury
C Vary the tasks through job rotation
C Make the machine serve the human
CSSBB 2014

Processing
Processing muda consists of additional steps or
activities in the manufacturing process. Examples
include the following:
C Having to remove burrs from a product

C Reshaping a piece due to poor dies
C Adding an extra handling process
C Performing an inspection step
C Repeating product changes that are unnecessary
C Maintaining extra copies of information
Transport
All forms of transportation are muda. This describes the
use of forklifts, conveyors, pallet movers, and trucks.
This can be caused by poor plant layouts, poor cell
designs, use of batch processing, long lead times, large
storage areas, or scheduling problems. Transport
should be eliminated whenever possible.
CSSBB 2014

Waiting
The muda of waiting occurs when an operator is ready
for the next operation, but must remain idle. The
operator is idle due to machine downtime, lack of parts,
unwarranted monitoring activities, or line stoppages. A
maintenance operator waiting at a tool bin for a part is
muda. The muda of waiting can be characterized by:
C Idle operators
C Breakdowns in machinery
C Long changeover times
C Uneven scheduling of work
C Batch material flows
C Long and unnecessary meetings
CSSBB 2014

Additional Classes of Wastes

In addition to the seven waste categories previously
listed, some other sources of waste include:
C Misused resources
C Underutilized resources
C Counting activities
C Looking for tools or parts
C Multiple systems
C Multiple hand-offs
C Unnecessary approvals
C Machine breakdowns
C Sending bad product to customers
C Providing bad service to customers
The classes of wastes can also be differentiated

between production, administrative, and transactional
activities. Other varieties of waste exist.
CSSBB 2014
VIII. ANALYZE
QUESTIONS
8.2. Failure modes and effects analysis involves what activity?
a. The determination of the probability of failure in a specified period

of time
b. The expected number of failures in a given time interval
c. The study of how a product fails and what causes the failure
d. A study of the probability of success in a given time period
8.5. Identify a non-value added activity that would be considered a form

of processing muda:
a. Any and all forms of transportation

b. An ergonomically unsound workplace
c. Reshaping a product due to poor dies
d. Producing more than needed by the customer
8.8. When finding a confidence interval for mean  based on a sample

size of n:
a. Increasing n increases the interval

b. Having to use Sx instead of n decreases the interval
c. The larger the interval, the better the estimate of μ
d. Increasing n decreases the interval
Answers 8.2 c, 8.5 c, 8.8 d

CSSBB 2014
VIII. ANALYZE
QUESTIONS
8.12. Which of the following is a valid null hypothesis?
a. p > 1/8
b. mu < 98
c. The mean of population A is not equal to the mean of population B
d. mu = 110
8.14. Assume that data for a goodness-of-fit test has been structured
into effective cells for a chi square calculation. Which of the
following distributions loses the most degrees of freedom?
a. Normal
b. Poisson
c. Binomial
d. Uniform
8.20. A dendogram, as used in cluster analysis, can be described as a:
a. Classification tree
b. Pareto diagram
c. Decision tree
d. Line diagram
Answers 8.12 d, 8.14 a, 8.20 a

CSSBB 2014
VIII. ANALYZE
QUESTIONS
8.21. An assessment technique to compare a company’s current

performance to a desired or potential performance is called:
a. Benchmarking
b. SWOT analysis
c. Gap analysis
d. Marketplace response
8.26. When a problem solving team applies the 5 whys technique, they
are attempting to:
a. Determine if the interviewee is telling the truth

b. Understand the basics of the problem
c. Eliminate areas not to investigate
d. Determine the root cause of the problem
8.29. Consider the SS and MS columns of an analysis of variance table

for a single factor design. The appropriate ratio for testing the null
hypothesis of no treatment effect is:
a. SS treatments divided by SS residual

b. MS treatments divided by MS residual
c. SS treatments divided by MS residual
d. MS treatments divided by SS residual
Answers 8.21 c, 8.26 d, 8.29 b

CSSBB 2014
VIII. ANALYZE
QUESTIONS
8.34. A small change truly exists between a current process and an

experimental process. However, an insufficient sample size is
collected to determine this difference. What type of error would
result?
a. A type II error
b. A 1-alpha decision
c. A type I error
d. A 1-beta decision
8.35. Identify the subjective root cause analysis tool from the options
below:
a. 5 whys
b. Pareto analysis
c. Control charts
d. Regression analysis
8.36. Why should an experimenter plot data points and graph the least
squares line if a probabilistic regression model exists?
a. To visually present the relationship to others

b. To check for fit; there may be a calculation error
c. There’s no rational reason for doing so
d. To permit a projection outside of the test area
Answers 8.34 a, 8.35 a, 8.36 b

CSSBB 2014
VIII. ANALYZE
QUESTIONS
8.43. If an investigator were interested in performing a chi-square

hypothesis test of the equality of measurement data, which of the
following would be an option?
a. Runs test
b. Mood's Median test
c. Mann Whitney test
d. Kruskal-Wallis test
8.45. The alpha critical region, to determine if a new pen lasts longer,
would be placed:
a. In the upper tail

b. In both upper and lower tails
c. In the lower tail
d. In neither tail
8.47. The probabilistic regression model for any particular observed

value of Y contains a term o, which represents:
a. The Y axis intercept, when X = 0

b. The Y axis intercept, when X = 1
c. The slope of the model
d. The X axis intercept, when Y = 0
Answers 8.43 b, 8.45 c, 8.47 c

CSSBB 2014
VIII. ANALYZE
QUESTIONS
8.51. A product was yielding 90% recovery before an improvement was

made. To determine if a 2% change (in either direction) has been
made at the 95% confidence level, what sample size should be
taken?
a. 468
b. 648
c. 864
d. 1,728
8.54. From a gap analysis perspective, why does an organization need

performance metrics?
a. To provide information for goal achievement

b. To indicate the number of and return on improvement projects
c. To determine the best internal and external benchmarks
d. To measure customer retention and loyalty
8.59. If two sigma limits are substituted for conventional three sigma
limits on a control chart, which of the following will occur?
a. Alpha risk decreases

b. Beta risk increases
c. Alpha risk increases
d. Sample size increases
Answers 8.51 c, 8.54 a, 8.59 c

© QUALITY COUNCIL OF INDIANA IX-1 (813)
CSSBB 2014
IX. IMPROVE
THE STARTING POINT FOR

IMPROVEMENT IS TO
RECOGNIZE THE NEED.
IMAI
CSSBB 2014
IX. IMPROVE VII. EXTRA

DOE/INTRODUCTION
Improve
Improve is presented in the following topic areas:
C Design of experiments (DOE)

C Lean methods
C Implementation
Design of Experiments
DOE is discussed in the following topic areas:
C Introduction
C Terminology
C Design principles
C Planning experiments
C One factor designs
C Fractional factorial designs
C Full factorial designs
CSSBB 2014

DOE/INTRODUCTION
DOE Introduction
Classical experiments focus on 1FAT (one factor at a
time) at two or three levels and attempt to hold
everything else constant. DOE can focus on a wide
range of key input factors or variables and will
determine the optimum levels of each of the factors.
The Pareto principle applies to the world of
experimentation. 20% of the input factors generally
make 80% of the impact on the result. The classical
approach to experimentation, changing just one factor
at a time, has shortcomings:
C Too many experiments are necessary.
C The optimum combination of all the variables may

never be revealed.
C The interaction between factors cannot be

determined.
C Conclusions may be wrong or misleading.
C Many of the observed effects tend to be mysterious.
C Time and effort may be wasted by studying the

wrong variables or obtaining improper data.
CSSBB 2014

DOE/INTRODUCTION
DOE Introduction (Continued)

Design of experiments overcomes these problems by
careful planning. Advantages of DOE include:
C Many factors can be evaluated simultaneously,

making the DOE process economical.
C Input factors can be controlled to make the output

insensitive to noise factors.
C In-depth, statistical knowledge is not always

necessary to get big benefit.
C One can look at a process with few experiments.
C Since the designs are balanced, there is confidence

in the conclusions.
C If important factors are overlooked in an

experiment, the results will indicate that fact.
C Precise statistical analysis can be run using

standard computer programs.
C Results can be improved with low costs.

CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
DOE Terms
Alias An alias occurs when two factor effects

are confounded with each other.
Balanced A fractional factorial design, in which an
design equal number of trials (at every level
state) is conducted for each factor.
Block A subdivision of the experiment into
relatively homogenous.
Blocking When structuring fractional factorial
experimental test trials, blocking is used
to account for variables that the
experimenter wishes to avoid.
Collinear A collinear condition occurs when two
variables are totally correlated.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
DOE Terms (Continued)
Confounded Confounding exists when the effects of

two factors are not separable. In the
following example, A, B, and C are input
factors and columns AB, AC, & BC
represent interactions. Confounded
columns are identified by arrows,
indicating the setting of one cannot be
separated from the setting of the other.
A B C AB AC BC A B C AB AC BC
- - + + - - + + - + - -
- + - - + - + - + - + -
+ - - or
- - + - + + - - +
+ + + + + + - - - + + +
A is confounded with BC, B with AC, and C with AB
Correlation A number between -1 and 1 that indicates

coefficient the degree of linear relationship between
(r) two sets of numbers. Zero (0) indicates
no linear relationship.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
Covariates Things which change during an

experiment which had not been planned
to change.
Curvature Refers to non-straight line behavior
between one or more factors and the
response. Curvature is usually expressed
in mathematical terms involving the
square or cube of the factor. For
example, in the model:
Y = B0 + B1X1 + B11 (X1 C X1) + ε

B11 (X1 C X1) describes curvature.
Degrees of The terms used are DOF, DF, df or ν. The
freedom number of measurements that are
independently available for estimating a
population parameter.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
Efficiency One estimator is more efficient than

another if it had a smaller variance.
EVOP Stands for evolutionary operation, a term

that describes the way sequential
experimental designs can be made to
adapt to system behavior by learning
from present results and predicting
future treatments for better responses.
Experiment A test undertaken to make an
improvement in a process or to learn
previously unknown information.
Experiment Variation in response or outcome of
error virtually identical test conditions. This is
also called residual error.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
First-order Refers to the power to which a factor

appears in a model. If “X1” represents a
factor and “B” is its factor effect, then
the model:
Y = B0 + B1X1 + B2X2 + ε
is first-order in both X1 and X2. First-

order models cannot account for
curvature or interaction.
Fractional An adjective that means fewer
experiments than the full design calls for.
Full Describes experimental designs which
factorial contain all combinations of all levels of
all factors. No possible treatment
combinations are omitted.
Input factor An independent variable which may
affect a (dependent) response variable
and is included at different levels in the
experiment.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
Interaction An interaction occurs when the effect of

one input factor on the output depends
upon the level of another input factor.
Refer to the following diagrams:
No Moderate Strong Strong
Interaction Interaction Interaction Interaction
A Low
A High
L H L H L H L H
Factor B Factor B Factor B Factor B
Interactions can be readily examined with

full factorial experiments. Often,
interactions are lost with fractional
factorial experiments.
Level A given factor or a specific setting of an
input factor. Four levels of a heat
treatment may be 100EF, 120EF, 140EF,
and 160EF.
Main effect An estimate of the effect of a factor
independent of any other factors.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
Multi- This occurs when two or more input

collinearity factors are expected to independently
affect the value of an output factor, but
are found to be highly correlated.
Nested An experimental design in which all trials
trials are not fully randomized. There is
generally a logical reason for taking this
action.
Orthogonal A design is orthogonal if the main and
interaction effects in a given design can
be estimated without confounding the
other main effects or interactions.
Parallel These experiments are done at the same
tests time, not one after another. Parallel
experimentation is the opposite of
sequential experimentation.
Precision The closeness of agreement between test
results.
Qualitative This refers to descriptors of category
and/or order, but not of interval or origin.
Different machines, operators, materials,
etc. represent qualitative levels or
treatments.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
Quanti- This refers to descriptors of order and

tative interval (interval scale) and possibly also
of origin (ratio scale). As a quantitative
factor, “temperature” might describe the
interval value 27.32EC. As a quantitative
response, “yield” might describe the ratio
value 87.42%.
Random Frees an experiment from the
trials environment and eliminates biases. This
technique avoids the undue influences of
systematic changes that are known or
unknown.
Repeated Trials that are conducted to estimate the
trials pure trial-to-trial experimental error so
that lack of fit may be judged. Also called
replications.
Residual The difference between the observed and
error the predicted value for that result, based
(ε) or (E) on an empirically determined model. It
can be variation in outcomes of virtually
identical test conditions.
Residuals The difference between experimental
responses and predicted model values.
CSSBB 2014
IX. IMPROVE VII.A.1

DOE/TERMINOLOGY
Response The variable that shows the observed

variable results of an experimental treatment.
Also known as the output or dependent
variable.
Screening A technique to discover the most
experiment (probable) important factors in an
experimental system. Most screening
experiments employ two-level designs.
Sequential Experiments are done one after another,
tests not at the same time. This is often
required by the type of experimental
design being used. Sequential
experimentation is the opposite of parallel
experimentation.
Test The percentage of all possible
coverage combinations of input factors in an
experimental test.
Treatments In an experiment, the various factor levels
that describe how an experiment is to be
carried out. A pH level of 3 and a
temperature level of 37E Celsius describe
an experimental treatment.
CSSBB 2014
IX. IMPROVE VII.A.2

DOE/DESIGN PRINCIPLES
Applications of DOE
Situations, where experimental design can be effectively
used include:
C Choosing between alternatives

C Selecting the key factors affecting a response
C Response surface modeling to:
C Hit a target
C Reduce variability
C Maximize or minimize a response
C Make a process robust
C Seek multiple goals
DOE Steps
Getting good results from a DOE involves a number of
steps:
C Set objectives
C Select process variables
C Select an experimental design
C Execute the design
C Ensure the data is consistent with the assumptions
C Analyze and interpret the results
C Use/present the results (may lead to further runs)
CSSBB 2014
IX. IMPROVE VII.A.2

DOE Steps
Important practical considerations in planning and
running experiments are:
C Check the performance of measuring devices

C Keep the experiment as simple as possible
C Check that all planned runs are feasible
C Watch for process drifts and shifts during the run
C Avoid unplanned changes
C Allow some time for unexpected events
C Obtain buy-in from all parties involved
C Maintain ownership of each experimental step
C Preserve all the raw data - not just averages!
C Record everything that happens
C Reset equipment after experimentation
CSSBB 2014
IX. IMPROVE VII.A.2

Experimental Objectives
Choosing an experimental design depends on the
objectives of the experiment and the number of factors
to be investigated. Experimental design objectives
include the following:
C Comparative objective
C Screening objective
C Response surface objective
C Optimizing mixture responses
C Optimal fitting of a regression model
CSSBB 2014
IX. IMPROVE VII.A.2

Select and Scale the Process Variables

Process variables include both inputs and outputs, i. e.
factors and responses. The selection of these variables
is best done as a team effort. The team should:
C Include all important factors (based on judgment)

C Be bold, but not foolish, in choosing factor levels
C Avoid impractical or impossible factor settings
C Include all relevant responses
C Avoid using combine responses
When choosing the range of settings for input factors,

it is wise to avoid extreme values. In some cases,
extreme values will give runs that are not feasible; in
other cases, extreme ranges might move the response
surface into some erratic region.
CSSBB 2014
IX. IMPROVE VII.A.2

Design Guidelines
Number Comparative Screening Objective

of Factors Objective
1 1-factor completely ____
randomized design
2-4 Randomized block Full or fractional
design factorial
5 or more Randomized block Fractional factorial
design or Plackett-Burman
It is a good idea to choose a design that requires

somewhat fewer runs than the budget permits, so that
additional runs can be added to check for curvature and
to correct any experimental mishaps.
CSSBB 2014
IX. IMPROVE VII.A.3

DOE/PLANNING EXPERIMENTS
A Typical DOE Checklist

Every experimental investigation will differ in detail, but
the following checklist will be helpful.
C Define the objective of the experiment.
C The principle experimenter should learn as many

facts about the process as possible.
C Brainstorm a list of the key independent and

dependent variables with knowledgeable people.
C Run “dabbling experiments” to debug equipment or

determine measurement capability.
C Assign levels to each independent variable in the

light of all available knowledge.
C Select a standard DOE plan or customize one.
C Run the experiments in random order.
C Draw conclusions and verify by replication.
An iterative approach is usually the most economical.

CSSBB 2014
IX. IMPROVE VII.A.3

Experimental Assumptions
In all experimentation, one makes assumptions. Some
of the engineering and mathematical assumptions an
experimenter can make include:
C Is the measurement system capable?

C Is the process stable?
C Are the residuals well behaved?
Is the Measurement System Capable?
It is not a good idea to find, after finishing an

experiment, that the measurement devices are
incapable. This should be confirmed before embarking
on the experiment itself.
Is the Process Stable?
Experimental runs should have control runs which are

done at the “standard” process set points, or at least at
some identifiable operating conditions. The experiment
should start and end with such runs.
CSSBB 2014
IX. IMPROVE VII.A.3

Experimental Assumptions (Continued)

Are the Residuals Well Behaved?
Residuals are estimates of experimental error obtained

by subtracting the observed response from the
predicted response. The predicted response is
calculated from the chosen model, after all the unknown
model parameters have been estimated from the
experimental data.
The overall pattern of the residuals should be similar to

the bell-shaped pattern observed when plotting a
histogram of normally distributed data. Graphical
methods are used to examine residuals. Departures
from assumptions usually mean that the residuals
contain structure that is not accounted for in the model.
Any graph suitable for displaying the distribution of a

set of data is suitable for judging the normality of the
distribution of a group of residuals. The three most
common types are: histograms, normal probability
plots, and dot plots.
CSSBB 2014
IX. IMPROVE VII.A.3

Interaction Case Study

An interaction case study is presented in the Primer.
The graphical results are plotted below.
9
8.5
8
RH 37%
7.5 RH 41%
7
RH RH Low High Low High
37% 41% Ion Ion Ion Ion
The interaction plot demonstrates that if the goal is to

reduce breaks, an economic choice could be made
between low ion/low RH and high ion/high RH.
CSSBB 2014
IX. IMPROVE VII.A.4

DOE/ONE FACTOR DESIGNS
Randomized Block Plans

In comparing a number of factor treatments, it is
desirable that all other conditions be kept as nearly
constant as possible. The required number of tests may
be too large to be carried out under similar conditions.
In such cases, one may be able to divide the experiment
into blocks, or planned homogeneous groups. A
randomized block design for air permeability response
is shown below:
Fabric Types I, II, III, & IV
I II III IV
B(15.1) D(11.6) A(15.4) C(9.9)
Chemical C(12.2) C(13.1) B(16.3) D(9.4)
Applications
A, B, C, D A(19.0) B(17.6) D(16.0) B(8.6)
D(11.5) A(13.0) C(10.8) A(11.5)
CSSBB 2014
IX. IMPROVE VII.A.4

Randomized Block Plans (Continued)

A randomized incomplete block (tension response)
design is shown below.
Treatment
Block (Days) A B C D
1 -5 Omitted -18 -10
2 Omitted -27 -14 -5
3 -4 -14 -23 Omitted
4 -1 -22 Omitted -12
Only treatments A, C, and D are run on the first day. B,

C, and D on the second, etc. In the whole experiment,
note that each pair of treatments, such as BC, occur
twice together. The order in which the three treatments
are run on a given day follows a randomized sequence.
Blocking factors are commonly environmental

phenomena outside of the control of the experimenter.
CSSBB 2014
IX. IMPROVE VII.A.4

Latin Square Designs

A Latin square design is called a one-factor design
because it attempts to measure the effects of a single
key input factor on an output factor. The experiment
further attempts to block (or average) the effects of two
or more nuisance factors. Such designs were originally
applied in agriculture. The square was literally a plot of
ground.
In Latin square designs, a third variable, the

experimental treatment, is then applied to the source
variables in a balanced fashion. The Latin square plan
is restricted by two conditions:
C The number of rows, columns, and treatments must

be the same.
C There should be no expected interactions between

row and column factors, since these cannot be
measured.
A Latin square design is essentially a fractional factorial

experiment which requires less experimentation to
determine the main treatment results.
CSSBB 2014
IX. IMPROVE VII.A.4

Latin Square Designs (Continued)

Consider the following 5 x 5 Latin square design:
Carburetor Type
Driver I II III IV V
1 A B C D E
2 B C D E A
3 C D E A B
4 D E A B C
5 E A B C D
In the above design, five drivers and five carburetors

were used to evaluate gas mileage from five cars. Only
twenty-five of the potential 125 combinations are tested.
Thus, the resultant experiment is a one-fifth fractional
factorial. Similar 3 x 3, 4 x 4, and 6 x 6 designs may be
utilized.
In some situations, what is thought to be a nuisance

factor can end up being very important.
CSSBB 2014
IX. IMPROVE VII.A.4

Graeco-Latin Designs
Graeco-Latin square designs are sometimes useful to
eliminate more than two sources of variability in an
experiment. A Graeco-Latin design is an extension of
the Latin square design, but one extra blocking variable
is added for a total of three blocking variables.
Consider the following 4 X 4 Graeco-Latin design:
Carburetor Type
Driver I II III IV Cars
1 Aα Bβ Cγ Dδ A,B,C,D
2 Bδ Aγ Dβ Cα
3 Cβ Dα Aδ Bγ Days
4 Dγ Cδ Bα Aβ α,β,γ,δ
The output (response) variable could be gas mileage for

the 4 cars (A, B, C, D).
Hyper-Graeco-Latin Designs
A hyper-Graeco-Latin square design permits the study
of treatments with more than three blocking variables.

CSSBB 2014
IX. IMPROVE VII.A.5

DOE/FRACTIONAL FACTORIAL DESIGNS
Fractional Factorial Experiments

The ASQ BOK refers to two-level fractional factorial
designs. The authors have chosen to include one three-
level example and an explanation of EVOP.
Two-Level Fractional Factorial Example

The basic steps for a two-level fractional factorial design
will be examined via the following example. The
following seven step procedure will be followed:
1. Select a process
2. Identify the output factors of concern
3. Identify the input factors and levels
4. Select a design
5. Conduct the experiment under desired conditions
6. Collect the data (relative to the identified outputs)
7. Analyze the data and draw conclusions
CSSBB 2014
IX. IMPROVE VII.A.5

CSSBB Test Success

Step 1: Select a process
Our ASQ Section wants to investigate CSSBB

exam success using students of comparable
educational levels.
Step 2: Identify the output factors
Student performance will be based on two results

(output factors):
(1) Did they pass the test?

(2) What grade score did they receive?
Step 3: Establish the input factors and levels to be

investigated
Our ASQ Section wants to study the effect of

seven variables at two-level that they suspect
may effect CSSBB student performance.
CSSBB 2014
IX. IMPROVE VII.A.5

CSSBB Test Success (Continued)

Step 3: (continued) (7 factors at 2-levels)
Input Factors Level 1 (-) Level 2 (+)

A. Refresher Course No Yes
B. Study Time Morning Afternoon
C. Problems Worked 200 800
D. Primary Reference Book A Book B
E. Method Of Study Sequential Random
F. Work Experience (Time) 4 Years 12 Years +
G. Duration Of Study 50 Hours 135 Hours
Note: The above inputs are both variable (quantitative)

and attribute (qualitative).
CSSBB 2014
IX. IMPROVE VII.A.5


Step 4: Select a design
A screening plan is selected from a design catalogue.

Only eight (8) tests are needed to evaluate the main
effects of all 7 factors at 2-levels. The design is:
Input Factors
Test A B C D E F G
#1 - - - - - - -
#2 - - - + + + +
#3 - + + - - + +
#4 - + + + + - -
#5 + - + - + - +
#6 + - + + - + -
#7 + + - - + + -
#8 + + - + - - +
CSSBB 2014
IX. IMPROVE VII.A.5


Step 5: Conduct the experiment
Step 6: Collect the data
Input Factors Outputs

(Test Results)
Test A B C D E F G Pass Score
#1 - - - - - - - no (-) 31
#2 - - - + + + + no (-) 59
#3 - + + - - + + yes (+) 74
#4 - + + + + - - no (-) 56
#5 + - + - + - + yes (+) 87
#6 + - + + - + - no (-) 69
#7 + + - - + + - no (-) 44
#8 + + - + - - + yes (+) 72
Avg 61.5
Step 7: Analyze the data and draw conclusions
The pass/fail pattern of (+)s and (-)s does not track with
any single input factor. It visually appears that there is
some correlation with factors C and G.
CSSBB 2014
IX. IMPROVE VII.A.5


What Factors Are Critical to Passing?
Input Factors
A B C D E F G
Test #1 -31 -31 -31 -31 -31 -31 -31
Test #2 -59 -59 -59 59 59 59 59
Test #3 -74 74 74 -74 -74 74 74
Test #4 -56 56 56 56 56 -56 -56
Test #5 87 -87 87 -87 87 -87 87
Test #6 69 -69 69 69 -69 69 -69
Test #7 44 44 -44 -44 44 44 -44
Test #8 72 72 -72 72 -72 -72 72
(∆) 52 0 80 20 0 0 92
(∆ ÷ 4) 13 0 20 5 0 0 23
(+) means level 2 has a positive effect. (-) means level 2

has a negative effect. 0 means level 2 has no effect.
CSSBB 2014
IX. IMPROVE VII.A.5


(continued)
C Factor A, taking refresher course, will improve the

exam results by 13 points
C Factor B, study time of day, has no effect
C Factor C, problems worked, will improve the exam

results by 20 points
C Factor D, primary reference, will improve the exam

results by 5 points
C Factor E, method of study, has no effect
C Factor F, work experience, has no effect
C Factor G, duration of study, will improve the exam

results by 23 points
CSSBB 2014
IX. IMPROVE VII.A.5


(continued)
To calculate the optimum student performance:
1. Sum the arithmetic value of the significant

differences (∆) and divide the total by two. Note
that the absolute value is divided by 2 because the
experiment is conducted in the middle of the high
and low levels and only one-half the difference (∆)
can be achieved.
A B C D E F G
13 + 0 + 20 + 5 + 0 + 0 + 23 = 61
Improvement = 61 ÷ 2 = 30.5. There were no

significant negative effects (-) in this experiment. If
there were, they would have been included (added)
in determining the total effect. In this particular
DOE format, the sign indicates direction only.
2. Average the test scores.
Average = 61.5
CSSBB 2014
IX. IMPROVE VII.A.5


(continued)
3. Add the improvement to the average to predict the

optimum performance.
Optimum = Average + Improvement

= 61.5 + 30.5
= 92
The optimum performance would be obtained by

running the following trial:
A B C D E F G
+ 0 + + 0 0 +
+ = Level 2 - = Level 1 0 = Doesn’t matter
The above trial was one of the 120 tests not performed
out of 128 possible choices. DOE is almost magical.
Obviously, the predicted student scores can be
confirmed by additional experimentation.
CSSBB 2014
IX. IMPROVE VII.A.5


70
60
50
-A +A -B +B -C +C -D +D -E +E -F +F -G +G
Factor Levels
Plot of CSSBB Factor Levels vs. Test Scores

CSSBB 2014
IX. IMPROVE VII.A.5

Plackett-Burman Designs
Plackett-Burman designs are used for screening
experiments. Plackett-Burman designs are very
economical. The run number is a multiple of four rather
than a power of 2.
Plackett-Burman geometric designs are two-level

designs with 4, 8, 16, 32, 64, and 128 runs and work best
as screening designs. Each interaction effect is
confounded with exactly one main effect.
All other two-level Plackett-Burman designs (12, 20, 24,

28, etc.) are non-geometric designs. In these designs a
two-factor interaction will be partially confounded with
each of the other main effects in the study. Thus, the
non-geometric designs are essentially “main effect
designs,” when there is reason to believe that any
interactions are of little significance. A Plackett-Burman
design in 12 runs may be used to conduct an experiment
containing up to 11 factors.
CSSBB 2014
IX. IMPROVE VII.A.5

Plackett-Burman Designs (Continued)

Plackett-Burman non-geometric design (12 runs/11
factors):
Factors
Ex X1 X2 X3 X4 X5 X6 X7 X8 X9 X1 X1 Result
1 + + + + + + + + + + +
2 - + - + + + - - - + -
3 - - + - + + + - - - +
4 + - - + - + + + - - -
5 - + - - + - + + + - -
6 - - + - - + - + + + -
7 - - - + - - + - + + +
8 + - - - + - - + - + +
9 + + - - - + - - + - +
10 + + + - - - + - - + -
11 - + + + - - - + - - +
12 + - + + + - - - + - -
With a 20-run design, an experimenter can do a

screening experiment for up to 19 factors. As many as
27 factors can be evaluated in a 28-run design.
CSSBB 2014
IX. IMPROVE VII.A.5

A Design from a Design Catalogue

The preferred DOE approach examines (screens) a large
number of factors with highly fractional experiments.
Interactions are then explored or additional levels
examined once the factors have been reduced.
Factors
Test A B C D E F G
1 0 0 0 0 0 0 0
2 0 0 0 1 1 1 0
3 0 0 1 0 1 1 1
4 0 0 1 1 0 0 1
5 0 1 0 0 0 1 1
6 0 1 0 1 1 0 1
7 0 1 1 0 1 0 0
8 0 1 1 1 0 1 0
9 1 0 0 0 1 0 1
10 1 0 0 1 0 1 1
11 1 0 1 0 0 1 0
12 1 0 1 1 1 0 0
13 1 1 0 0 1 1 0
14 1 1 0 1 0 0 0
15 1 1 1 0 0 0 1
16 1 1 1 1 1 1 1
Note: 0 = low level and 1 = high level.

CSSBB 2014
IX. IMPROVE VII.A.5

Three-Factor, Three-Level Experiment

Often, a three-factor experiment is required after
screening a large number of variables. These
experiments may be full or fractional factorial. A one-
third fractional factorial design is shown below.
Generally the (-) and (+) levels in two-level designs are
expressed as 0 and 1 in most design catalogues. Three-
level designs are often represented as 0, 1, and 2.
Concentration ( 222 )
( 200 )
( 122 )
( 100 )
Pressure
( 022 )
( 012 )
Temperature
( 000 ) ( 001 ) ( 002 )
CSSBB 2014
IX. IMPROVE VII.A.5

Three-Factor, Three-Level (Continued)

From a design catalogue test plan, the selected
fractional factorial experiment looks like so:
Experiment Concentration Pressure Temperature

1 0 0 0
2 0 1 2
3 0 2 1
4 1 0 1
5 1 1 0
6 1 2 2
7 2 0 2
8 2 1 1
9 2 2 0
CSSBB 2014
IX. IMPROVE VII.A.5

EVOP Evolutionary Operations

EVOP (evolutionary operations) emphasizes a
conservative experimental strategy for continuous
process improvement. Refer to the Figure below:
91% 94%
E E
69% 83% 96%

B B D
71% 79% 88%

pH
A A C
E
92%
69% B D
A 87% 88%
A A C
63% 70% 84%
Concentration
Tests are carried out in phase A until a response pattern

is established. Then phase B is centered on the best
conditions from phase A. This procedure is repeated
until the best result is determined. When nearing a
peak, the experimenter will then switch to smaller step
sizes or will examine different variables. EVOP can
entail small incremental changes so that little or no
process scrap is generated. Large sample sizes may be
required to determine the direction of improvement.
CSSBB 2014
IX. IMPROVE VII.A.6

DOE/FULL FACTORIAL DESIGNS
A Full Factorial Example

Suppose that pressure, temperature, and concentration
are three suspected key variables affecting the yield of
a chemical process which is currently running at 64%.
An experimenter may wish to fix these variables at two-
levels (high and low) to see how they influence yield. In
order to find out the effect of all three factors and their
interactions, a total of 2 x 2 x 2 = 23 = 8 experiments
must be conducted. This is called a full factorial
experiment. The low and high levels of input factors are
noted below by (-) and (+).
Experiment No. Temperature Pressure Concentration % Yield

1 - - - 55
2 + - - 77
3 - + - 47
4 + + - 73
5 - - + 56
6 + - + 80
7 - + + 51
8 + + + 73
Average 64
Temperature: (-) = 120EC (+) = 150EC

Pressure: (-) = 10 psi (+) = 14 psi
Concentration: (-) = 10N (+) = 12N
CSSBB 2014
IX. IMPROVE VII.A.6

A Full Factorial Example (Continued)

To find the effect of temperature, sum the yield values
when the temperature is high and subtract the sum of
yields when the temperature is low, dividing the results
by four.
 77 + 73 + 80 + 73  -  55 + 47 + 56 + 51
Temperature effect: =
4
Temperature effect: = 23.5
When the temperature is set at a high level rather than

at a low level, one gains 23.5% yield. All of this
improvement can be attributable to temperature alone.
 47 + 73 + 51 + 73  -  55 + 77 + 56 + 80 
Pressure effect: =
4
Pressure effect: = - 6
The effect of changing pressure from low level to high

level is a loss of 6% yield.
 56 + 80 + 51 + 73  -  55 + 77 + 47 + 73 
Concentration effect: =
4
Concentration effect: = 2
Higher concentration levels results in a relatively minor

2% improvement in yield.
CSSBB 2014
IX. IMPROVE VII.A.6


The interaction effects between the factors can be
checked by using the T, P, and C columns to generate
the interaction columns by the multiplication of signs:
Interactions
EXP. T P C TXP PXC TXC TXPXC YIELD
1 - - - + + + - 55
2 + - - - + - + 77
3 - + - - - + + 47
4 + + - + - - - 73
5 - - + + - - + 56
6 + - + - - + - 80
7 - + + - + - - 51
8 + + + + + + + 73
CSSBB 2014
IX. IMPROVE VII.A.6


Following the same principles used for the main effects:
 55 + 73 + 56 + 73  -  77 + 47 + 80 + 51
T x P interaction =
4
T x P interaction = 0.5
The T x P interaction shows a marginal gain in yield.
 55 + 77 + 51 + 73  -  47 + 73 + 56 + 80 
PxC interaction =
4
PxC interaction = 0
 55 + 47 + 80 + 73  -  77 + 73 + 56 + 51
TxC interaction =
4
TxC interaction = - 0.5
 77 + 47 + 56 + 73  -  55 + 73 + 80 + 51
TxPxC interaction =
4
TxPxC interaction = - 1.5
In this example, the interactions have either zero or

minimal negative yield effects. If the interactions are
significant compared to the main effects, they must be
considered before choosing the final.
CSSBB 2014
IX. IMPROVE VII.A.6


Assume the following balanced design is chosen.
Experiment T P C Yield
2 + - - 77
3 - + - 47
5 - - + 56
8 + + + 73
Since a fractional factorial experiment is being

conducted, only the main effects of factors can be
determined. Please note that experiments 1, 4, 6, and 7
would have been equally valid.
The temperature effect =

 77 + 73  -  47 + 56 
= 23.5
2
The pressure effect =

 47 + 73  -  77+56  = -6.5
2
The concentration effect =

 56+73  -  47 + 56  = 2.5
2
The results are not exactly identical to what was

obtained by conducting eight experiments previously.
But, the same relative conclusions can be drawn.
CSSBB 2014
IX. IMPROVE VII.A.6

MINITAB Results
Most people don’t analyze experimental results using
manual techniques. A synopsis of the effects of
temperature, pressure, and concentration on yield
results using MINITAB is presented in the Primer. This
analysis represents the very same data for the
previously presented examples.
CSSBB 2014
IX. IMPROVE VII.A.6

Experimental Notation
Often experiments are abbreviated numerically. For
example, a 25 experiment means five factors at two
levels (or 32 experiments). A mnemonic aid is: levels
are level to the line, while factors fly.
Fractional designs can be expressed as Lf-g, where L is

the number of levels, f is the number of factors, and g
represents the number of generators. A generator of a
fraction determines what effects are combined or
confounded with one another. A design with g such
generators is called a 1/Lg fractional factorial.
For example a 25-2 design is a two-level, five-factor, one-

fourth fractional design. Rather than 32 runs, this
experiment will require only eight runs.
CSSBB 2014
IX. IMPROVE VII.B.1

LEAN METHODS/WASTE ELIMINATION
Waste Elimination
Various lean manufacturing techniques are used in the
six sigma control phase for waste elimination: 5S, visual
controls, kaizen, kanban, poka-yoke, and standard work.
Also important in this effort is the identification and
minimization of the seven classic wastes reviewed in
Section VIII. The application of many of these lean tools
is summarized below.
5S Workplace Organization
Implementing 5S is a fundamental first step for any
manufacturing company wishing to call itself world
class. The presence of a 5S program is indicative of the
commitment of senior management to workplace
organization, lean manufacturing, and the elimination of
muda (Japanese for waste). The 5S program mandates
that resources be provided in the required location, and
be available as needed to support work activities. The
five Japanese “S” words for workplace organization are:
C Seiri (proper arrangement)

C Seiton (orderliness)
C Seiso (cleanup)
C Seiketsu (standardize)
C Shitsuke (personal discipline)
CSSBB 2014
IX. IMPROVE VII.B.1

5S (Continued)
For American companies, the translated English
equivalents are:
C Sort: Separate out and eliminate all that is unneeded

C Straighten: Put things in order
C Scrub (Shine): Make the workplace spotless
C Standardize: Make cleaning and checking routine
C Sustain: Commit to the previous 4 steps
The 5S approach exemplifies a determination to

organize the workplace, keep it neat and clean, establish
standardized conditions, and maintain the discipline that
is needed to do the job.
Modifications have been made on the 5S structure. It

can be reduced to 4S. It can be modified to a 5S + 1S or
6S program, where the sixth S is safety. The 5S concept
requires that a discipline of will be installed and
maintained.
CSSBB 2014
IX. IMPROVE VII.B.1

5S (Continued)
Details of a 5S program are itemized below in a step-by-
step approach.
Step 1: Sort (Organize)
C Set up a schedule to target each area

C Remove unnecessary items in the workplace
C Red tag unneeded items,
C Record everything that is thrown out
C Keep repaired items that will be needed
C Major housekeeping and cleaning is done by area
C Inspect the facility for problems and grime
C List everything which needs repair
C Deal with causes of filth and grime
C Red tag grime areas and prioritize corrections
C Perform management reviews of all steps
CSSBB 2014
IX. IMPROVE VII.B.1

5S (Continued)
Step 2: Straighten
C Have a place for everything

C Put everything in its place
C Analyze the existing conditions
C Create location for everything
C Decide how things are stocked (with locations)
C Use labels, tool outlines, and color codes
C Determine everyday controls
C Determine out-of-stock conditions
C Define who does the reordering
C Reduce inventories
C Determine who has missing items or if they are lost
C Use aisle markings, and equipment placements
C Establish pallet zones for work in process (WIP)
CSSBB 2014
IX. IMPROVE VII.B.1

5S (Continued)
Step 3: Scrub (Shine and Clean)
C This includes ways to keep things clean

C Establish a commitment to be responsible
C Clean everything in the workplace
C Remedy machinery and equipment problems
C Complete training on maintenance basics
C Assign individual area responsibilities
C Rotate difficult or unpleasant jobs
C Implement 3 and 10-minute 5S activities
C Use inspection checklists
CSSBB 2014
IX. IMPROVE VII.B.1

5S (Continued)
Step 4: Standardize
C Make 5S activities routine

C Determine the important areas to manage
C Maintain and monitor facilities for cleanliness
C Make abnormal conditions obvious
C Use visual controls
C Set standards, and determine necessary tools
C Determine inspection methods
C Determine short-term countermeasures
C Determine long-term remedies
C Use visual tools such as color coding and labels
C Provide equipment markings, maps, and charts
CSSBB 2014
IX. IMPROVE VII.B.1

5S (Continued)
Step 5: Sustain
C Commit to the 4 previous steps

C Continually improve on the 4 previous steps
C Acquire self-discipline
C Establish standards for each of the 5S steps
C Establish and perform evaluations of each step
Management will determine the control and self-

discipline areas for an organization. A well run 5S
program will result in a factory that is in control.
CSSBB 2014
IX. IMPROVE VII.B.1

Poka-Yoke / Mistake Proofing

Shigeo Shingo is widely associated with a Japanese
concept called poka-yoke (pronounced poker-yolk-eh)
which means to mistake proof the process. The
success of poka-yoke is to provide some intervention
device or procedure to catch the mistake before it is
translated into nonconforming product.
Shingo lists the following characteristics of poka-yoke

devices:
C They permit 100% inspection

C They avoid sampling for monitoring and control
C They are inexpensive
Poka-yoke devices can be combined with other

inspection systems to obtain near zero defect
conditions. Errors can occur in many ways:
C Skipping an operation
C Positioning parts in the wrong direction
C Using wrong parts or materials
C Failing to properly tighten a bolt
CSSBB 2014
IX. IMPROVE VII.B.1

Poka-Yoke (Continued)
There are numerous adaptive approaches. Devices can
stop machines from working if a part or operation
sequence has been missed by an operator. A
specialized tray or dish can be used prior to assembly to
ensure that all parts are present.
Mechanical screening devices can be utilized.

Applications can be based on length, width, height, and
weight. Cash registers at many fast food outlets have
descriptions or schematics of the product purchased.
This system, in addition to the use of bar codes at
supermarkets has eliminated data entry errors and
saves time. Mistake proofing is a preventive technique.
Mistake proofing can be accomplished through control

methods by preventing human errors or by using a
warning mechanism to indicate an error.
CSSBB 2014
IX. IMPROVE VII.B.1

A signaling mechanism can alert a worker of possible
sources of error. A buzzer or light will signal that an
error has occurred, requiring immediate action. Root
cause analysis and corrective action are required before
work resumes.
Other than eliminating the opportunity for errors,

mistake proofing is relatively inexpensive to install and
engages the operator in a contributing way. Work teams
can often contribute by brainstorming potential ways to
thwart error-prone activities. A disadvantage is, in many
cases, that technical or engineering assistance is
required during technique development.
CSSBB 2014
IX. IMPROVE VII.B.1

Design improvements to mistake proof products and
processes include:
C Elimination of error-prone components

C Amplification of human senses
C Redundancy in design (back-up systems)
C Simplification by using fewer components
C Consideration of environmental factors
C Providing fail-safe cut-off mechanisms
C Enhancing product producibility and maintainability
C Selecting components and circuits that are proven
Examples of Poka-Yoke
C Gas cap attached to a car

C Gas pumps with automatic shut-off nozzles
C 110V electrical plugs and polarized sockets
C Microwave stops when door is opened
C Seatbelt buzzer to warn drivers and passengers
C Lawn mower safety shut-off when bar is released
C Symmetrical car keys to allow two-way insertion
C Product drawings on cash registers
C Bar codes for product identification
CSSBB 2014
IX. IMPROVE VII.B.1

Potential countermeasures for human errors include:
Errors Cause Countermeasures

Forgetfulness Poor Checklists/Visual aids
concentration
Misunderstanding Unfamiliar Training/Checklists
situations Work standardization
Visual aids/Work instructions
Identification Similar Training/Visual aids
appearance
Amateur errors Inexperience Training/Skill building
Beginner errors Work standardization
Willful errors Ignoring rules Training/Work instructions
Inadvertent errors Absent Discipline/Skill building
mindedness Work standardization
Visual aids/Checklists
Slowness Judgment delays Work standardization
Lack of standards Inadequate Work standardization
instructions Work instructions
Surprise errors Erratic equipment TPM/Work standardization
Intentional errors Crimes / Education/Discipline
Sabotage
CSSBB 2014
IX. IMPROVE VII.B.1

Standard Work
The operation of a plant depends on the use of policies,
procedures, and work instructions. These could be
referred to as standards. Maintaining and improving
standards leads to improvement of the processes and
plant effectiveness.
The following are some examples of standards that go

beyond procedures and work instructions:
C Yellow lines on the floor

C Color coding
C Production control board
C Indicators for inventory levels
C Cross-training matrices
C Trouble lights
CSSBB 2014
IX. IMPROVE VII.B.1

Standard Work (Continued)

Standard work is regarded to be one of the most
important techniques for achieving a perfect process.
This approach provides the discipline for attaining
perfect flow in a process. Under normal work
conditions, with no abnormalities in the system, the flow
is perfect. Standard work conditions are determined for:
C Takt time
C Ergonomics
C Parts flow
C Maintenance procedures
C Routines
Standard work is the documentation of each action

required to complete a specified task. Standard work
should always be displayed at the workplace.
A standard work sheet combines the 3 elements of

materials, workers, and machines in a work
environment. Toyota refers to it as a work combination.
CSSBB 2014
IX. IMPROVE VII.B.1


Standard work sheets, that operators will have
confidence in, should consider the following:
C Resource availability
C Machine arrangements
C Process improvements
C Worker ideas valued
C Tooling improvements
C Minimized transport
C Optimized inventory
C Defective prevention
C Operational mistakes deterred
C Safe workplace concepts
C Autonomous systems installed
CSSBB 2014
IX. IMPROVE VII.B.1


The elements that comprise standard work are:
C Cycle time: This is the time allowed to make a piece

of production. This will be based on the takt time.
The actual time will be compared to the required
takt time to see if improvements are needed.
C Work sequence: This is the order of operations that

the worker must use to produce a part. The same
order of work must be done every time. A standard
time, work combination sheet, and work layout
sheet are provided to the operator.
C Standard inventory: This is the minimum allowable

in-process material in the work area, needed to
maintain a smooth flow. For continuous flow, one
piece in the machine and one piece for hand off is
optimal.
CSSBB 2014
IX. IMPROVE VII.B.1


Shingo and Sharma indicate that standard charts will
also include:
C Capacity charts by part

C Standard task combination sheets
C Task manuals
C Task instruction manuals
C Standard operating sheets
Sharma provides a final definition of standard work:
“The best combination of machines and people

working together to produce a product or provide a
service at a particular point in time.”
CSSBB 2014
IX. IMPROVE VII.B.1


Standards have the following features:
1. Standards are the easiest and safest way to work

2. They preserve employee know-how and expertise
3. They provide a way to measure performance
4. They show work related relationships
5. They assist maintenance and improvement
6. They provide a set of visual signs
7. Standards are a basis for training
8. They are a basis for auditing
9. They are a means to prevent errors
10. Standards minimize variability
CSSBB 2014
IX. IMPROVE VII.B.1

Takt Time
In the operation of a continuous flow manufacturing
(CFM) line, takt time takes on great importance. Takt
time is a time element that equals the demand rate. In a
CFM or one-piece flow line, the time allowed for each
line operation is limited. The line is ideally balanced so
that each operator can perform their work in the time
allowed. The word, takt, is a German word for baton,
used by an orchestra conductor. This provides a
rhythm to the process, similar to a heartbeat.
CSSBB 2014
IX. IMPROVE VII.B.1

Kanban-Pull
Kanban is the Japanese word for “sign” and is a method
of material control in the factory. It is intended to
provide product to the customer with the shortest
possible lead times. Inventory and lead times are
reduced through Heijunka (leveling of production).
Using a kanban system the method of feeding an

assembly line is achieved using the following process:
1. Parts are used on the assembly line and a

withdrawal kanban is placed in a designated area.
2. A worker takes the withdrawal kanban to the

previous operation to get additional parts. The
WIP kanban is removed from the parts pallet and
put in a specified spot. The original withdrawal
kanban goes back to the assembly line.
3. The WIP kanban card instructs the operator to

produce more parts. This may require a kanban
card to pull material from an earlier operation.
4. The next operation will see that it has a kanban

card and will produce more parts.
5. This sequence can continue further upstream.

CSSBB 2014
IX. IMPROVE VII.B.1

Kanban-Pull (Continued)
The order to produce parts at any one station is
dependent on receiving an instruction, the kanban card.
This system aims at simplifying paperwork, minimizing
WIP and finished goods inventories. Examples of
kanban cards are shown below.
Production Instruction Production Instruction

KANBAN KANBAN
Code Color Code Color
RZC 5 MBT 8
Type Manual Type Automatic
Quantity 1 Set Quantity 5 Each
Style Standard Style Deluxe
Control 4M539ALR Control 1Z2673YQP
A kanban system can not have production halted by

machine failures or quality problems. A stoppage will
cause distress throughout the production system.
Every effort is made to eliminate causes of downtime
and errors. Kanbans are normally cards, but they can
be flags or spaces on the floor, etc.
CSSBB 2014
IX. IMPROVE VII.B.2

LEAN METHODS/CYCLE TIME REDUCTION
Cycle Time Reduction

Cycle time is defined as the amount of time needed to
complete a single task and to move it forward in the
process. Cycle times may differ by task, but to make the
line flow, all operations must be completed under a
given takt time.
A reduction in cycle time is undertaken for many of the

following reasons:
C To please a customer
C To reduce internal or external wastes
C To increase capacity
C To simplify operations
C To reduce product damage (improve quality)
C To remain competitive
CSSBB 2014
IX. IMPROVE VII.B.2

Cycle Time Reduction

Training
Some cycle time training topics are listed below:
C Introduction to the total systems concept

C Problem solving tools such as the “5 Whys”
C Importance of the next process as the customer
C Non-judgmental attitude to problem solving
C Identification of value and non-value added work
C Identification of muda (the seven wastes)
C Principles of motion study
C Work flow patterns (straight, T-shaped, U-shaped)
C Standard operations
C 5S workplace organization
C Visual management principles
C Just-in-time (JIT) production
C Poka-yoke principles
C Team dynamics
CSSBB 2014
IX. IMPROVE VII.B.2

Cycle Time Reduction (Continued)

The concept of cycle time reduction is discussed in the
Primer with an example of a line with 5 stations and
observed cycle times.
Various alternatives are considered including work

times and delays.
CSSBB 2014
IX. IMPROVE VII.B.2


A study of the stations reveals the motions used by the
operators. An approximation of the operator effort will
suffice. The Shingo technique of classifying human
motions is divided into 4 grades:
1. Assemble, disassemble, and use (true value

added)
2. Transport empty, grasp, transport loaded, and

release load (non-value added)
3. Search, find, select, reposition, hold, inspect, and

pre-position (non-value added, lower on the
grading scale than above)
4. Rest, frequent planning, unavoidable delays,

avoidable delays (non-valued added elements of
the lowest rank)
The concept in studying human motions is to reduce the

stress and strain upon the operators, creating a more
efficient operation.
CSSBB 2014
IX. IMPROVE VII.B.2


In the hypothetical Primer example an improvement
team prepares a workplace layout of the line. This
layout will include operators, WIP inventory, raw
materials, and equipment in the workspace. A charting
of the current flow of the product may reveal a
“spaghetti-like” flow as shown below.
In
Station 4 Station 1
Operator 4 Operator 1
Station 2 Station 5 Station 3
Operator 2 Operator 5 Out Operator 3
The idea is to arrange the production line using either a

U-shape, L-shape, C-shape, or straight line arrangement
in order to create continuous flow. The various lines
must reduce the distance traveled by the part, reduce
the amount of WIP inventory between stations, and still
meet the required takt time.
CSSBB 2014
IX. IMPROVE VII.B.2


Perhaps a U-shaped line can be constructed as shown
below.
Station 2 Station 1 In
Operator 2
Station 3 Operator 1
Operator 3
Station 4 Station 5 Out
An analysis of the walking distance and material flow

reveals significantly less wasted motion.
CSSBB 2014
IX. IMPROVE VII.B.2


The team improvement goals are usually:
C Reducing cycle times

C Meeting takt times
C Reducing space
C Reducing inventory
C Increasing line balance
C Maintaining a safe work environment
CSSBB 2014
IX. IMPROVE VII.B.2

Continuous Flow Manufacturing

The main principle in CFM is that material should always
be moving one-piece at a time, at a rate determined by
the needs of the customer. The flow of product must be
smooth and uninterrupted by:
C Quality issues
C Setups
C Machine reliability
C Breakdowns
C Distance
C Handling methods
C Transportation arrangements
C Staging areas
C WIP inventory problems, etc.
CSSBB 2014
IX. IMPROVE VII.B.2

CFM (Continued)
The mass production, or large lot production, world is a
series of operations that produce goods in large
batches. The sequence of operations used in producing
large batch sizes results in waiting time between
operations. Large lot production has the following
faults:
C Longer lead times for customer delivery

C Additional internal resources are needed
C Additional product transportation expenses
C Increased product damage or deterioration costs
Continuous flow or one-piece flow will:
C Deliver product flow product with less delay

C Require less storage and transport
C Lower the risk of losses
C Provide a mechanism to solve other problems
CSSBB 2014
IX. IMPROVE VII.B.2

CFM (Continued)
Ideally, in a continuous flow manufacturing layout, the
production steps are arranged in a tight sequence, such
as a straight line or U-shaped cell, without WIP, using
single piece flow. Inside this flow concept, each station
and operator must operate with complete reliability to
achieve continuous flow and the desired takt time.
The following concepts are important:
C Poka-yoke (mistake proofing): To prevent defects

from proceeding to the next step.
C Source inspection: To catch errors that cause

defects and to correct the process.
C Self-check: Checks by the operator to catch defects

and to correct the process.
C Successive checks: Checks by the next process to

catch errors and to correct the process.
C Total productive maintenance is used to help

achieve high machine capability
CSSBB 2014
IX. IMPROVE VII.B.2

Setup Reduction
SUR is an acronym for setup reduction. SMED is an
acronym for single minute exchange of dies. In this
discussion, the two terms will be used interchangeably.
SUR is one of the most important tools in the lean
manufacturing system. The concept is to take a long
setup change of perhaps hours in length and reduce it
to minutes. There are 3 myths regarding setup times:
C The skill for setup changes comes from

considerable practice and experience
C Long production runs are more efficient because

they save setup times
C Long production runs are economically better

CSSBB 2014
IX. IMPROVE VII.B.2

Setup Reduction (Continued)

SUR systems reduce the dependence on the long-term
experience of operators to perform an effective
changeover. SUR systems reduce the skill level needed
for setup changes. Long runs will reduce problems with
setup changes, but lead to excess inventories, extra
handling, extra storage, etc. The reduction of setup
times:
C Expands production capacity

C Reduces inventories and minimize wastes
C Prevents quicker responses to demand changes
C Increases operating flexibility
C Makes more effective use of floor space
C Improves the utilization of capital equipment
C Reduces material handling
C Increases operator efficiency and safety
CSSBB 2014
IX. IMPROVE VII.B.2


To achieve quick changeover one determines operations
that must be done while the machine is stopped, called
internal setup (IS), and distinguish these from those
which can be done while the machine runs, called
external setup (ES). Any useless steps are removed.
The sequence consists of:
C Removing useless operations

C Converting IS to ES
C Simplifying fittings and installations
C Suppressing adjustments and trials
C Working continually on improvement ideas
Quick changeover methods involve the following

sequence of actions:
C Document all elements of current setup

C Separate internal from external operations
C Convert internal setups to external setups
C Generate ideas for reducing external setups
C Generate ideas for reducing internal setups
C Evaluate/test new ideas
C Prepare for the next changeover using new ideas
C Standardize new actions and procedures
C Continuously improve the process
CSSBB 2014
IX. IMPROVE VII.B.2


In planning a SUR project, the actual conditions and
steps of the die changeover must be detailed. This can
be done by:
C Use of a stopwatch for continuous observation

C Use of a work sampling study
C Worker interviews
C Videotaping the entire setup operation
CSSBB 2014
IX. IMPROVE VII.B.2


Every step in the setup process (from start to finish) is
broken down and classified. A major item is to separate
items that can be done when the machine is running
(external setup) and to separate the items that can only
be done when the machine is down (internal setup).
External setup operations should include:
C Preparation of parts
C Finding parts
C Measuring parts
C Maintenance of dies and spares
C Cleaning of spares, etc.
The break down of initial elements into internal and

external setup operations is just a start. The existing
internal setup elements should be reexamined to
convert more of those elements into external setup. The
goal is to reduce the time to under 1 digit. It may take a
series of SUR projects to lower the time to 1 digit.
Once a SUR procedure is agreed upon, the setup team

should practice the process and critique itself for
additional improvements. A SUR case study is
presented in the Primer.
CSSBB 2014
IX. IMPROVE VII.B.2

Quick Response Manufacturing

Quick Response Manufacturing helps companies use
speed and the reduction of cycle times to deliver
products and services faster than their competitors.
This methodology can be applied to both the shop floor
and the office.
In many cases, QRM requires that the managerial mind

set must change. QRM is especially useful for a product
line that has a large variety of highly engineered
products with variable demand.
A specialized material planning technique that is a

combination of both “push” and “pull” termed
“POLCA”, is used for controlling material flow. POLCA
stands for paired-cell overlapping loops of cards with
authorization. This is a material control system that
operates in conjunction with MRP and a cellular
arrangement.
CSSBB 2014
IX. IMPROVE VII.B.2

QRM (Continued)
The QRM methodology focuses on speed. Suri
highlights these principles:
C Change the management mindset

C Complete a job by minimizing lead times
C Operate critical resources at 80% not 100%
C Measure reduction of lead time not utilization
C Do not measure efficiencies or utilizations
C Lead time reduction trumps on-time delivery
C Install the POLCA material control system
C Move the suppliers to QRM
C Educate customers on QRM
C Use quick response cells for product families
C QRM will lead to a truly lean company
CSSBB 2014
IX. IMPROVE VII.B.2

LEAN METHODS/KAIZEN
Kaizen
Kaizen is Japanese for continuous improvement. The
word kaizen is taken from the Japanese kai “change”
and zen “good.” Kaizen is usually referred to as
incremental improvement, but on a continuous basis,
involving everyone. Western management is enthralled
with radical innovations. They enjoy seeing major
breakthroughs, the home runs of business. Kaizen is an
umbrella term for:
C Productivity
C Total quality control
C Zero defects
C Just-in-time
C Suggestion systems
The kaizen strategy involves:
C Maintaining and improving operating standards

C Improving processes
C Using the PDCA/PDSA cycles
C Making quality the highest priority
C Solving problems with hard data.
C Considering the next process as the customer
CSSBB 2014
IX. IMPROVE VII.B.2

LEAN METHODS/KAIZEN
The Kaizen Blitz

While most kaizen activities are considered to be of a
long-term nature by numerous individuals, a different
type of kaizen strategy can occur. This has been termed
a kaizen event or kaizen blitz, which involves a kaizen
activity in a specific area (involving planning, training,
and implementation) within a short time period.
The kaizen blitz, using cross functional volunteers in a

3 to 5 day period, results in a rapid workplace change on
a project basis. The volunteers come from various
groups, such as accounting, marketing, engineering,
maintenance, quality and production. If the work
involves a specific department, more team members are
selected from that department.
Depending on the experience levels of the group, a 5

day kaizen blitz starts with 2 days of intense sessions
on continuous improvement concepts. This is followed
by 3 days of hands on data collection, analysis, and
implementation at the source. Management must trust
the decision-making process as determined by the
kaizen blitz team and facilitator.
CSSBB 2014
IX. IMPROVE VII.B.2

LEAN METHODS/KAIZEN
The Kaizen Blitz (Continued)

A significant amount of time and money is involved at
the implementation stage. The team makes a final
presentation of the project to the plant manager and all
interested plant employees. Every project has the
possibility of bringing immediate changes and benefits.
Various metrics are used to measure the outcomes of a

kaizen blitz:
C Floor space saved

C Line flexibility
C Improved work flow
C Improvement ideas
C Increased quality levels
C Safe work environment
C Reduced non-value added time
CSSBB 2014
IX. IMPROVE VII.B.4

LEAN METHDOS/OTHER TOOLS
Theory of Constraints
The theory of constraints is an intuitive framework for
managing based on the desire to continually improve a
company. Using TOC, a definition of the goals of the
company are established along with metrics for critical
measures.
Goldratt reminds readers that there are three basic

measures to be used in the evaluation of a system.
C Throughput
C Inventory
C Operational expenses
These measures are more reflective of the true system

impact than machine efficiency, equipment utilization,
downtime, or balanced plants.
CSSBB 2014
IX. IMPROVE VII.B.4

Theory of Constraints (Continued)

A few of the most widely used TOC concepts are
detailed below:
C Bottleneck resources are resources whose capacity

is equal to or less than the demand placed upon it.
If a resource presents itself as a bottleneck, then
things must be done to lighten the load. One
should beware of lost production at a bottleneck,
due to poor quality or rejects.
C Balanced plants are not always a good thing. One

should not balance capacity with demand, but
balance the flow of product through the plant with
demand from the market. One can do more with
less by just producing what the market requires at
the time.
CSSBB 2014
IX. IMPROVE VII.B.4


C Dependent events and statistical fluctuations are
important. A subsequent event depends upon the
ones prior to it. A bottleneck will restrain the entire
throughput.
C Throughput is the rate at which the system

generates money through sales. The finished
product must be sold before it can generate money.
C Inventory is all the money that the system has

invested in purchasing things that it intends to sell.
This can also be defined as sold investments.
C Operational expenses are all the money that the

system spends in order to turn inventory into
throughput.
C The terms throughput, inventory, and operational

expenses define money as incoming money, money
stuck inside, and money going out.
CSSBB 2014
IX. IMPROVE VII.B.4


Goldratt recommends that the following 5-step method
be used for TOC implementation:
1. Identify the system’s constraints.
2. Decide how to exploit the system’s constraints.
3. Subordinate everything else to the above

decisions.
4. Elevate the system’s constraints.
5. Back to step 1. After the constraint has been

broken, look for new constraints.
CSSBB 2014
IX. IMPROVE VII.B.4

Drum - Buffer - Rope

The drum-buffer-rope concept relates to TOC step 3
“Subordinate everything else to the above decisions.”
Most discussions on bottlenecks or constraints center
on increasing the capacity or removing factors that slow
the bottleneck. However, if the capacity of the
bottleneck cannot be increased, then one must accept
it, and then work to maximize the bottleneck’s output.
One must ensure a smooth source of materials to the

bottleneck. The ideal situation is to always have enough
WIP for the bottleneck (which controls the pace of the
line) to keep production rates moving. Therefore, a set
amount of inventory (a buffer) is needed ahead of the
bottleneck.
CSSBB 2014
IX. IMPROVE VII.B.4

Drum - Buffer - Rope (Continued)

To maintain a proper buffer level, a feedback mechanism
is necessary to control the release of raw materials to
the downstream equipment. One such technique is
called the drum-buffer-rope concept, as described
below:
C Drum: This is the constraint that controls the pace

of the process. The “beat” of this operation sets the
pace of the line.
C Buffer: This is the work-in-process, or inventory, for

the bottleneck. It must be available to keep the
bottleneck operating at full performance.
C Rope: This is the feedback mechanism from the

buffer to the raw material input point. The
dispatching point will release only enough material
to keep the buffer inventory at the proper level.
CSSBB 2014
IX. IMPROVE VII.B.4

Drum - Buffer - Rope (Continued)

The drum-buffer-rope technique can be thought of as a
“pull-push” system. Refer to the Figure below.
Buffer Drum
Shipping
Material
Step 1 Step 2 Inventory Bottleneck or next
Release
operation
Feedback loop
Rope
The DBR model can work very well in a job shop with its
wide variety of products, routings, and process times.
In this environment, bottlenecks can be everywhere.
The use of DBR methods has led to excellent
performances in some situations, because WIP is kept
low, and lower system cycle times are achieved.
CSSBB 2014
IX. IMPROVE VII.C

IMPLEMENTATION
Implementing Improved Processes

Listed below are a number of methods used to evaluate
process and product improvements prior to full scale
deployment.
Pilot run or pilot study: This is a trial of a changed

product, process, equipment, or system to gain
experience and collect data about the change.
Bothe calls a pilot study a temporary introduction of a

solution designed to confirm its effectiveness and
uncover any potential problems with its eventual
implementation. Some common objectives of such a
study include:
C Identifying potential implementation problems

C Discovering any adverse side effects
C Learning how to optimize the solution
CSSBB 2014
IX. IMPROVE VII.C

IMPLEMENTATION
Implementing Improved
Processes (Cont’d)
Simulation: This is a time-dependent trial of a new
product, process, or system. This can often be
accomplished by mathematical or computer-based
modeling. This form of testing is conducted when an
actual demonstration is too difficult, time consuming,
expensive, massive, or dangerous. Normally parts,
units, or operational people are not physically involved.
Demonstration: In the six sigma context, a

demonstration can be an execution of an improved
process or system. If a product is involved, there may
be a requirement to obtain an approval for any real or
potential changes. In the case of some products or
processes, various human factors such as sight, touch,
audio, thermal, or vibration may require investigation.
Prototype testing: This is the evaluation of a

developmental model or unit that is close to production.
It should be highly representative of the final equipment,
parts, and processes. There may be additional
manufacturing design changes but a prototype should
allow for full mechanical and electrical evaluation.
CSSBB 2014
IX. IMPROVE
QUESTIONS
9.1. In the drum-buffer-rope methodology, what role does the drum

play?
a. The process output

b. The process constraint
c. The process feedback
d. The process inventory
9.5. An incomplete block design may be especially suitable when:
a. There is missing data

b. There is need for fractional replication
c. It may not be possible to apply all treatments in every block
d. There is a need to estimate the parameters during the
experimentation
9.8. Identify the lean enterprise technique in which the videotaping of

a segment of the operation is helpful:
a. SUR/SMED
b. TPM
c. Takt time determination
d. FIFO
Answers 9.1 b, 9.5 c, 9.8 a

CSSBB 2014
IX. IMPROVE
QUESTIONS
9.13. Considering that all of the following terms have benefits, which
would most likely improve product quality?
a. Waste elimination
b. Takt time
c. Setup reduction
d. Standard work sheets
9.16. When performing one experiment with five repetitions, what are the
six experiments called?
a. Randomization
b. Replications
c. Planned grouping
d. Sequential
9.17. The tool/technique most widely used by a number of automotive

manufactures that supply products to Toyota is called:
a. Kanban
b. Muda
c. Poka-yoke
d. An andon board
Answers 9.13 a, 9.16 b, 9.17 a

CSSBB 2014
IX. IMPROVE
QUESTIONS
9.24. Review the following 5S elements and identify the step that is
being referenced.
Determine who has missing items

Create a name and location for everything
Use aisle and material placement markings
Use labels, tool boards, and color codes
a. Sort
b. Straighten
c. Scrub
d. Standardize
9.26. Randomized block designs are best suited for:
a. Screening objectives
b. Comparative objectives
c. Response surface objectives
d. Regression model objectives
9.28. A 2 level 5 factor experiment is being conducted to optimize the

reliability of an electronic control module. A half replicate of the
standard full factorial experiment is proposed. The number of
treatment combinations will be:
a. 10
b. 16
c. 25
d. 32
Answers 9.24 b, 9.26 b, 9.28 b

CSSBB 2014
IX. IMPROVE
QUESTIONS
9.33. Consider the following statement: "The best combination of

machines and people working together to produce a product or
service at a particular point in time." What lean concept is being
described?
a. Standard work
b. A future state map
c. The value stream
d. Ultimate cycle time
9.38. To achieve rapid improvement, which of the following concepts is

appropriate?
a. Standard work
b. Poka-yoke
c. Kaizen blitz
d. 5S activities
9.40. Identify the best way to error proof activities?
a. By corrective actions
b. By preventive actions
c. By containment actions
d. By temporary actions
Answers 9.33 a, 9.38 c, 9.40 b

CSSBB 2014
IX. IMPROVE
QUESTIONS
9.42. How does poka-yoke respond to human error?
a. By eliminating human error

b. By punishing human error
c. By rewarding defect detection due to human error
d. By catching human error before it becomes a defect
9.46. The theory of constraints concentrates mainly on:
a. Understanding customer needs

b. Developing a value stream map
c. Achieving on-time goals
d. Removing process bottlenecks
9.48. Red tagging is used during which 5S stage?
a. Standardize
b. Sustain
c. Straighten
d. Sort
Answers 9.42 d, 9.46 d, 9.48 d

CSSBB 2014
IX. IMPROVE
QUESTIONS
9.51. Why are pilot runs, simulations, demonstrations, and prototype

testing important to six sigma teams?
a. They are great ways to define problems

b. They can save time and money during problem analysis
c. They are important techniques to evaluate potential improvements
d. They indicate the optimal way to control a number of processes
9.54. Which of the following 5S stages is primarily the responsibility of

top management?
a. Shine
b. Sustain
c. Sort
d. Straighten
9.56. What is considered an ideal batch size in a continuous flow

operation?
a. Large batches
b. It depends on the bin size
c. One piece at a time
d. The takt time batch size
Answers 9.51 c, 9.54 b, 9.56 c

© QUALITY COUNCIL OF INDIANA X-1 (919)
CSSBB 2014
X. CONTROL
A STATE OF STATISTICAL
CONTROL IS NOT A NATURAL
STATE FOR A MANUFACTURING
PROCESS. IT IS INSTEAD AN
ACHIEVEMENT, ARRIVED AT BY
ELIMINATING ONE BY ONE, BY
DETERMINED EFFORT, THE
SPECIAL CAUSES OF EXCESSIVE
VARIATION.
W. EDWARDS DEMING
CSSBB 2014
X. CONTROL VIII.A.1
SPC/OBJECTIVES
Control
Control is presented in the following topic areas:
C Statistical process control

C Other control tools
C Maintain controls
C Sustain improvements
Statistical Process Control
Statistical process control is described in the

following topics:
C Objectives
C Selection of variables
C Rational subgrouping
C Control chart selection
C Control chart analysis
CSSBB 2014
X. CONTROL VIII.A.1
SPC/OBJECTIVES
Objectives
Statistical process control (SPC) is a technique for
applying statistical analysis to measure, monitor, and
control processes. The major component of SPC is the
use of control charting methods. The assumption made
in SPC is that all processes are subject to variation.
This variation may be classified as one of two types,
chance cause variation and assignable cause variation.
When assignable cause variation does occur, SPC

facilitates the identification of the source so it can be
eliminated. Statistical process control also provides the
ability to determine process capability, monitor
processes, and identify whether the process is
operating as expected, or whether the process has
changed and corrective action is required.
CSSBB 2014
X. CONTROL VIII.A.1
SPC/OBJECTIVES
Objectives (Continued)
Control chart information can be used to determine the
natural range of the process, and to compare it with the
specified tolerance range. If the natural range is wider,
then either the specification range should be expanded,
or improvements will be necessary to narrow the natural
range.
Benefits from control charting are derived from both

attribute and variable charts. Once the control chart
shows that a process is in control, and within
specification limits, it is often possible to eliminate
costs relating to inspection.
Control charts may be used as a predictive tool to

indicate when changes are required in order to prevent
the production of out of tolerance material.
CSSBB 2014
X. CONTROL VIII.A.1
SPC/OBJECTIVES
Objectives (Continued)
An additional benefit of control charts is the ability to
monitor continuous improvement efforts. When process
changes are made which reduce variation, the control
chart can be used to determine if the changes were
effective.
The benefits of statistical process control are not

without costs. Costs associated with SPC include the
selection of the variable(s) or attribute(s) to monitor,
setting up the control charts and data collection system,
training personnel, and investigating and correcting the
cause when data values fall outside control limits. Many
companies find that the benefits of statistical process
control far outweigh the related costs.
CSSBB 2014
X. CONTROL VIII.A.2
SPC/SELECTION OF VARIABLES
Selection of Variables
Given the benefits of control charting, one might be
tempted to control chart every characteristic or process
variable. The logic is if any characteristic changes, then
the process can be stopped. This decision would also
eliminate the need to determine if one characteristic is
more important than another.
The risk of charting many parameters is the operator will

spend so much time and effort completing the charts,
that the actual process becomes secondary. When a
change does occur, it will most likely be overlooked.
When more than a few charts are used for a process, the
benefits may decrease, as quickly as the costs increase.
CSSBB 2014
X. CONTROL VIII.A.2
SPC/SELECTION OF VARIABLES
Selection of Variables (Continued)

Some considerations for the selection of a control chart
variable include:
C Items that protect human safety

C Items that protect the environment or community
C Items that are running at a high defective rate
C Key process variables that impact the product
C Major sources of customer complaints
C Items that show adherence to applicable standards
C Items that are requested by key customers
C Variables that have caused processing difficulties
C Variables that can be measured
C Items that can be counted by the person charting
C Items that contribute to high internal costs
C Variables that help control the process
CSSBB 2014
X. CONTROL VIII.A.3
SPC/RATIONAL SUBGROUPING
Rational Subgrouping
A control chart provides a statistical test to determine if
the variation from sample-to-sample is consistent with
the average variation within the sample. The key idea in
the Shewhart control chart is the division of
observations into what are called rational subgroups.
Generally, subgroups are selected in a way that makes

each subgroup as homogeneous as possible, and that
gives the maximum opportunity for variation from one
subgroup to another.
In control charting, it is very important to maintain the

order of production. A charted process which shows
out of control conditions (and resulting opportunities for
correction) may be mixed to create new X̄ - R charts
which demonstrate remarkable control. By mixing,
chance causes are substituted for the original
assignable causes as a basis for the differences among
subgroups.
CSSBB 2014
X. CONTROL VIII.A.3
Rational Subgrouping (Continued)

Where order of production is used as a basis for
subgrouping, two fundamentally different approaches
are possible:
C The first subgroup consists of product produced as

nearly as possible at one time. This method follows
the rule for selection of rational subgroups by
permitting a minimum chance for variation within a
subgroup and a maximum chance for variation from
subgroup-to-subgroup.
C Another subgroup option consists of product

intended to be representative of all the production
over a given period of time. Product may
accumulate at the point of production, with a
random sample chosen from all the product made
since the last sample.
In most cases, more useful information will be obtained

from, five subgroups of 5 rather than from one subgroup
of 25. In large subgroups, such as 25, there is likely to
be too much opportunity for a process change within
the subgroup.
CSSBB 2014
X. CONTROL VIII.A.3
Sources of Variability
The long-term variation in a product is termed the
product (or process) spread. There will be some
difference between the process average and variation
from lot-to-lot. One of the objectives of control charting
is to reduce lot-to-lot variability.
It may be necessary to analyze each stream-to-stream

entity separately. Another main objective of control
charting is to reduce the time-to-time variation.
Measurements taken at a number of different points on

a given unit define within-piece variability. Significant
positional variation may necessitate changes in material
or machinery.
Another source of variability is the piece-to-piece

variability of a single production unit. Often, the
inherent error of measurement is significant. This error
consists of both human and equipment components.
The remaining variability is referred to as the inherent
process capability. It is the instant reproducibility of the
machine under ideal conditions.
CSSBB 2014
X. CONTROL VIII.A.3
Breakdown of Variation
Process Spread
Minus
Within Lot Variation Lot-to-Lot Variation
Minus
Within Stream Variation Stream-to-Stream Variation
Minus
Within Time Variation Time-to-Time Variation
Minus
Within Piece Variation Piece-to-Piece Variation
Minus
Inherent Process Variation

Error of Measurement
Equipment Human
CSSBB 2014
X. CONTROL VIII.A.4
SPC/CONTROL CHART SELECTION
Control Charts
Control charts are the most powerful tools to analyze
variation in most processes. A process which is in
statistical control is characterized by plot points that do
not exceed the control limits. When a process is in
control, it is predictable. There are many variations of
possible control charts. The two primary types are for
variables and attributes.
Control Charts for Variables
Plots measurements of a process characteristic. Types

include:
X̄ - R charts (when data is readily available)

Run charts (limited single point data)
MX̄ - MR charts (limited data)
X - MR charts (or I - MR charts) (limited data)
X̄ - s charts (when sigma is readily available)
Median charts
Moving average
CSSBB 2014
X. CONTROL VIII.A.4
Control Charts (Continued)

Control Charts for Attributes
Control charts for attributes plot a general measurement

of the total process (the number of complaints per order,
number of orders on time, absenteeism frequency,
number of errors per letter, etc.). Types include:
p charts (fraction defective)

np charts (number of defectives)
c charts (number of defects)
u charts (number of defects per unit)
In some cases, the relatively larger sample sizes

associated with attribute charts can prove to be
expensive.
CSSBB 2014
X. CONTROL VIII.A.4
X and R Chart Terms

n Sample size (subgroup size).
X A reading (the data).
X Average of readings in a sample.
X Average of all the X s. It is the value of the

central line on the X chart.
R The range. The difference between the largest

and smallest value in each sample.
R Average of all the Rs. It is the value of the

central line on the R chart.
UCL Upper and lower control limits. They are

LCL boundaries for 99.73% of the population. They
are not specification limits.
CSSBB 2014
X. CONTROL VIII.A.4
Typical X - R Control Chart
20.5
UCLX = 20.0
20.0
19.5
X 19.0 X = 18.9
Average 18.5
18.0
LCLX = 17.8
17.5
1 5 10 15 20 25 30
4.5
4.0 UCLR = 4.0
3.5
R
3.0
Range 2.5
2.0 R = 1.9
1.5
1.0
0.5
0 LCLR = 0
1 5 10 15 20 25 30
CSSBB 2014
X. CONTROL VIII.A.4
Steps for Constructing X̄ - R Charts

1. Determine the sample size and frequency
2. Collect 20 to 25 sets of time sequenced samples
3. Calculate the average for each set of samples
4. Calculate the range for each set of samples
5. ¯ (the average of the X̄ values), this is

Calculate X̄
the center line of the X̄ chart.
6. Calculate R̄ (the average of the R values). This is

the center line of the R chart.
7. Calculate the control limits:
¯ + A2 R̄
X̄ chart: UCLX̄ = X̄ ¯ - A2 R̄
LCLX̄ = X̄
R chart: UCLR = D4 R̄ LCLR = D3 R̄
8. Plot the data and interpret the chart for special or

assignable causes.
CSSBB 2014
X. CONTROL VIII.A.4
X̄ - R Control Limit Factors
n A2 D3 D4 d2
2 1.88 0 3.27 1.13
3 1.02 0 2.57 1.69
4 0.73 0 2.28 2.06
5 0.58 0 2.11 2.33
6 0.48 0 2 2.53
CSSBB 2014
X. CONTROL VIII.A.4
X̄ - R Control Chart Data

Measurements Sample Sample
Sample
1 2 3 4 5 Avg. X̄ Range R
1 12 12 13 15 12 12.8 3
2 15 17 16 17 18 16.6 3
3 13 18 14 14 15 14.8 5
4 10 12 11 10 11 10.8 2
5 13 16 14 15 14 14.4 3
6 15 12 13 15 11 13.2 4
7 15 16 15 16 15 15.4 1
8 15 17 16 14 12 14.8 5
9 22 17 15 17 14 17.0 8
10 16 15 17 15 18 16.2 3
11 16 18 16 16 16 16.4 2
12 15 16 17 17 14 15.8 3
13 17 15 16 15 16 15.8 2
14 16 15 18 18 16 16.6 3
15 17 19 17 15 17 17.0 4
16 19 17 15 15 17 16.6 4
17 16 19 16 15 14 16.0 5
18 16 15 17 16 18 16.4 3
19 17 13 17 15 14 15.2 4
20 19 18 17 15 16 17.0 4
21 14 17 16 14 13 14.8 4
Totals 323.6 75
¯ + A2 R̄ = 15.4 + (0.58 x 3.6) = 17.5
UCLX̄ = X̄ ¯
X̄ 15.4
¯ - A2 R̄
LCLX̄ = X̄ = 15.4 - (0.58 x 3.6) = 13.3 R̄ 3.6
UCLR = D4 R̄ = 2.11 x 3.6 = 7.6
LCLR = D3 R̄ = 0 x 3.6 = 0
CSSBB 2014
X. CONTROL VIII.A.4
X̄ - R Control Chart
Chart No. 1
Product Name: Tablets Process Closure Department Operator Bill
Variable: Removal Torques Specification Limits: LSL = 10 LBS USL = 22 LBS Units of Measure: LBS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
18
UCLX =17.5
17
run
16
X = 15.4
15
Averages
14
13 LCLX =13.3
special
12
special
11
10.5
special
8
UCLR = 7.6
7 special
Ranges
6
5
4
3
R = 3.6
2
1
0
LCLR = 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Date Start 1/12/10 1/12 1/13
Time 7:05 7:10 7:35 8:10 8:15 9:10 9:12 9:33 11:40 11:43 12:05 13:05 13:45 13:55 14:20 14:55 7:00 7:55 9:00 9:12 9:32
1 12 15 13 10 13 15 15 15 22 16 16 15 17 16 17 19 16 16 17 19 14
2 12 17 18 12 16 12 16 17 17 15 18 16 15 15 19 17 19 15 13 18 17
3 13 16 14 11 14 13 15 16 15 17 16 17 16 18 17 15 16 17 17 17 16
Sample 4 15 17 14 10 15 15 16 14 17 15 16 17 15 18 15 15 15 16 15 15 14
Measurement 15 11
5 12 18 14 11 15 12 14 18 16 14 16 16 17 17 14 18 14 16 13
Total 64 83 74 54 72 66 77 74 85 81 82 79 79 83 85 83 80 82 76 85 74
Average x 12.8 16.6 14.8 10.8 14.4 13.2 15.4 14.8 17 16.2 16.4 15.8 15.8 16.6 17 16.6 16 16.4 15.2 17 14.8
Range R 3 3 5 2 3 4 1 5 8 3 2 3 2 3 4 4 5 3 4 4 4
This start up process smoothed out from data set 10 on.

The chart would need new control limits from that point.
CSSBB 2014
X. CONTROL VIII.A.4
X-Bar and Sigma Charts

X-bar (X̄) and sigma (s) charts are often used for
increased sensitivity to variation (especially when larger
sample sizes are used). Often, s comes from automated
process equipment so the charting process is much
easier. The formula is:
  X - X
2
s=
n-1
The X̄ chart is constructed in the same way as described

earlier, except that ( s̄ ) is used for the control limit
calculations via the following formulas:
¯ + A3 s̄
UCLX̄ = X̄ ¯ - A3 s̄
LCLX̄ = X̄
The control limits for the sigma (s) chart are calculated
using the following formulas and table:
UCLs = B4 s̄ LCLs = B3 s̄
CSSBB 2014
X. CONTROL VIII.A.4
Sigma Chart Factors

n 2 3 4 5 6 7 8 9 10 25
A3 2.66 1.95 1.63 1.43 1.29 1.18 1.10 1.03 0.98 1
B3 * * * * 0.03 0.12 0.19 0.24 0.28 1
B4 3.3 2.6 2.3 2.1 2 1.9 1.8 1.8 1.7 1
*The lower control limit for a sigma chart when (n) is less than 6 is zero.
Capability from X̄-s Charts

^ called sigma hat,
The estimated standard deviation σ
can be calculated by:
s
ˆ =
c4
If both X̄ and s charts are in control process capability

can be assessed using the following c4 factors.
n 2 3 4 5 6 7 8 9 10 25
c4 0.8 0.9 0.9 0.9 1 1 1 1 1 1
CSSBB 2014
X. CONTROL VIII.A.4
Median Control Charts

There are several varieties of median control charts.
One type plots only the individual measured data on a
single chart. The middle value is circled. Median charts
may use an odd number of readings to make the median
value more obvious.
Another variety records the data and plots the median

value and range on two separate charts. Minimal
calculations are needed for each subgroup. The control
limits for the median chart are calculated using the same
formulas as the X̄ - R chart:
 +A
UCL X = X  R  -A
LCL X = X  R
2 2
The A 2 values are somewhat different than the A 2 values

for the X̄ - R chart since the median is less efficient and,
therefore, exhibits more variation.
n 2 3 4 5 6

A 1.9 1.2 0.8 0.7 0.6
2
CSSBB 2014
X. CONTROL VIII.A.4
Median Control Charts (Continued)

The range factors (D3 and D4) and process standard
deviation factor (d2) are the same as used for the X̄ - R
chart.
The specific advantages of a median chart are:
C It is easy to use and requires fewer calculations

C It shows the process variation
C It shows both the median and the spread
CSSBB 2014
X. CONTROL VIII.A.4
MX̄ - MR Charts
MX̄ - MR charts are a variation of X̄ - R charts where data
is less readily available. There are several construction
techniques. An example for n = 3 is shown below.
Product Name Liquid Process Filling O per ation Chart No. 2 Operator Bill
Variable W eight Specification Limit USL = 3.1 LSL= 3.0 Units of Measure Grams
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
UCLX = 3.05
3.05
Averages
3.00
X = 2.96
2.95
2.90
LCLX = 2.88
2.85
.3
Range
UCLR = .217
.2
R = .084 .1
LCLR = 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Date 5/1/08
Time 7:18 7:26 7:33 7:45 8:00 8:05 8:20 8:31 8:45 9:00 9:15 9:30 9:45 9:59
2.83 2.80 2.95 2.92 2.93 2.97 2.95 2.86 2.89 2.78 2.96 3.02 2.93 2.88 2.97 3.04 3.05 3.03 3.04 3.04 3.01 3.02 3.00 3.00 3.03
1
2 2.80 2.95 2.92 2.93 2.97 2.95 2.86 2.89 2.78 2.96 3.02 2.93 2.88 2.97 3.04 3.05 3.03 3.04 3.04 3.01 3.02 3.00 3.00 3.03 3.07
3 2.95 2.92 2.93 2.97 2.95 2.86 2.89 2.78 2.96 3.02 2.93 2.88 2.97 3.04 3.05 3.03 3.04 3.04 3.01 3.02 3.00 3.00 3.03 3.07 2.92
Sample 4
Measurements 5
Average, X 2.86 2.89 2.93 2.94 2.95 2.93 2.90 2.84 2.88 2.92 2.97 2.94 2.93 2.96 3.02 3.04 3.04 3.04 3.03 3.02 3.01 3.01 3.01 3.03 3.01
Range, R .15 .15 .03 .05 .04 .11 .09 .11 .18 .24 .09 .14 .09 .16 .08 .02 .02 .01 .03 .03 .02 .02 .03 .07 .15
Notes
Note, as the above chart is plotted, the oldest

measurement is dropped, 2 measurements are reused
and a new measurement is added.
CSSBB 2014
X. CONTROL VIII.A.4
X - MR Charts
Control charts plotting individual readings and a moving
range may be used for short runs and in the case of
destructive testing. X - MR charts are also known as
I - MR, individual moving range charts. The control
limits are calculated using the formulas and factors
below:
UCL X = X + E2 MR UCLMR = D4 MR
LCL X = X - E2 MR LCLMR = D3 MR
n 2 3 4 5
D4 3.27 2.57 2.28 2.11
D3 0 0 0 0
E2 2.66 1.77 1.46 1.29
The control limits for the range chart are calculated

exactly as for the X̄ - R chart.
The X - MR chart (for individuals and moving ranges) is

the only control chart which may have specification
limits shown.
CSSBB 2014
X. CONTROL VIII.A.4
X - MR Charts (Continued)
There are drawbacks in the interpretation and use of X -
MR chart:
C All interpretation is faulty if the data distribution is

not normal.
C Individual's charts do not separate piece-to-piece

repeatability of the process.
C Variability patterns cannot be assured until 80-100

readings are taken.
C Individual’s charts are not as sensitive to changes

in the process as the X̄ - R chart (or MX̄-MR, when n
= 3 ).
MX̄ - MR charts with n = 3 are recommended by the

authors when information is limited.
CSSBB 2014
X. CONTROL VIII.A.4
X - MR Chart Example
Product Name Apple S trudel Process Line A Chart No. 7 Operator you
Variable S tick Weights Specification Limit T85, High 88, Low 82 Units of Measure Grams
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
95
UCL = 94.4
90
X = 85.4
85
Measurement
80
LCL = 76.4
75
15
UCL=11.1
Range 10
5
MR = 3.4
0 LCL = 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Date 4/16/98
Time
1
2
3
Sample 4
Measurements 5
Individuals, X 85 87 86 86 77 83 84 87 90 84 89 82 84 86 88 85 90 83 84 87 87 X = 85.4
Range, R 2 1 0 9 6 1 3 3 6 5 7 2 2 2 3 5 7 1 3 0 MR = 3.4
Notes
1794 68
X= = 85.4 MR = = 3.4
21 20
UCL X = X + E2 MR = 85.4 + (2.66)(3.4) = 94.4
LCL X = X - E2 MR = 85.4 - (2.66)(3.4) = 76.4
UCLMR = D4 MR = (3.27)(3.4) = 11.1
LCLMR = D3 MR = (0)(3.4) = 0
CSSBB 2014
X. CONTROL VIII.A.4
Attribute Charts
An attribute chart plots characteristics such as short or
tall, fat or thin, blue or brown, pass or fail, okay or not
okay, good or bad, etc. Attributes are discrete, counted
data. Unlike variables charts, only one chart is plotted
for attributes. There are four types of attribute charts,
as summarized below:
Chart Records Subgroup

size
p Fraction defective Varies
np Number of defectives Constant
c Number of defects Constant
u Number of defects per unit Varies
100p Percent defectives Varies
The best uses of an attribute chart are to:
C Follow trends and cycles

C Evaluate any change in the process
CSSBB 2014
X. CONTROL VIII.A.4
Attribute Charts (Continued)

Key points to consider when using attribute charts:
C Normally the subgroup size is greater than 50 (for p

charts).
C The average number of defects/defectives is equal

to or greater than 4 or 5.
C If the actual p chart subgroup size varies by more

than ± 20% from the average subgroup size, the
data point must either be discarded or the control
limits calculated for the individual point.
C The most sensitive attribute chart is the p chart.

The most sensitive and expensive chart is the X̄ - R
chart.
C The defects and defectives plotted in attribute

charts are often categorized in Pareto fashion to
determine the vital few. To actually reduce the
defect or defective level, a fundamental change in
the system is often necessary.
CSSBB 2014
X. CONTROL VIII.A.4
Attribute Chart Formulas

Defectives (Binomial Distribution)
p Chart np Chart
% Defectives Defectives
p=
np
for fraction np =
 np
n k
np
p=
n
x 100 for % UCLnp = np + 3 np 1-p  
p(1-p) LCLnp = np - 3 np  1-p 
UCLp = p + 3
n k = number of samples
p(1-p)
LCLp = p - 3
n
n=
n p=
 np
k n
Defects (Poisson Distribution)
u Chart c Chart
Number of Defects/Unit Number of Defects
c
u=
n c=
c
k
u
UCLu = u + 3 UCLc = c + 3 c
n
LCLc = c - 3 c
u
LCLu = u - 3 k = number of samples
n
n=
n u=
c
k n
Sample Size Varies Sample Size Fixed
CSSBB 2014
X. CONTROL VIII.A.4
p Chart Example Calculations
22,500
n= = 1125
20
total defective 52
p= = = 0.23 %
total inspected 22,500
UCLp , LCLp = p  3

p 100 - p 
n
0.23%  100% - 0.23% 

UCLp , LCLp = 0.23%  3
1125
UCLp , LCLp = 0.23%  (3)(0.143%)
UCLp = 0.66% LCLp = -0.20% = 0%

CSSBB 2014
X. CONTROL VIII.A.4
p Chart Example
Attributes Control Chart Form p np c u
PART : DESCRIPTION: Uncoa ted Tablets CHARACTERISTIC: % Defectives DATE: 6/1 - 6/19
SOURCE: Tablet Department OPERATOR: Various INSPECTOR: You
UCL: 0.66 LCL: 0 AVE RAGE: 0.23
1.0
0.9
0.8
0.7
UCLp = .66
0.6
0.5
0.4
0.3
p = .23
0.2
0.1
0 LCLp = 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Sample
(n) See prior data sheet
Number
(np,c) 8 0 12 3 5 0 2 2 1 3 0 5 10 0 0 0 0 0 1 0
Frac tion
% (p,u) .64 0 .89 .25 .43 0 .18 .15 .08 .50 0 .45 .95 0 0 0 0 0 .09 0
Date/ 6 6
Time 1 19
Notes
A system change occurred at plot point 14. If continued,

the sample size should be increased substantially.
CSSBB 2014
X. CONTROL VIII.A.4
np Chart Example Calculations

total defective 112
k = 25 np = = = 4.5
no. of lots 25
 np 
UCLnp , LCLnp = np  3 np  1 - 
 n 
= 4.5  3 (4.5)(0.955)
UCLnp = 10.7 LCLnp = -1.72 = 0

CSSBB 2014
X. CONTROL VIII.A.4
np Chart Example
CSSBB 2014
X. CONTROL VIII.A.4
c Chart Example Calculations
total defects 114

k = 25 c= = = 4.6
no. of lots 25
UCLc = c + 3 c = 4.6 + 3 4.6 = 11

LCLc = c - 3 c = 4.6 - 3 4.6 = -1.8 = 0
CSSBB 2014
X. CONTROL VIII.A.4
c Chart Example
Attributes Control Chart Form p np c u
PART : Encyc lopedia DESCRIPTION: SPC Chec klis t CH ARAC TERISTIC : De fe cts DATE: 10 /1
SOURCE: Binding Department OPERATOR: INSPEC TOR : You
UC L: LCL: 0 AVERAGE:
14
12
UCL = 11
10
0 LCL = 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Sample
(n) One Fixed Standard Sample
Numbe r
(np,c) 5 8 7 5 7 3 3 4 2 2 3 3 2 3 1 9 6 7 7 4 7 1 6 5 4
Fra ction
% (p,u) c = 4.6
Da te / 10 Shift
Time 1 Cha nge
Notes
Note, a shift change occurred between plot points 15

and 16. Is this significant?
CSSBB 2014
X. CONTROL VIII.A.4
Moving Average Charts

Past data may be summarized by computing the mean
of successive sets of data. Single moving average is a
method of smoothing the data to be used as an estimate
of future values. The formula for single moving average
is:
1
Mt = X t + X t-1 +  + X t-N+1 
N 
X is individual data values

t is the current time period
N is the moving group size
As N is increased, there is less sensitivity to changes in

individual X data values. Moving average is best used
when the process mean is stable, but is a poor predictor
when the process exhibits trends.
CSSBB 2014
X. CONTROL VIII.A.4
Moving Average Charts (Continued)

An example of single moving average with N = 3 is
shown below:
10.0
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
0 5 10 15 20 25
Data period t
Xt Mt
Note that there are relatively small changes in the Mt

while there are large changes in the Xt values. The first
Mt value is not calculated until there are N data points.
The Mt values are plotted at the t value corresponding to
the last data point in each group of N data. The data
values also appear to exhibit a downward trend.
CSSBB 2014
X. CONTROL VIII.A.4
Short Run SPC

Most traditional SPC techniques require long,
reasonably stable, production runs. Short run charting
may be desirable when the production lot size is
extremely small (10 to 20 pieces) or when the sample
size is small. Two limited data charts have already been
discussed:
X - MR Charts
MX̄ - MR Charts
According to Bothe, other options exist: first and last

piece inspection, 100% inspection, or a separate chart
for each part number or measurement. These
alternatives can be costly or messy.
Various techniques have been suggested by a number

of authors. However, the recommendations of some are
not without controversy. The emphasis has been on
short runs and multiple variables per chart, as this is
increasingly the greatest need in an era of
customization.
The Primer includes examples of set target-set range

and coded variation charts. Bothe also recommends
normal variable charts and short run attribute charts.
CSSBB 2014
X. CONTROL VIII.A.5
SPC/CONTROL CHART ANALYSIS
Control Chart Analysis

If a process is out-of-control, then special causes of
variation are present in either the average chart or range
chart, or both. These special causes must be found and
eliminated in order to achieve an in control process. A
process out-of-control is detected on a control chart by
either having any point outside the control limits or by
unnatural patterns of variation.
± 1S = 68.27 % 3 Upper Control Limit

2
1
± 2S = 95.45 % 1 Grand Average
2
3
± 3S = 99.73 % Lower Control Limit

CSSBB 2014
X. CONTROL VIII.A.5
Control Chart Analysis (Continued)

Control limits are 3 standard deviations above and
below the grand average. If the process is in control,
99.73% of the averages will fall inside these limits. The
same is true for the range control limits. Because there
are two components to every control chart -- the average
and the range -- there are four possible conditions which
could occur in the process.
1. Average Out-of-Control
Range In Control
2. Average In Control Process

Range Out-of-Control Out-of-Control
3. Average Out-of-Control
Range Out-of-Control
4. Average In Control Process

Range In Control In Control
CSSBB 2014
X. CONTROL VIII.A.5
1. Process Average Out-of-control

Average Shifting
Variation Stable
2. Process Variation Out-of-control

Average Stable
Variation Changing
CSSBB 2014
X. CONTROL VIII.A.5
3. Average & Variation Out-of-control

Average Shifting
Variation Changing
4. Process In Control
Average Stable
Variation Stable
CSSBB 2014
X. CONTROL VIII.A.6
Control Chart Interpretation

Five Common Rules
(Rule 2) 4 out of 5 (Rule 4) 8 or more consecutive
points in zone B points on one side of center line
UCL
Zone A
Zone B
Zone C
Zone C
Zone B
Zone A
(Rule 3) 2 UCL
(Rule 5) A trend is 6
(Rule 1) A point out of 3 points or more consecutive points
beyond the control limit in zone A increasing or decreasing
Comment: Some authorities say 7 or more consecutive points for both Rules 4 and 5.
Other Unusual Patterns

UCL
Zone A
Zone B
Zone C
Zone C
Zone B
Zone A
(Rule 6) Stratification UCL
(Rule 7) Mixture or
15 or more points in zone C
systematic variation
CSSBB 2014
X. CONTROL VIII.A.5
Process In Control
This is an example of a process which is in control.

Notice that it looks good, but not too good.
Trends
X CHART CAUSES R CHART CAUSES
C Deterioration of machine C Improvement or deterioration of operator skill

C Tired operator C Tired operator
C Tool wear C Change in incoming material quality
CORRECTIVE ACTION
C Repair or use alternate machine if available

C Discuss operation with operator to find cause
C Rotate operator
C Change, repair, or sharpen tool
C Investigate material
CSSBB 2014
X. CONTROL VIII.A.5
Jumps in Process Level
C Materials coming from different sources C Change in material

C New operator or machine C Change in method
C Modification of production method or process C Change in operator
C Change in inspection device or method C Change in inspection
CORRECTIVE ACTION
C Keep material supply consistent C Examine operator methods and instruction

C Investigate source of material C Check calibration of measurement device
C Check out machine capability
Recurring Cycles
C Physical environment (Temperature, Humidity) C Scheduled maintenance

C Tired operator C Tired operator
C Regular rotation of machine or operator C Tool wear
CORRECTIVE ACTION
C If environment is controllable, adjust it C Evaluate machine maintenance

C Service equipment C Replace, sharpen, or repair tool
C Rotate operators
CSSBB 2014
X. CONTROL VIII.A.5
Points Near or Outside Limits
C Over control C Mixture of material of distinctly

C Large systematic differences in material quality different quality
C Large differences in test methods or
equipment
CORRECTIVE ACTION
C Check control limits C Evaluate inspection frequency or methods

C Investigate material variation C Eliminate operator overadjustment of the process
C Evaluate test procedures
Lack of Variability
C Incorrect calculation of control limits C Collecting measurements from

C An improvement in the process widely differing lots
C Employee may not be making checks C Improvement in process since
limits were calculated
CORRECTIVE ACTION
C Check control limits C Verify proper employee measurement

C Validate rational sample subgroupings C Congratulate someone for improvement
C Verify checking procedure, gages, etc.
CSSBB 2014
X. CONTROL VIII.A.5
Runs Test for Randomness

A run is a sequence of data that exhibit the same
characteristic. This analysis can apply to both variable
and attribute data. For variable data, the points of
interest may be above or below an average or median
line. For attribute data, the measurement can be
illustrated by those individuals who prefer Diet Pepsi ®
or Diet Coke ®. Consider the results of two surveys:
Test I PPPPPPPPPCCCCCCCCC
Test II PCPCPCPCPCPCPCPCPC
In both examples, eighteen samples were taken. In Test

I, there were only 2 runs. In Test II, there were 18 runs.
Both examples suggest non-random behavior. To
perform a runs test, the following sequence applies:
1. Determine the value of n1 and n2 (either the total

of two attributes or the readings above and below
the center line on a run or control chart).
2. Determine the number of runs (R).
3. Consult a critical value table for a test statistic.

CSSBB 2014
X. CONTROL VIII.A.5
Runs Test (Continued)

Refer to the following Table for the expected numbers of
runs. The expected number of runs can be
approximated by adding the smallest and largest values
together and dividing by two.
n1 + n2 Smallest Average Largest Run
Plotted Points Run Limit # Runs Limit
8 1 5 9 (not possible)
10 2 6 10
12 3 7 11
14 3 8 13
16 4 9 14
18 5 10 15
20 6 11 16
22 7 12 17
24 7 13 19
26 8 14 20
28 9 15 21
30 10 16 22
34 11 18 25
40 14 21 28
50 20 26 32
The Table entries are critical values of R assuming a two

tailed test with a significance level of α= 0.05. The null
hypothesis is rejected if the total number of runs (R) is
less than the smaller entry or greater than the larger
entry.
CSSBB 2014
X. CONTROL VIII.A.5
Runs Test for Randomness (Continued)
Runs Test Example

As an illustration of the use of the critical value table,
consider the following run chart.
11
10
5
5 10 15 20 24
There are 24 plot points with 5 total runs above and

below the center line. The critical value table 10.22
indicates that between 8 and 18 total runs are expected.
If there are 7 or fewer or 19 or more runs, then it can be
said with 95% confidence that non-random variation
exists. The above example would fail the runs test.
CSSBB 2014
X. CONTROL VIII.B.1
OTHER CONTROL TOOLS/TPM
Other Control Tools
Other control tools are described in the following

topics:
C Total productive maintenance (TPM)

C Visual controls
Total Productive Maintenance (TPM)

Total productive maintenance (TPM) is an activity that
promotes coordinated group activities for greater
equipment effectiveness and requires operators to share
responsibility for routine machine inspection, cleaning,
maintenance, and minor repairs. The professional
maintenance staff retains responsibility for major
maintenance activities and serve as coaches for the
routine and minor items.
CSSBB 2014
X. CONTROL VIII.B.1
TPM (Continued)
The most important features of total productive
maintenance (TPM) are:
1. Efforts to maximize equipment effectiveness
2. A system of productive maintenance for a

machine’s life span
3. Implementation by engineering, operations, and

maintenance
4. Involvement of every employee, from top

management to the floor employees
5. Autonomous maintenance by operators
6. Company led small group activities

CSSBB 2014
X. CONTROL VIII.B.1
TPM (Continued)
The “total” in total productive maintenance has the
following meanings:
1. Total effectiveness in the pursuit of economic

efficiency and profitability.
2. Total maintenance prevention, maintainability,

and preventive maintenance.
3. Total participation of all employees includes

autonomous maintenance by operators.
The goal of maximizing equipment effectiveness

requires the complete elimination of failures, defects,
waste and loss due to equipment related operations.
The objectives of TPM are zero breakdowns and zero
defects. There are reports of firms achieving just 2% of
the original number of breakdowns and a 90% reduction
in process defects through the use of TPM.
CSSBB 2014
X. CONTROL VIII.B.1
TPM (Continued)
There are six big losses that contribute negatively to
equipment effectiveness:
1. Equipment failure: Breakdowns cause time losses

reducing productivity, and quality losses due to
defective products. The mind set should be zero
breakdowns.
2. Setup and adjustment: These are losses from

setup changes. Reduce the setup times and have
better adjustment periods.
3. Idling and minor stoppages: Defective sensors,

parts caught on a conveyor, etc., will cause
slowdowns and losses. Zero minor stoppages
are essential for unmanned production.
4. Reduced speed: This is the loss from the

differences between designed and actual speeds.
5. Process defects: Malfunctioning equipment will

produce scrap and quality defects.
6. Reduced yield: Product losses occur from

machine shut down and startup.
CSSBB 2014
X. CONTROL VIII.B.1
TPM (Continued)
The following Table shows the goals for each of the six
big losses.
Six Big Losses Goal

Breakdown losses 0
Setup and adjustment losses <10 minutes
Speed losses 0
Idling and minor stoppage 0
losses
Quality defect losses 0
Yield losses minimize
Elimination of the six big losses, will lead to dramatically

improved plant conditions. The lean manufacturing
system cannot exist without TPM.
CSSBB 2014
X. CONTROL VIII.B.1
Designing for Maintainability

The following guidelines will increase maintainability
and availability:
C Standardization: Look for compatibility of mating

parts and minimize the number of different parts in
the system. This will reduce spare parts inventory.
C Modularization: Have standards on sizes, shapes,

modular units. This will allow for standardized
assembly and disassembly procedures.
C Functional packaging: Place all needed components

of an item into a kit or package.
C Interchangeability: Control functional tolerances.

This refers to plug-in devices where spares are
instantly interchangeable with failed parts.
CSSBB 2014
X. CONTROL VIII.B.1
Designing for Maintainability (Cont’d)

C Accessibility: Allow room for workers to be able to
perform the task properly. All items should be
accessible. A part should be easy to replace.
C Malfunction annunciation: Provide a means to

notify the operator when the unit fails. This could
include gages, instrument panels, lights or sound.
C Fault isolation: A malfunction can be traced. This is

the most time-consuming task of all maintenance
work. This problem can be minimized by preventive
maintenance procedures, built-in test equipment
(BITE), simplicity in design of parts, and trained
personnel.
C Identification: Have a unique identification of all

components and a method of recording corrective
and preventive maintenance.
CSSBB 2014
X. CONTROL VIII.B.1
TPM Metrics
Overall equipment effectiveness is the prime measure
used to evaluate TPM. There are several formula
variations. The formulas are:
 Overall Equipment  =  Availability  x Performance  x Rate of Quality 
 Effectiveness   Efficiency   Products 
     
Operation Time Loading Time - Downtime

Availability = =
Loading Time Loading Time
Loading time is the available time per shift or per unit

minus planned downtime. Planned downtime includes
scheduled maintenance and morning meetings.
Operation time is loading time minus unscheduled
downtime.
CSSBB 2014
X. CONTROL VIII.B.1
TPM Metrics (Continued)

If there are 480 minutes per shift (available time), 15
minutes of setup time, 10 minutes of planned downtime,
30 minutes of unscheduled equipment failures, what is
the loading time and the availability?
Loading Time = Available Time per Shift - Planned Downtime
Loading Time = 480 - 10 = 470 minutes
470 -  30 + 15) 
Availability = = 0.904 = 90.4 %
470
CSSBB 2014
X. CONTROL VIII.B.1

Performance efficiency is defined as the operating
speed rate multiplied by the net operating rate. The
operating speed rate is the ratio of the theoretical cycle
time to its actual operating cycle time.
Theoretical Cycle Time

Operating Speed Rate =
Actual Cycle Time
If the theoretical cycle time is 1 minute per unit, and the

actual cycle time is 1.5 minutes per unit, what is the
operating speed rate?
1 minute
Operating Speed Rate = = 0.667 = 66.7%
1.5 minutes
CSSBB 2014
X. CONTROL VIII.B.1

The net operating rate measures the stability of the
equipment, the losses from minor stoppages, small
problems and adjustment losses.
Actual Processing Time

Net Operating Rate =
Operating Time
Processed Amount x Actual Cycle Time

Net Operating Rate =
Operating Time
If the processed amount is 185 units, the actual cycle
time is 1.5 minutes per unit, and the operation time is
425 minutes, what is the net operating rate?
185 x 1.5
Net Operating Rate = = 65.3%
425
Now, the performance efficiency can be calculated:
Performance Efficiency = Operating Speed Rate x Net

Operating Rate
For the previous example:
Performance Efficiency = 0.667 x 0.653 = 43.6%

CSSBB 2014
X. CONTROL VIII.B.1

Overall equipment effectiveness (OEE) is equal to
(availability) x (performance efficiency) x (rate of quality
products).
If the rate of quality products is 95%, the OEE is:
Overall Equipment Effectiveness = 0.904 x 0.436 x 0.95

= 0.374 = 37.4%
The overall equipment effectiveness (OEE) for this

example is a poor 37.4%. TPM prize winning companies
have OEE’s above 85%. The ideal conditions are:
C Availability greater than 90%

C Performance efficiency greater than 95%
C Rate of quality greater than 99%
CSSBB 2014
X. CONTROL VIII.B.1
Steps to Implement TPM

The Japan Institute of Plant Maintenance (JIPM) awards
an annual preventative maintenance (PM) prize. The
criteria is based on improvements from TPM, such as
increased productivity and quality. Factors include the
following:
1. Reduced costs
2. Reduced inventory
3. Accident reduction/elimination
4. Pollution control
5. Work environment
The proper integration of the TPM philosophy within the

company will bring about improved worker and
equipment utilization. These changes are aided by
improving employee attitudes, increasing their skills,
and providing a supporting work environment.
Steps to implement TMP and autonomous TPM small

group activities are reviewed in the Primer.
CSSBB 2014
X. CONTROL VIII.B.2
OTHER CONTROL TOOLS/VISUAL CONTROLS
Visual Controls
Visual control systems can be described as the use of
production boards, schedule boards, tool boards,
jidohka devices, and kanban cards on the factory floor.
The intent of these techniques is to provide
management and workers with a visible display of what
is happening at any moment. Visual controls assist the
operators by providing them with information that they
would not normally receive.
Imai provides three reasons for using visual

management tools:
C To make problems visible
C To help workers and management stay in direct

contact with gemba (the workplace)
C To clarify targets for improvement

CSSBB 2014
X. CONTROL VIII.B.2
Visual Controls (Continued)

Production boards and schedule boards are examples
of a visual factory. These generally include the posting
of daily production, maintenance items, or quality
problems for everyone to see and understand.
Jidoka (jidohka) is a device that stops a machine

whenever a defective product is produced. It is
autonomation, that is, a form of automation with human
elements attached.
When an equipment malfunction occurs, a light turns

red or a signal comes on to indicate a problem. The
operator or maintenance personnel must respond to find
the source of the problem and to resolve it.
CSSBB 2014
X. CONTROL VIII.B.2
Visual Controls (Continued)

The kanban system and station tool boards are forms of
visual control.
The visual factory places an emphasis on setting and

displaying targets for improvement. The concept is that
various operations have a target or goal to achieve.
The visual factory enables management and employees

to see the status of the factory floor at a glance. The
current conditions and progress are evident.
CSSBB 2014
X. CONTROL VIII.C.1
MAINTAIN CONTROLS/MEASUREMENT SYSTEM RE-ANALYSIS
Maintain Controls
Maintain controls is described in the following

topics:
C Measurement system re-analysis

C Control plan
Measurement System Re-analysis

Process improvements often result in reduced variation.
With an increased understanding of a process, variation
that previously was considered to be random or chance
cause variation, can be reduced. Continuous
improvement efforts resulting in reduced process
variation, may require a re-evaluation of the
measurement systems.
CSSBB 2014
X. CONTROL VIII.C.1
MSA (Continued)
The new quality system standards require “...an
estimation of the measurement uncertainty as well as
statistical techniques for analysis of test and/or
calibration data.” Similar wording was in MIL-STD-
45662A “...the collective uncertainty of the measurement
standards shall not exceed 25 percent of the acceptable
tolerance for each characteristic being calibrated.”
CSSBB 2014
X. CONTROL VIII.C.1
MSA (Continued)
The Measurement Systems Analysis Reference Manual
does not state specific requirements for the
measurement system requirements, but does provide
guidance for calculation of the bias, linearity, stability,
repeatability and reproducibility. For example,
guidelines for acceptance of gage repeatability and
reproducibility (% R&R) are:
C < 10% error The measurement system is

acceptable.
C 10% to 30% May be acceptable based upon

error importance of application, cost of gage,
cost of repairs, etc.
C > 30% error Measurement system needs

improvement. Make every effort to
identify the problems and have them
corrected.
The expression of measurement uncertainty includes

both a range and the level of confidence at which the
statement is made.
CSSBB 2014
X. CONTROL VIII.C.2
MAINTAIN CONTROLS/CONTROL PLAN
Control Plan
A control plan is a document describing the critical to
quality characteristics, the critical Xs or Ys, of the part
or process. Through this system of monitoring and
control, customer requirements will be met and the
product or process variation will be reduced. However,
the control plan should not be a replacement for detailed
operator instructions in the form of work instructions or
standard operating procedures. Each part or process
must have a control plan. A group of common parts
using a common process can be covered by a single
control plan.
CSSBB 2014
X. CONTROL VIII.C.2
Types of Control Plans

For the automotive sector, ISO/TS 16949 and the
Advanced Product Quality Planning APQP, identify three
control plan phases:
C Prototype
C Pre-launch
C Production
A prototype control plan is used in the early

development stages when the part or process is being
defined or configured. This control plan will list the
controls for the necessary dimensional measurements,
types of materials, and required performance tests.
CSSBB 2014
X. CONTROL VIII.C.2
Types of Control Plans (Continued)

A pre-launch control plan is used after the prototype
phase is completed and before full production is
approved. It lists the controls for the necessary
dimensional measurements, types of materials, and
performance tests.
A production control plan is used for the full production

of a part. It contains all of the line items for a full control
plan: part or product characteristics, process controls,
tests, measurement system analysis, and reaction plans.
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Organization

The project control phase is necessary in order to
sustain the project gains. The control plan must truly be
a “living document” for it to remain an effective
mechanism to monitor and control the process.
A responsible person must be placed in charge of the

control plan. This ensures successful monitoring and
updating. A black belt may or may not be a suitable
person for the role, as he/she may be replaced or
transferred to a different position. A better selection
would be the process owner.
The current process owner can be listed on the control

plan, but in reality it is a functional role that is to be
passed on to the next individual in that same
organizational position. If the control plan is not
maintained, the benefits of the project could slowly be
lost.
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Organization (Continued)

Some considerations in the closing phase of the project
include:
C Identify the process owner

C Involve the team in the control plan
C Create updated work instructions and procedures
C Notify and train the affected personnel
C Ensure that the control plan training is effective
C Place the plan in the proper system document
C Attain agreement between the team members and
process owner
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Input Sources

A number of inputs or sources contribute to
understanding, manufacturing, and controlling the part
or process. Many of the following are included:
C Process flow diagrams

C System FMEAs, DFMEAs, and PFMEAs
C Cause-and-effect analysis
C Special customer characteristics
C Historical data
C Lessons learned
C Team process knowledge
C Design reviews
C Quality function deployment
C Designed experiments
C Statistical applications
C Multi-vari studies
C Regression analysis
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Input Sources (Continued)

Customer requirements may dictate the exact form of
the control plan. Often, there is some flexibility in the
construction of the forms. An illustrative control plan is
shown below.
Control Plan (Sample)
Control Plan for:
Control number: Team members: Page:
Original date:
Contact person (typically process owner): Revision date:
characteristic note
Sample frequency
Key input variable
Subprocess step
Control method
Gage capability
gage technique
responsible for
Specifications
Reaction plan
measurement
Measurement
Part/Process
Sample size
variable (Y)
Key output
Initial Cpk
Special
Person
(X)
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Line Items

1. Control plan: Provide a title for the control plan.
The control plan will often be placed into another
document, such as an operating instruction.
2. Control number: Provide a reference number. This

number may be supplied by the responsible party.
3. Team members: If a cross functional team is

involved, provide the member’s names.
4. Contact person: This could be the black belt in

charge of the project, however, the name and
function of the process owner are more important.
5. Page: Provide page numbers if the control plan

exceeds one page. Control plans may run up to 20
pages.
6. Original date: Indicate the original date of issue of

the control plan.
7. Revision date: Provide the latest revision date of

the control plan.
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Line Items (Continued)

8. Part/process: List the part number or the process
flow being charted.
9. Subprocess step: Indicate the subprocess step

being described (if applicable).
10. Key input variable (X): Note the key input variable,
when appropriate. On any line item, only the X or Y
variable is filled out, not both.
11. Key output variable (Y): Note the key output

variable, when appropriate.
12. Special characteristic note: Indicate if a special

characteristic is to be monitored and controlled.
13. Specifications: For manufacturing applications, the

engineering specifications for the part should be
monitored and controlled. For other applications,
one would provide specification limits, as well as
the target value.
CSSBB 2014
X. CONTROL VIII.C.2

14. Measurement gage technique: The gage or
measurement technique should be described. The
test device used for data collection must be in
conformance with the needed measurement system
analysis requirements.
15. Gage capability: Provide the capability of the

measurement system. The MSA manual lists:
C Under 10% error as acceptable

C 10% to 30% error may be acceptable
C Over 30% error is not acceptable
Devices may need uncertainty determinations.
16. Sample size: Provide the sample size for each

subgroup.
17. Sample frequency: List how often the inspection or

monitoring of the part or process is required.
CSSBB 2014
X. CONTROL VIII.C.2

18. Initial Cpk: This provides an indication of process
capability.
19. Person responsible for measurement: Indicate who

will make and record the measurement.
20. Control method: Note how this X or Y variable will

be controlled. Examples include control charts,
checklists, visual inspections, automated
measurements, etc.
21. Reaction plan: Describe what will happen if the

variable goes out of control. How should the
responsible person respond to the situation?
CSSBB 2014
X. CONTROL VIII.C.2
Illustrative Control Plan

Control Plan (Example)
Control Plan for: Black Belt Primer
Control number: BB-001 Team members: Glenn, Wes, Page: 1 of 1
Tim, Bob, Odis, Bill
Original date: August 20, 2012
Contact person (typically Process Owner): Bill Revision date: January 20, 2013
Key output variable (Y)
Person responsible for

Special characteristic
Key input variable (X)
Sample frequency
Subprocess step
Control method
Gage capability
gage technique
Specifications
Measurement/
Reaction plan
measurement
Part/Process
Sample size
Initial Cpk
note
Receive ring Heavy Notify Bill

Primer binders metal duty slant visual NA 5 lot NA clerk checklist contact
10320 Notify Bill
binder Gunny Arrestoxx contact
fabric Sak Master 10% 5 lot 1.5 clerk checklist supplier
Notify Bill
binder PMS 492 Pantone contact
print (Red) color 10% 5 lot 1.5 clerk checklist supplier
Notify Bill
binder 3.13" steel contact
width +/- 0.03" ruler 6% 5 lot 1.7 clerk checklist supplier
Notify Bill
binder 11.63" steel contact
height +/- 0.03" ruler 6% 5 lot 1.7 clerk checklist supplier
Example Control Plan for Receiving Primer Binders
In the example above, note that only the key input

column is controlled.
CSSBB 2014
X. CONTROL VIII.C.2
Control Plan Summary

Control plan construction is often led by the black belt
in charge of the six sigma project. The team is usually
cross functional with individuals from different areas,
including the process owner. The team will ensure that
the control plan contains the critical variables, the Xs
and the Ys, of the product or process. The control plan
must show compliance and control before project
closure.
A successful control plan will remain a living document

to ensure that the benefits of the project will be fully
realized.
CSSBB 2014
X. CONTROL VIII.D.1
SUSTAIN IMPROVEMENTS/LESSONS LEARNED
Sustain Improvements
Sustain improvements tools is described in the

following topics:
C Lessons learned
C Training plan deployment
C Documentation
C Ongoing evaluation
Lessons Learned
One of the last project closure steps is the postmortem
analysis (also called lessons learned, and post-project
appraisal). This review is often a formal and
documented critique conducted by a committee of
qualified company personnel. The project review
extends over all phases of development.
CSSBB 2014
X. CONTROL VIII.D.1
SUSTAIN IMPROVEMENTS/LESSONS LEARNED
Lessons Learned (Continued)

Some of the fundamental review topics include:
C The adequacy of personnel, time and money

C The effectiveness of the entire project
C How well the project was tracked
C How well management and was informed of status
C How well the project team performed together
C How well project team efforts were recognized
C The effectiveness of actions
C The quality level of the delivered product or service
The review process is often established by management

policy. The membership and responsibilities in a project
review will vary considerably, based on the type of
review under consideration.
CSSBB 2014
X. CONTROL VIII.D.2
SUSTAIN IMPROVEMENTS / LESSONS LEARNED

The success or failure of a project is usually measured
in three dimensions:
C Were the specified goals and objectives achieved?

C Was it achieved within the time deadlines?
C Was the project at or below cost constraints?
Well executed project plans meet all of the above

criteria. However, it is possible for a project to be
considered a success, even when the project is late,
over budget, and not meeting all objectives.
Nearly every project encounters unanticipated events or

problems, but this is not an acceptable excuse for failing
to meet the performance standards. Performance is
measured on results, not effort.
CSSBB 2014
X. CONTROL VIII.D.2
SUSTAIN IMPROVEMENTS / LESSONS LEARNED

Often, within the six sigma framework, the assigned
improvement team will identify what the team did well
and what they should do differently in the future. The
team should indicate if the additional process
improvements justify extending the project. Equally
important, other processes or applications should be
reviewed for similar potential improvements.
Project results are presented by the improvement team

to upper management. Often outside black belts and
master black belts are included in this presentation.
When the best corrective actions have been identified,
these should be standardized by the company.
Results of this project review will be retained, along with

the other project documentation, and archived for future
reference.
CSSBB 2014
X. CONTROL VIII.D.2
SUSTAIN IMPROVEMENTS/TRAINING PLANS
Training Plans
Training is the process aimed at improving the skills and
expanding the knowledge of employees. Training
provides very specific employee development intended
to close the gap between current and desired abilities.
George indicates, for six sigma to truly take hold in a

company, both formal learning and practical knowledge
should be communicated and shared broadly within a
company. He details the following strategies:
C Build a robust curriculum

C Provide a means for people to interact
C Exploit technology
C Document and share best practices
Top-down support for training should come from the

realization that training can provide the human
resources needed to attain the organization’s goals.
CSSBB 2014
X. CONTROL VIII.D.2
Training Plans (Continued)

Upper management’s support for training may be driven
by the objective of stability in available work skills.
Trained workers become more valuable to the
organization.
Training programs must be supported by all levels of

management if they are to be successful. The desire for
training by a subordinate cannot possibly overcome
resistance by higher management for that training.
After the creation of a new or revised process, training

deployment requires identification of individuals that
will need to be trained in order to sustain and maintain
the change. It is also important to determine the depth
of knowledge required for this training.
One of the most effective ways to gain acceptance of the

change is to include a number of key personnel on the
initial improvement team.
All affected personnel should receive hands-on training

emphasizing not only what must be done differently, but
why it is necessary to do it differently.
CSSBB 2014
X. CONTROL VIII.D.2
Training Plans (Continued)

Employees must understand the benefits of process
changes, and understand their responsibilities in the
activities. The benefits may include the following:
C Better understanding of the commitment to improve

C Better knowledge of the process and its intent
C Clear understanding of expectations
C Better knowledge of how to do a task consistently
C Better knowledge of how to adjust a process for the
desired performance
C Clear understanding of where to get help, if needed,
to control a process
Completion of the training will also provide an

opportunity to identify and correct any errors or
clarification needed in the documentation.
CSSBB 2014
X. CONTROL VIII.D.3
SUSTAIN IMPROVEMENTS/DOCUMENTATION
Documentation
An effective control system is characterized by formal
documents. These documents provide directions to the
employees on how to accomplish a task, who is
responsible for performing the task, or how the
company systems work. There are various names for
these documents including:
C Manuals (including the quality manual)

C Procedures
C Standard operating procedures
C Work instructions
C Records
Many companies organize the documentation into a

hierarchy. The manual is the highest level document in
the system. Procedures are at the second level and
describe the responsibilities of various personnel and
the administrative system used to accomplish the tasks.
The manual details what is to be done and procedures
describe who will do it.
The third level is the work instructions that describe

how to do the tasks. The work instructions detail the
specific steps to accomplish the goals defined in the
manual and the procedures.
CSSBB 2014
X. CONTROL VIII.D.3
Documentation (Continued)
The basic content of any good procedure or instruction
should include:
C Purpose of the document

C Basis of the document
C Scope of the document
Documentation is necessary for the continued success

of a company. Formal procedures or instructions have
a number of benefits and characteristics as described in
the Primer.
Documented procedures allow the improvement of

processes (both administrative and technical) by first
establishing a baseline that can be used for subsequent
improvements.
CSSBB 2014
X. CONTROL VIII.D.3
Documentation (Continued)
Pande provides the following general guidelines for
documentation:
C Keep the documentation simple

C Keep the documentation clear and inviting
C Include options and instructions for “emergencies”
C Keep the documentation brief
C Keep the documentation handy
C Have a process for updates and revisions
Documentation must be written to the level that is

understood by the users. It should also reflect the
current processes and methods. After completion of a
process improvement, the documentation should
correspond to the new methods, and the users should
be trained on the new documentation.
CSSBB 2014
X. CONTROL VIII.D.4
SUSTAIN IMPROVEMENTS/ONGOING EVALUATION
Ongoing Evaluation
Effective project or process improvement activities
should ultimately lead to the advancement of company
operations. However, this is not automatic. Continuous
improvement takes the concentrated and continuing
efforts of everyone. The improvement process is
illustrated below.
IMPLEMENT
PROCESS
DOCUMENT EVALUATE
PROCESS PROCESS
IMPROVE
PROCESS
CSSBB 2014
X. CONTROL VIII.D.4
SUSTAIN IMPROVEMENTS/ONGOING EVALUATION
Ongoing Evaluation (Continued)

In addition to changes resulting from improvements,
there should be a balanced mix of measurements to
monitor overall process performance. Examples
include:
C Performance results
C Quality results
C Changes in customer requirements
C Financial results
C Benchmarking results
C Process capability measurements
C Audit results
C SWOT analysis
The above results are often reported in management

reports. In almost all situations, graphs and charts are
preferable to texts and columns of numbers.
The organization should have an oversight committee to

respond to both problems and opportunities.
CSSBB 2014
X. CONTROL
QUESTIONS
10.2. The basic objective of visual factory techniques is to:
a. Utilize production and schedule boards

b. Enable everyone to maintain contact with the workplace
c. Regulate production using kanban cards
d. Make problems visible using tool boards and jidohka devices
10.3. A control chart is used to:
a. Determine if defective parts are being produced

b. Measure process capability
c. Determine causes of process variation
d. Detect non-random variation in processes
10.5. After the creation of a new or revised process, what action step
should follow next?
a. Management should be trained in the techniques

b. A human interaction curriculum must be developed
c. A group of adequately qualified instructors should be
recruited
d. The appropriate individuals needing training should be
identified
Answers 10.2 b, 10.3 d, 10.5 d

CSSBB 2014
X. CONTROL
QUESTIONS
10.11. A six sigma project has progressed to the point that a control
plan is required. Control plan activities can be considered closed
after which of the following?
a. A process owner is named for the control plan

b. A responsible engineer is designated
c. The cross functional team signs off on the control plan
d. The control plan is a “living document” and is rarely closed
10.13. An average chart has been in control for some time. If the
sample ranges suddenly and significantly increase, the sample
mean will:
a. Usually increase
b. Stay the same
c. Always decrease
d. Occasionally show out-of-control of either limit
10.14. In a visual factory setting, kanban cards are typically used to:
a. Identify areas needing housekeeping attention

b. Clarify targets for future improvement
c. Regulate the flow of production and inventory
d. Make production problems visible
Answers 10.11 d, 10.13 d, 10.14 c

CSSBB 2014
X. CONTROL
QUESTIONS
10.24. An X-bar chart has been in control for a long time. However, the
points for the last 50 samples are all very near the center line on
the chart. In fact, they are all within one sigma of the center line.
This probably indicates that:
a. It is a desirable situation
b. It is an undesirable situation
c. The process standard deviation has decreased recently
d. The control limits are incorrectly computed
10.26. Tool boards, jidohka devices, and red lights all combine to:
a. Make problems visible

b. Prevent defective products
c. Maintain management control
d. Display targets for improvement
10.31. What is the equipment availability if there are 7.5 hours available
per shift, 30 minutes of setup time, 15 minutes of planned
downtime, and 15 minutes of unscheduled equipment failure?
a. 87%
b. 93%
c. 90%
d. 85%
Answers 10.24 b, 10.26 a, 10.31 c

CSSBB 2014
X. CONTROL
QUESTIONS
10.35. Identify the most logical reason for using short run SPC charts:
a. There is limited operator time

b. There is limited data
c. There is a need to plot multiple variables per chart
d. There is a need to simplify the charting process
10.37. When a new process change is introduced in an area, what is the

most critical element in the successful implementation of that
change?
a. Training
b. Documentation
c. Education
d. Procedures
10.40. Assume an operating speed rate of 80%. If 40 units are produced

at 2 minutes/unit in two hours, what is the performance efficiency
of the work unit?
a. 0.800
b. 0.667
c. 0.534
d. 0.435
Answers 10.35 b, 10.37 a, 10.40 c

© QUALITY COUNCIL OF INDIANA XI-1 (1017)
CSSBB 2014
XI. DESIGN FOR SIX SIGMA
WHEN WE BUILD, LET US THINK THAT

WE BUILD FOREVER.
JOHN RUSKIN
CSSBB 2014
XI. DESIGN FOR SIX SIGMA IX.A

COMMON DFSS METHODOLOGIES
Design for Six Sigma (DFSS)
Design for Six Sigma is covered in the following topic

areas:
C Common DFSS methodologies

C Design for X (DFX)
C Robust design
CSSBB 2014

Common DFSS Methodologies

Design for six sigma is the suggested method to bring
order to product design. 70% to 80% of all quality
problems are design related. Emphasis on the
manufacturing side alone will concentrate at the tail end
of the problem solving process.
One of the ways to increase revenues must include

introducing more new products for sale to customers.
Cooper provides details of how winning products are

obtained:
1. A unique, superior product

2. A strong market orientation
3. Predevelopment work
4. Good product definition
5. Quality of execution
6. Team effort for product development
7. Proper project selection
CSSBB 2014

Common DFSS Methodologies (Cont.)

8. Prepare for the launch
9. Top management leadership
10. Speed to market
11. A new product process (stage gate)
12. An attractive market
13. Strength of company abilities
There are many product development processes to

choose from. Multi-functional team activities involving
all departments are necessary for effectiveness and
speed to market. The process has two parts, idea
generation and new product development (NPD). NPD
includes 5 activities:
C Concept study
C Feasibility investigations
C Development of the new product
C Maintenance
C Continuous learning
CSSBB 2014

Stage Gate Process

A stage gate process is used by many companies to
screen and pass projects as they progress through
development stages. Each stage of a project has
requirements that must be fulfilled.
The gate is a management review of the particular stage

in question. It is at the various gates that management
should make the “kill” decision.
Product Development Stages:
C Get an idea
C Prove it works
C Financial assessment
C Develop and test
C Scale up
C Launch
C Post delivery support
C Continuous learning
The individual organization should customize their

process and allow a suitable time period for it to
stabilize.
CSSBB 2014

Product Development
In the area of new product management, some
commonly accepted new product terms are:
1. New-to-the-world products: These are inventions

and discoveries.
2. New category entries: These are company products

that are new to the company.
3. Additions to product lines: Extensions of the

organization’s existing product line.
4. Product improvements: Current products made

better.
5. Repositionings: Products that are retargeted for a

new use.
6. Cost reductions: New products are designed to

replace existing products, but at a lower cost.
CSSBB 2014

IDOV
Four step IDOV model:
C Identify: Use a team charter, VOC, QFD, FMEA, and

benchmarking.
C Design: Emphasize CTQs, identify requirements,

develop alternatives, evaluate, and select.
C Optimize: Use process capability information,

statistical tolerancing, and robust design.
C Validate: Test and validate the design.
DMADV
Five step define, measure, analyze, design and validate
(DMADV) process for six sigma design:
C Define: Define project goals and customer needs

C Measure: Measure and determine customer needs
C Analyze: Analyze the process to meet needs
C Design: Develop the process to meet needs
C Verify: Verify and validate the design performance
CSSBB 2014

DMADOV
The process steps for a DMADOV (define, measure,
analyze, design, optimize, and verify) project include:
C Define the project

C Measure the opportunity
C Analyze the process options
C Design the process
C Optimize the process
C Verify the performance
CSSBB 2014

The French Design Model

The design engineer will select a design process, for
example, The French model:
Analysis of Statement of Conceptual

Need
problem problem design
Selected Embodiment Working

Detailing
schemes of schemes drawings, etc.
The design team captures the needs, provides analysis,

and produces a statement of the problem. The
conceptual design will generate a variety of solutions to
the problem and produce a working drawing from the
abstract concept.
The detailing step consolidates and coordinates the fine

points of producing a product.
The designer of a new product is responsible for taking

the initial concept to final launch. The project manager,
product manager, or general manager will need to
manage the process.
CSSBB 2014
XI. DESIGN FOR SIX SIGMA IX.B

DESIGN FOR X
Design for X (DFX)

Design for X (DFX) is defined as a knowledge-based
approach for designing products to have as many
desirable characteristics as possible. The desirable
characteristics include: quality, reliability, serviceability,
safety, user friendliness, etc.
AT&T Bell Laboratories coined the term DFX to describe

the process of designing a product to meet the above
characteristics. The life cycle cost of a product and the
lowering of downstream manufacturing costs are
addressed.
The DFX toolbox has continued to grow in number from

its inception 15 years ago to include hundreds of tools
today.
CSSBB 2014

DESIGN FOR X
Usage of DFX Techniques and Tools

1. Design guidelines: Rules of thumb provide broad
design rules and strategies. The rule to increase
assembly efficiency requires a reduction in the part
count and types.
2. DFX analysis tools: Each DFX tool involves some

analytical procedure that measures the
effectiveness of the selected tool.
3. Determine DFX tool structure: A technique may

require other calculations before the technique can
be considered complete. An independent tool will
not depend on the output of another tool.
4. Tool effectiveness and context: Each tool can be

evaluated for usefulness by the user based on
accuracy of analysis, reliability characteristics
and/or integrity of the information generated.
5. The focus of activity and the product development

process: Understanding the process activities will
help determine when a particular tool can be used.
6. Mapping tool focus by level: Several levels of

analysis may be involved with one individual tool.
CSSBB 2014

DESIGN FOR X
DFX Characteristics
The following characteristics and attributes should be
considered by DFX projects.
C Function and performance

C Safety
C Quality
C Reliability
C Testability
C Manufacturability
C Assembly (Design for Assembly, DFA)
C Environment
C Serviceability (Maintainability and Repairability)
C Maintainability
C User Friendliness or Ergonomics
C Appearance (Aesthetics)
C Packaging
C Features
C Time to Market
CSSBB 2014
XI. DESIGN FOR SIX SIGMA IX.C

ROBUST DESIGN
Robust Design Introduction

Dr. Genichi Taguchi wrote that the United States has
coined the term “Taguchi Methods” to describe his
system of robustness for the evaluation and
improvement of the product development processes.
He has stated that he preferred the term “quality

engineering” to describe the process.
CSSBB 2014

ROBUST DESIGN
Robust Design Approach

Robust design processes can produce extremely
reliable designs both during manufacture and in use.
Robust design uses the concept of parameter control to

place the design in a position where random “noise”
does not cause failure.
The concept is that a product or process is controlled by

a number of factors to produce the desired response.
The signal factor is the signal used for the intended

response. The success of obtaining the response is
dependent on control factors and noise factors.
Noise Factors
Signal Factor Products/Procedures Response
Control Factors
CSSBB 2014

ROBUST DESIGN
Robust Design Approach (Continued)

Control factors are those parameters that are
controllable by the designer. These factors are the
items in the product or process that operate to produce
a response when triggered by a signal.
Noise factors are parameters or events that are not

controllable by the designer. These are generally
random, in that only the mean and variance can be
predicted. Noise factors have the ability to produce an
error in the desired response.
CSSBB 2014

ROBUST DESIGN
Robust Design Example

The most celebrated case of design of experiments was
that of a parameter design experiment at a tile
manufacturing company in Japan, as documented by
Genichi Taguchi.
Factors which were less expensive to control were fixed

at levels such that the variation in tile dimension was
made insensitive to a noise factor, temperature
variation.
The problem was extreme variation in the dimensions of

the tile produced. Tiles in the kiln toward the outside of
the stack tended to have a different average dimension
and exhibited more variation than those toward the
inside of the stack.
Burners Burners
Inside Outside
Tiles Tiles
CSSBB 2014

ROBUST DESIGN
Robust Design Example (Continued)

The cause of variation was an uneven temperature
profile inside the kiln. The company would have to
redesign the kiln, which was very expensive. The
company budget didn't allow such costly action, but the
kiln was creating a tremendous financial loss.
Although temperature was an important factor, it was

treated as a noise factor. People having knowledge
about the process brainstormed and identified seven
major controllable factors which they thought could
affect the tile dimension. These were: (1) limestone
content in the raw mix, (2) fineness of the additives, (3)
amalgamate content, (4) type of amalgamate, (5) raw
material quantity, (6) waste return content, and (7) type
of feldspar.
After testing these factors using an orthogonal design,

they discovered that factor #1 (limestone content) was
the most significant factor, although other factors had
smaller effects. By increasing limestone content from
1% to 2%, the percent warpage could be reduced from
30% to less than 1%. Limestone was the cheapest
material in the tile mix. They found that they could use
a smaller amount of amalgamate without adversely
affecting the tile dimension. Amalgamate was the most
expensive material in the tile.
CSSBB 2014

ROBUST DESIGN
Concept Design
Concept design is the selection of the process or
product architecture based on technology, cost,
customer, or other important considerations.
Parameter Design
The design is established using the lowest cost
components and manufacturing techniques. The
response is then optimized for control and minimized for
noise.
Tolerance Design
The tolerances are reduced until the design
requirements are met.
With robust design approaches, the designer has the

ability to produce a design with either the lowest cost,
the highest reliability or an optimized combination of
cost and reliability.
CSSBB 2014

ROBUST DESIGN
Functional Requirements
In the development of a new product, the product
planning department must determine the functions
required. The designer will have a set of requirements
that a new product must possess. The designer will
develop various concepts, embodiments, or systems
that will satisfy the customer’s requirements.
The product design must be “functionally robust,”

which implies that it must withstand variation in input
conditions and still achieve desired performance
capabilities. The designer has two objectives:
C Develop a product that can perform the desired

functions and be robust under various operating or
exposure conditions
C Have the product manufactured at the lowest

possible cost
CSSBB 2014

ROBUST DESIGN
Functional Requirements (Continued)

Parameter Design
Parameter designs improve the functional robustness of

the process so that the desired dimensions or quality
characteristics are obtained. The process is considered
functionally robust if it produces the desired part with a
wide variety of part dimensions. The steps to obtain
robustness are:
1. Determine the signal factors and the uncontrollable

noise factors and ranges.
2. Choose as many controllable factors as possible,

select levels for these factors, and assign these
levels to appropriate orthogonal arrays.
3. Calculate S/N ratios from the experimental data.
S 1  S - Ve 
η= = 10 log10  β 
N r  VN 
4. Determine the optimal conditions for the process

derived from the experimental data.
5. Conduct actual production runs.

CSSBB 2014

ROBUST DESIGN

Signal-to-Noise Ratio
A signal-to-noise ratio (S/N) is used to evaluate system

performance. The combinations of the design variables
that maximize the S/N ratio are selected for
consideration as product or process parameter settings.
Case 1: S/N ratio for “smaller is better”:

S/N = -10 log (mean-squared response)
 n 2
S   yi 
η= = -10 log10  i=1 
N  n 
 
 
Case 2: S/N ratio for “larger is better”:
S/N = -10 log (mean-squared of the reciprocal
response)

n
1 
S   y2 
η= = -10 log10  i=1 i 
N  n 
 
 
Case 3: S/N ratio for “nominal is best”:
S   mean 2   y2 
η= = 10 log10   = 10 log10  2 
N  
 variance  s 
CSSBB 2014

ROBUST DESIGN

Parameter Design Case Study
A parameter design case study is provided in the

CSSBB Primer. The case study has 4 controllable
factors and 3 noise factors. The controllable factors are
at 3 levels; the noise factors at 2 levels.
Given 4 factors at 3 levels, amounts to 81 experiments.

Taguchi provided orthogonal arrays to reduce the
amount of testing required. They are fractional factorial
experiments without regard for interactions, in most
cases.
The case study includes an example orthogonal design

layout. The L9 array is called the inner array, while the
L8 array is the outer array.
The completed matrix contains the mean response

results. The larger the S/N ratio the better. S/N ratios
are computed for each of the 9 experimental conditions.
The optimum combination of factors and levels can be

determined from the analysis. A confirmation run
should be conducted to verify the results.
CSSBB 2014

ROBUST DESIGN

The Loss Function
The loss function is used to determine the financial loss

that will occur when a quality characteristic, y, deviates
from the target value, m. The quality loss is zero when
the quality characteristic, y, is at the target value, m.
L(y) = k  y - m 
2
cost of a defective product A

k= =
 tolerance 
2
∆2
= mean value of  y - m 
2
σ2
The loss function L(y) shows that the further the quality
characteristic is away from the target, the greater the
quality loss. “A” is the cost due to a defective product.
The amount of deviation from the target, or “tolerance,”
is the delta (∆) value. The constant k is derived as
shown.
The mean square deviation from the target (σ2), as used

by Taguchi, does not indicate a variance.
CSSBB 2014

ROBUST DESIGN

The Loss Function (Continued)
Given that Professor Barker wished to buy a pair of size

7 shoes. The store was out of size 7 and he had to settle
for a pair of 7-1/2 shoes. After 2 days, he found them to
be ill-fitting and had to discard them. The original cost
of the shoes was $50. Size 6-1/2 shoes were also not
suitable.
The target value m is 7.0

The existing quality characteristic y is 7.5
The cost of a defective product A is $50.
The hypothetical tolerance (7.5 - 7.0) is 0.5
L(y) = k  y - m 
2
cost of a defective product A $50

k= = = = $200
 tolerance   0.5 
2 2
∆2
L(y) = k  y - m  =  $200  0.5  = $50
2 2
The above calculations shows the quality loss to be $50.

CSSBB 2014

ROBUST DESIGN
Tolerance Design
The tolerances for all system components must be
determined. In tolerance design, there is a balance
between a given quality level and cost of the design.
The measurement criteria is quality losses. Quality
losses are estimated by the functional deviation of the
products from their target values plus the cost due to
the malfunction of these products. Taguchi described
the approach as using economical safety factors.
The functional limit ∆0 must be determined by methods

like experimentation and testing. Taguchi uses a LD50
point as a guide to establish the upper and lower
functional limits. The LD50 point is where the product
will fail 50% of the time. The 50% point is called the
median.
An example from Taguchi illustrates the determination

of the functional limit: A spark plug has a nominal
ignition voltage of 20 kV. The lower functional limit ∆01
is -12 kV. The upper functional limit ∆02 is +18 kV.
These values are determined by testing. The resulting
specifications will have a lower tolerance (∆1) of 8kV and
upper tolerance (∆2) of 38 kV.
CSSBB 2014

ROBUST DESIGN
Tolerance Design (Continued)

The formulas for tolerance specifications, the functional
limit, and safety factors are as follows:
Functional Limit
Tolerance Specification =
Safety Limit
∆
Tolerance Specification = ∆ i = 0i (i = 1, 2)
φi
∆0
or commonly ∆=
φ
The economical safety factor φ is determined as follows:
Loss when exceeding functional limit A0

φ= =
Loss when exceeding tolerance specs A
Given the value of the quality characteristic at y, and the

target value at m, the quality loss function will appear as
follows:
A
L(y) = 20  y - m 
2
∆0
CSSBB 2014

ROBUST DESIGN
Tolerance Design (Continued)

A power supply for a TV set has the functional limits at
+/- 25% of output voltage. The average quality loss A0
after shipment of a bad TV is known to be $300. The
adjustment of a power supply in-house before shipping
is $1.00. Phi, φ, is the economical safety factor and is
calculated as:
A0 300
φ= = = 17.3
A 1
The tolerance specs for the output voltage will be:

∆0 25%
∆= = = 1.45%
φ 17.3
Therefore, the tolerance specification for the output

voltage of 120 volts will be:
120 ± (120)(0.0145) = 120 ± 1.74 volts
Although the functional limits were initially established

at 120 ± 25%, or ± 30 volts, the TV sets should have
output voltages within 1.74 volts of the nominal.
CSSBB 2014

ROBUST DESIGN
Taguchi’s Quality Imperatives

C Robustness is a function of product design. Quality
losses are a loss to society.
C Robust products have a strong signal with low

internal noise.
C For new products, use planned experiments and

orthogonal arrays to seek out the parameter targets.
C To build robust products, simulate customer-use

conditions.
C Tolerances are set before going to manufacturing.

The quality loss function can be measured.
C Products that barely meet the standard are only

slightly better than products that fail the
specifications.
C The factory must manufacture products that are

consistent by reducing variation.
C Reducing product failure in the field will reduce the

number of defectives in the factory.
C Proposals for capital equipment for on-line quality

efforts should include the average quality loss.
CSSBB 2014

ROBUST DESIGN
Statistical Tolerancing
Statistical tolerancing uses the square root of the sum
of variances to determine the tolerances required.
Assume that plus and minus four sigma is necessary

and that three components are assembled.
Part # Thickness 1 Sigma 4 Sigma

A 10" ± 0.0011" ± 0.0044"
B 12" ± 0.0010" ± 0.0040"
C 8" ± 0.0014" ± 0.0056"
Thus:  2 Assembly =  2A + B2 +  2C
 A = 0.0011" B = 0.0010"  C = 0.0014"
Therefore:
 2 Assembly =  0.0011"  +  0.0010"  +  0.0014" 
2 2 2
 0.0011"  +  0.0010"  +  0.0014" 

2 2 2
 Assembly =
 Assembly = 0.00204"
4 Assembly = 0.0082"
The final assembly, without special effort, will be:
30" ± 0.0082"
CSSBB 2014

QUESTIONS
11.1. When following the design robustness approach, what step follows
parameter design?
a. Tolerance design
b. System design
c. Concept design
d. Feature design
11.7. The principal purpose of robust design techniques is to:
a. Make the product less sensitive to noise effects

b. Use the tools of experimental design
c. Reduce the sources of variation
d. Improve manufacturing quality
11.8. In both the DMADV and DMADOV development methodologies

there is a measure step. What is being measured?
a. The project goals

b. The estimated cost of the project
c. The depth of the potential creative model solutions
d. The customer needs and specifications
Answers: 11.1 a, 11.7 a, 11.8 d

CSSBB 2014

QUESTIONS
11.13. The stage gate process is used by many companies to screen

and pass projects. Many companies may fail to use the process
properly. A common problem in the stage gate process that
could negate its benefits would be:
a. Failure to kill a project

b. Too many kills in the system
c. Teams are overloaded with projects
d. Requirements are not established
11.14. DFX has many internal and external customer-based objectives.

Which of the following would be considered an external target?
a. Design for assembly

b. Design for features
c. Design for testability
d. Design for manufacture
11.18. The most significant difference between the six step DMADOV
design process and the four step IDOV design process is:
a. The two additional design steps in the DMADOV process

b. IDOV uses validate and DMADOV uses verify
c. The IDOV process is quicker
d. IDOV includes measure and analyze in the design step
Answers: 11.13 a, 11.14 b, 11.18 d

© QUALITY COUNCIL OF INDIANA XII-1 (1048)
CSSBB 2014
XII. APPENDIX
INDEX LEARNING TURNS NO

STUDENT PALE, YET HOLDS THE
EEL OF SCIENCE BY THE TAIL.
ALEXANDER POPE
CSSBB 2014
XII. APPENDIX
Table I - Standard Normal Table
0 Z
Z X.X0 X.X1 X.X2 X.X3 X.X4 X.X5 X.X6 X.X7 X.X8 X.X9
0.0 0.5000 0.4960 0.4920 0.4880 0.4840 0.4801 0.4761 0.4721 0.4681 0.4641
0.1 0.4602 0.4562 0.4522 0.4483 0.4443 0.4404 0.4364 0.4325 0.4286 0.4247
0.2 0.4207 0.4168 0.4129 0.4090 0.4052 0.4013 0.3974 0.3936 0.3897 0.3859
0.3 0.3821 0.3783 0.3745 0.3707 0.3669 0.3632 0.3594 0.3557 0.3520 0.3483
0.4 0.3446 0.3409 0.3372 0.3336 0.3300 0.3264 0.3228 0.3192 0.3156 0.3121
0.5 0.3085 0.3050 0.3015 0.2981 0.2946 0.2912 0.2877 0.2843 0.2810 0.2776
0.6 0.2743 0.2709 0.2676 0.2643 0.2611 0.2578 0.2546 0.2514 0.2483 0.2451
0.7 0.2420 0.2389 0.2358 0.2327 0.2297 0.2266 0.2236 0.2206 0.2177 0.2148
0.8 0.2119 0.2090 0.2061 0.2033 0.2005 0.1977 0.1949 0.1922 0.1894 0.1867
0.9 0.1841 0.1814 0.1788 0.1762 0.1736 0.1711 0.1685 0.1660 0.1635 0.1611
1.0 0.1587 0.1562 0.1539 0.1515 0.1492 0.1469 0.1446 0.1423 0.1401 0.1379
1.1 0.1357 0.1335 0.1314 0.1292 0.1271 0.1251 0.1230 0.1210 0.1190 0.1170
1.2 0.1151 0.1131 0.1112 0.1093 0.1075 0.1056 0.1038 0.1020 0.1003 0.0985
1.3 0.0968 0.0951 0.0934 0.0918 0.0901 0.0885 0.0869 0.0853 0.0838 0.0823
1.4 0.0808 0.0793 0.0778 0.0764 0.0749 0.0735 0.0721 0.0708 0.0694 0.0681
1.5 0.0668 0.0655 0.0643 0.0630 0.0618 0.0606 0.0594 0.0582 0.0571 0.0559
1.6 0.0548 0.0537 0.0526 0.0516 0.0505 0.0495 0.0485 0.0475 0.0465 0.0455
1.7 0.0446 0.0436 0.0427 0.0418 0.0409 0.0401 0.0392 0.0384 0.0375 0.0367
1.8 0.0359 0.0351 0.0344 0.0336 0.0329 0.0322 0.0314 0.0307 0.0301 0.0294
1.9 0.0287 0.0281 0.0274 0.0268 0.0262 0.0256 0.0250 0.0244 0.0239 0.0233
2.0 0.0228 0.0222 0.0217 0.0212 0.0207 0.0202 0.0197 0.0192 0.0188 0.0183
2.1 0.0179 0.0174 0.0170 0.0166 0.0162 0.0158 0.0154 0.0150 0.0146 0.0143
2.2 0.0139 0.0136 0.0132 0.0129 0.0125 0.0122 0.0119 0.0116 0.0113 0.0110
2.3 0.0107 0.0104 0.0102 0.0099 0.0096 0.0094 0.0091 0.0089 0.0087 0.0084
2.4 0.0082 0.0080 0.0078 0.0075 0.0073 0.0071 0.0069 0.0068 0.0066 0.0064
2.5 0.0062 0.0060 0.0059 0.0057 0.0055 0.0054 0.0052 0.0051 0.0049 0.0048
2.6 0.0047 0.0045 0.0044 0.0043 0.0041 0.0040 0.0039 0.0038 0.0037 0.0036
2.7 0.0035 0.0034 0.0033 0.0032 0.0031 0.0030 0.0029 0.0028 0.0027 0.0026
2.8 0.0026 0.0025 0.0024 0.0023 0.0023 0.0022 0.0021 0.0021 0.0020 0.0019
2.9 0.0019 0.0018 0.0018 0.0017 0.0016 0.0016 0.0015 0.0015 0.0014 0.0014
3.0 0.00135
CSSBB 2014
XII. APPENDIX
Table II - Six Sigma Failure Rates

With a 1.5 σ Process Shift With No Process Shift
Z ppm Z ppm Z ppm Z ppm
1.0 697,672.15 3.6 17,864.53 1.0 317,310.52 3.6 318.29
1.1 660,082.92 3.7 13,903.50 1.1 271,332.20 3.7 215.66
1.2 621,378.38 3.8 10,724.14 1.2 230,139.46 3.8 144.74
1.3 581,814.88 3.9 8,197.56 1.3 193,601.10 3.9 96.23
1.4 541,693.78 4.0 6,209.70 1.4 161,513.42 4.0 63.37
1.5 501,349.97 4.1 4,661.23 1.5 133,614.46 4.1 41.34
1.6 461,139.78 4.2 3,467.03 1.6 109,598.58 4.2 26.71
1.7 421,427.51 4.3 2,555.19 1.7 89,130.86 4.3 17.09
1.8 382,572.13 4.4 1,865.88 1.8 71,860.53 4.4 10.83
1.9 344,915.28 4.5 1,349.97 1.9 57,432.99 4.5 6.80
2.0 308,770.21 4.6 967.67 2.0 45,500.12 4.6 4.23
2.1 274,412.21 4.7 687.20 2.1 35,728.71 4.7 2.60
2.2 242,071.41 4.8 483.48 2.2 27,806.80 4.8 1.59
2.3 211,927.71 4.9 336.98 2.3 21,448.16 4.9 0.960
2.4 184,108.21 5.0 232.67 2.4 16,395.06 5.0 0.574
2.5 158,686.95 5.1 159.15 2.5 12,419.36 5.1 0.340
2.6 135,686.77 5.2 107.83 2.6 9,322.44 5.2 0.200
2.7 115,083.09 5.3 72.37 2.7 6,934.05 5.3 0.116
2.8 96,809.10 5.4 48.12 2.8 5,110.38 5.4 0.067
2.9 80,762.13 5.5 31.69 2.9 3,731.76 5.5 0.038
3.0 66,810.63 5.6 20.67 3.0 2,699.93 5.6 0.021
3.1 54,801.40 5.7 13.35 3.1 1,935.34 5.7 0.012
3.2 44,566.73 5.8 8.55 3.2 1,374.40 5.8 0.007
3.3 35,931.06 5.9 5.42 3.3 966.97 5.9 0.004
3.4 28,716.97 6.0 3.40 3.4 673.96 6.0 0.002
3.5 22,750.35 6.1 2.11 3.5 465.35 6.1 0.001
CSSBB 2014
XII. APPENDIX
Table III - Poisson Distribution

Probability of r or fewer occurrences of an event that has an average
number of occurrences equal to np.
r
0 1 2 3 4 5 6 7 8 9
np
0.02 0.980 1.000

0.04 0.961 0.999 1.000
0.06 0.942 0.998 1.000
0.08 0.923 0.997 1.000
0.10 0.905 0.995 1.000
0.15 0.861 0.990 0.999 1.000

0.20 0.819 0.982 0.999 1.000
0.25 0.779 0.974 0.998 1.000
0.30 0.741 0.963 0.996 1.000
0.35 0.705 0.951 0.994 1.000

0.40 0.670 0.938 0.992 0.999 1.000
0.45 0.638 0.925 0.989 0.999 1.000
0.50 0.607 0.910 0.986 0.998 1.000
0.55 0.577 0.894 0.982 0.998 1.000

0.60 0.549 0.878 0.977 0.997 1.000
0.65 0.522 0.861 0.972 0.996 0.999 1.000
0.70 0.497 0.844 0.966 0.994 0.999 1.000
0.75 0.472 0.827 0.959 0.993 0.999 1.000
0.80 0.449 0.809 0.953 0.991 0.999 1.000

0.85 0.427 0.791 0.945 0.989 0.998 1.000
0.90 0.407 0.772 0.937 0.987 0.998 1.000
0.95 0.387 0.754 0.929 0.984 0.997 1.000
1.00 0.368 0.736 0.920 0.981 0.996 0.999 1.000
1.1 0.333 0.699 0.900 0.974 0.995 0.999 1.000

1.2 0.301 0.663 0.879 0.966 0.992 0.998 1.000
1.3 0.273 0.627 0.857 0.957 0.989 0.998 1.000
1.4 0.247 0.592 0.833 0.946 0.986 0.997 0.999 1.000
1.5 0.223 0.558 0.809 0.934 0.981 0.996 0.999 1.000
1.6 0.202 0.525 0.783 0.921 0.976 0.994 0.999 1.000

1.7 0.183 0.493 0.757 0.907 0.970 0.992 0.998 1.000
1.8 0.165 0.463 0.731 0.891 0.964 0.990 0.997 0.999 1.000
1.9 0.150 0.434 0.704 0.875 0.956 0.987 0.997 0.999 1.000
2.0 0.135 0.406 0.677 0.857 0.947 0.983 0.995 0.999 1.000
CSSBB 2014
XII. APPENDIX
Table III - Poisson Distribution (Cont.)

r
0 1 2 3 4 5 6 7 8 9
np
2.2 0.111 0.355 0.623 0.819 0.928 0.975 0.993 0.998 1.000
2.4 0.091 0.308 0.570 0.779 0.904 0.964 0.988 0.997 0.999 1.000
2.6 0.074 0.267 0.518 0.736 0.877 0.951 0.983 0.995 0.999 1.000
2.8 0.061 0.231 0.469 0.692 0.848 0.935 0.976 0.992 0.998 0.999
3.0 0.050 0.199 0.423 0.647 0.815 0.916 0.966 0.988 0.996 0.999
3.2 0.041 0.171 0.380 0.603 0.781 0.895 0.955 0.983 0.994 0.998
3.4 0.033 0.147 0.340 0.558 0.744 0.871 0.942 0.977 0.992 0.997
3.6 0.027 0.126 0.303 0.515 0.706 0.844 0.927 0.969 0.988 0.996
3.8 0.022 0.107 0.269 0.473 0.668 0.816 0.909 0.960 0.984 0.994
4.0 0.018 0.092 0.238 0.433 0.629 0.785 0.889 0.949 0.979 0.992
4.2 0.015 0.078 0.210 0.395 0.590 0.753 0.867 0.936 0.972 0.989
4.4 0.012 0.066 0.185 0.359 0.551 0.720 0.844 0.921 0.964 0.985
4.6 0.010 0.056 0.163 0.326 0.513 0.686 0.818 0.905 0.955 0.980
4.8 0.008 0.048 0.143 0.294 0.476 0.651 0.791 0.887 0.944 0.975
5.0 0.007 0.040 0.125 0.265 0.440 0.616 0.762 0.867 0.932 0.968
5.2 0.006 0.034 0.109 0.238 0.406 0.581 0.732 0.845 0.918 0.960
5.4 0.005 0.029 0.095 0.213 0.373 0.546 0.702 0.822 0.903 0.951
5.6 0.004 0.024 0.082 0.191 0.342 0.512 0.670 0.797 0.886 0.941
5.8 0.003 0.021 0.072 0.170 0.313 0.478 0.638 0.771 0.867 0.929
6.0 0.002 0.017 0.062 0.151 0.285 0.446 0.606 0.744 0.847 0.916
10 11 12 13 14 15 16
2.8 1.000
3.0 1.000
3.2 1.000
3.4 0.999 1.000
3.6 0.999 1.000
3.8 0.998 0.999 1.000
4.0 0.997 0.999 1.000
4.2 0.996 0.999 1.000

4.4 0.994 0.998 0.999 1.000
4.6 0.992 0.997 0.999 1.000
4.8 0.990 0.996 0.999 1.000
5.0 0.986 0.995 0.998 0.999 1.000
5.2 0.982 0.993 0.997 0.999 1.000

5.4 0.977 0.990 0.996 0.999 1.000
5.6 0.972 0.988 0.995 0.998 0.999 1.000
5.8 0.965 0.984 0.993 0.997 0.999 1.000
6.0 0.957 0.980 0.991 0.996 0.999 0.999 1.000
CSSBB 2014
XII. APPENDIX
Table III - Poisson Distribution (Cont.)

r
0 1 2 3 4 5 6 7 8 9
np
6.2 0.002 0.015 0.054 0.134 0.259 0.414 0.574 0.716 0.826 0.902
6.4 0.002 0.012 0.046 0.119 0.235 0.384 0.542 0.687 0.803 0.886
6.6 0.001 0.010 0.040 0.105 0.213 0.355 0.511 0.658 0.780 0.869
6.8 0.001 0.009 0.034 0.093 0.192 0.327 0.480 0.628 0.755 0.850
7.0 0.001 0.007 0.030 0.082 0.173 0.301 0.450 0.599 0.729 0.830
7.2 0.001 0.006 0.025 0.072 0.156 0.276 0.420 0.569 0.703 0.810
7.4 0.001 0.005 0.022 0.063 0.140 0.253 0.392 0.539 0.676 0.788
7.6 0.001 0.004 0.019 0.055 0.125 0.231 0.365 0.510 0.648 0.765
7.8 0.000 0.004 0.016 0.048 0.112 0.210 0.338 0.481 0.620 0.741
8.0 0.000 0.003 0.014 0.042 0.100 0.191 0.313 0.453 0.593 0.717
8.5 0.000 0.002 0.009 0.030 0.074 0.150 0.256 0.386 0.523 0.653
9.0 0.000 0.001 0.006 0.021 0.055 0.116 0.207 0.324 0.456 0.587
9.5 0.000 0.001 0.004 0.015 0.040 0.089 0.165 0.269 0.393 0.522
10.0 0.000 0.000 0.003 0.010 0.029 0.067 0.130 0.220 0.333 0.458
10 11 12 13 14 15 16 17 18 19
6.2 0.949 0.975 0.989 0.995 0.998 0.999 1.000
6.4 0.939 0.969 0.986 0.994 0.997 0.999 1.000
6.6 0.927 0.963 0.982 0.992 0.997 0.999 0.999 1.000
6.8 0.915 0.955 0.978 0.990 0.996 0.998 0.999 1.000
7.0 0.901 0.947 0.973 0.987 0.994 0.998 0.999 1.000
7.2 0.887 0.937 0.967 0.984 0.993 0.997 0.999 0.999 1.000
7.4 0.871 0.926 0.961 0.980 0.991 0.996 0.998 0.999 1.000
7.6 0.854 0.915 0.954 0.976 0.989 0.995 0.998 0.999 1.000
7.8 0.835 0.902 0.945 0.971 0.986 0.993 0.997 0.999 1.000
8.0 0.816 0.888 0.936 0.966 0.983 0.992 0.996 0.998 0.999 1.000
8.5 0.763 0.849 0.909 0.949 0.973 0.986 0.993 0.997 0.999 0.999
9.0 0.706 0.803 0.876 0.926 0.959 0.978 0.989 0.995 0.998 0.999
9.5 0.645 0.752 0.836 0.898 0.940 0.967 0.982 0.991 0.996 0.998
10.0 0.583 0.697 0.792 0.864 0.917 0.951 0.973 0.986 0.993 0.997
20 21 22
8.5 1.000
9.0 1.000
9.5 0.999 1.000
10.0 0.998 0.999 1.000
CSSBB 2014
XII. APPENDIX
Table IV - Binomial Distribution

Probability of r or fewer occurrences of an event in n trials
p (the probability of occurrence on each trial)
n r 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50
2 0 0.9025 0.8100 0.7225 0.6400 0.5625 0.4900 0.4225 0.3600 0.3025 0.2500
1 0.9975 0.9900 0.9775 0.9600 0.9375 0.9100 0.8775 0.8400 0.7975 0.7500
3 0 0.8574 0.7290 0.6141 0.5120 0.4219 0.3430 0.2746 0.2160 0.1664 0.1250
1 0.9928 0.9720 0.9392 0.8960 0.8438 0.7840 0.7182 0.6480 0.5748 0.5000
2 0.9999 0.9990 0.9966 0.9920 0.9844 0.9730 0.9571 0.9360 0.9089 0.8750
4 0 0.8145 0.6561 0.5220 0.4096 0.3164 0.2401 0.1785 0.1296 0.0915 0.0625
1 0.9860 0.9477 0.8905 0.8192 0.7383 0.6517 0.5630 0.4752 0.3910 0.3125
2 0.9995 0.9963 0.9880 0.9728 0.9492 0.9163 0.8735 0.8208 0.7585 0.6875
3 1.0000 0.9999 0.9995 0.9984 0.9961 0.9919 0.9850 0.9744 0.9590 0.9375
5 0 0.7738 0.5905 0.4437 0.3277 0.2373 0.1681 0.1160 0.0778 0.0503 0.0312
1 0.9774 0.9185 0.8352 0.7373 0.6328 0.5282 0.4284 0.3370 0.2562 0.1875
2 0.9988 0.9914 0.9734 0.9421 0.8965 0.8369 0.7648 0.6826 0.5931 0.5000
3 1.0000 0.9995 0.9978 0.9933 0.9844 0.9692 0.9460 0.9130 0.8688 0.8125
4 1.0000 1.0000 0.9999 0.9997 0.9990 0.9976 0.9947 0.9898 0.9815 0.9688
6 0 0.7351 0.5314 0.3771 0.2621 0.1780 0.1176 0.0754 0.0467 0.0277 0.0156
1 0.9672 0.8857 0.7765 0.6554 0.5339 0.4202 0.3191 0.2333 0.1636 0.1094
2 0.9978 0.9842 0.9527 0.9011 0.8306 0.7443 0.6471 0.5443 0.4415 0.3438
3 0.9999 0.9987 0.9941 0.9830 0.9624 0.9295 0.8826 0.8208 0.7447 0.6562
4 1.0000 0.9999 0.9996 0.9984 0.9954 0.9891 0.9777 0.9590 0.9308 0.8906
5 1.0000 1.0000 1.0000 0.9999 0.9998 0.9993 0.9982 0.9959 0.9917 0.9844
7 0 0.6983 0.4783 0.3206 0.2097 0.1335 0.0824 0.0490 0.0280 0.0152 0.0078
1 0.9556 0.8503 0.7166 0.5767 0.4449 0.3294 0.2338 0.1586 0.1024 0.0625
2 0.9962 0.9743 0.9262 0.8520 0.7564 0.6471 0.5323 0.4199 0.3164 0.2266
3 0.9998 0.9973 0.9879 0.9667 0.9294 0.8740 0.8002 0.7102 0.6083 0.5000
4 1.0000 0.9998 0.9988 0.9953 0.9871 0.9712 0.9444 0.9037 0.8471 0.7734
5 1.0000 1.0000 0.9999 0.9996 0.9987 0.9962 0.9910 0.9812 0.9643 0.9375
6 1.0000 1.0000 1.0000 1.0000 0.9999 0.9998 0.9994 0.9984 0.9963 0.9922
8 0 0.6634 0.4305 0.2725 0.1678 0.1001 0.0576 0.0319 0.0168 0.0084 0.0039
1 0.9428 0.8131 0.6572 0.5033 0.3671 0.2553 0.1691 0.1064 0.0632 0.0352
2 0.9942 0.9619 0.8948 0.7969 0.6785 0.5518 0.4278 0.3154 0.2201 0.1445
3 0.9996 0.9950 0.9786 0.9437 0.8862 0.8059 0.7064 0.5941 0.4770 0.3633
4 1.0000 0.9996 0.9971 0.9896 0.9727 0.9420 0.8939 0.8263 0.7396 0.6367
5 1.0000 1.0000 0.9998 0.9988 0.9958 0.9887 0.9747 0.9502 0.9115 0.8555
6 1.0000 1.0000 1.0000 0.9999 0.9996 0.9987 0.9964 0.9915 0.9819 0.9648
7 1.0000 1.0000 1.0000 1.0000 1.0000 0.9999 0.9998 0.9993 0.9983 0.9961
9 0 0.6302 0.3874 0.2316 0.1342 0.0751 0.0404 0.0207 0.0101 0.0046 0.0020
1 0.9288 0.7748 0.5995 0.4362 0.3003 0.1960 0.1211 0.0705 0.0385 0.0195
2 0.9916 0.9470 0.8591 0.7382 0.6007 0.4628 0.3373 0.2318 0.1495 0.0898
3 0.9994 0.9917 0.9661 0.9144 0.8343 0.7297 0.6089 0.4826 0.3614 0.2539
4 1.0000 0.9991 0.9944 0.9804 0.9511 0.9012 0.8283 0.7334 0.6214 0.5000
5 1.0000 0.9999 0.9994 0.9969 0.9900 0.9747 0.9464 0.9006 0.8342 0.7461
6 1.0000 1.0000 1.0000 0.9997 0.9987 0.9957 0.9888 0.9750 0.9502 0.9102
7 1.0000 1.0000 1.0000 1.0000 0.9999 0.9996 0.9986 0.9962 0.9909 0.9805
8 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.9999 0.9997 0.9992 0.9980
10 0 0.5987 0.3487 0.1969 0.1074 0.0563 0.0282 0.0135 0.0060 0.0025 0.0010
1 0.9139 0.7361 0.5443 0.3758 0.2440 0.1493 0.0860 0.0464 0.0232 0.0107
2 0.9885 0.9298 0.8202 0.6778 0.5256 0.3828 0.2616 0.1673 0.0996 0.0547
3 0.9990 0.9872 0.9500 0.8791 0.7759 0.6496 0.5138 0.3823 0.2660 0.1719
4 0.9999 0.9984 0.9901 0.9672 0.9219 0.8497 0.7515 0.6331 0.5044 0.3770
5 1.0000 0.9999 0.9986 0.9936 0.9803 0.9527 0.9051 0.8338 0.7384 0.6230
6 1.0000 1.0000 0.9999 0.9991 0.9965 0.9894 0.9740 0.9452 0.8980 0.8281
7 1.0000 1.0000 1.0000 0.9999 0.9996 0.9984 0.9952 0.9877 0.9726 0.9453
8 1.0000 1.0000 1.0000 1.0000 1.0000 0.9999 0.9995 0.9983 0.9955 0.9893
9 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.9999 0.9997 0.9990
CSSBB 2014
XII. APPENDIX
Table V - t Distribution
tα
d.f. t.100 t.050* t.025** t.010 t.005 d.f.

1 3.078 6.314 12.706 31.821 63.657 1
2 1.886 2.920 4.303 6.965 9.925 2
3 1.638 2.353 3.182 4.541 5.841 3
4 1.533 2.132 2.776 3.747 4.604 4
5 1.476 2.015 2.571 3.365 4.032 5
6 1.440 1.943 2.447 3.143 3.707 6
7 1.415 1.895 2.365 2.998 3.499 7
8 1.397 1.860 2.306 2.896 3.355 8
9 1.383 1.833 2.262 2.821 3.250 9
10 1.372 1.812 2.228 2.764 3.169 10
11 1.363 1.796 2.201 2.718 3.106 11
12 1.356 1.782 2.179 2.681 3.055 12
13 1.350 1.771 2.160 2.650 3.012 13
14 1.345 1.761 2.145 2.624 2.977 14
15 1.341 1.753 2.131 2.602 2.947 15
16 1.337 1.746 2.120 2.583 2.921 16
17 1.333 1.740 2.110 2.567 2.898 17
18 1.330 1.734 2.101 2.552 2.878 18
19 1.328 1.729 2.093 2.539 2.861 19
20 1.325 1.725 2.086 2.528 2.845 20
21 1.323 1.721 2.080 2.518 2.831 21
22 1.321 1.717 2.074 2.508 2.819 22
23 1.319 1.714 2.069 2.500 2.807 23
24 1.318 1.711 2.064 2.492 2.797 24
25 1.316 1.708 2.060 2.485 2.787 25
26 1.315 1.706 2.056 2.479 2.779 26
27 1.314 1.703 2.052 2.473 2.771 27
28 1.313 1.701 2.048 2.467 2.763 28
29 1.311 1.699 2.045 2.462 2.756 29
inf. 1.282 1.645 1.960 2.326 2.576 inf.
* one tail 5% α risk ** two tail 5% α risk

CSSBB 2014
XII. APPENDIX
Table VI - Critical
Chi-Square (O2) Values f(F )
X2
0 .95 X2 0.0 5
"
F "
DF O20.99 O20.95 O20.90 O20.10 O20.05 O20.01
1 0.00016 0.0039 0.0158 2.71 3.84 6.63

2 0.0201 0.1026 0.2107 4.61 5.99 9.21
3 0.115 0.352 0.584 6.25 7.81 11.34
4 0.297 0.711 1.064 7.78 9.49 13.28
5 0.554 1.15 1.61 9.24 11.07 15.09
6 0.872 1.64 2.20 10.64 12.59 16.81
7 1.24 2.17 2.83 12.02 14.07 18.48
8 1.65 2.73 3.49 13.36 15.51 20.09
9 2.09 3.33 4.17 14.68 16.92 21.67
10 2.56 3.94 4.87 15.99 18.31 23.21
11 3.05 4.57 5.58 17.28 19.68 24.73
12 3.57 5.23 6.30 18.55 21.03 26.22
13 4.11 5.89 7.04 19.81 22.36 27.69
14 4.66 6.57 7.79 21.06 23.68 29.14
15 5.23 7.26 8.55 22.31 25.00 30.58
16 5.81 7.96 9.31 23.54 26.30 32.00
18 7.01 9.39 10.86 25.99 28.87 34.81
20 8.26 10.85 12.44 28.41 31.41 37.57
24 10.86 13.85 15.66 33.20 36.42 42.98
30 14.95 18.49 20.60 40.26 43.77 50.89
40 22.16 26.51 29.05 51.81 55.76 63.69
60 37.48 43.19 46.46 74.40 79.08 88.38
120 86.92 95.70 100.62 140.23 146.57 158.95
Table VII - Distribution of F

CSSBB 2014
XII. APPENDIX
F Table α = 0.05
f(F )
"
ν1(DF)
F "
ν2(DF) 1 2 3 4 5 6 7 8 9 10 12 15
1 161.4 199.5 215.7 224.6 230.2 234.0 236.8 238.9 240.5 241.9 243.9 245.9
2 18.51 19.00 19.16 19.25 19.30 19.33 19.35 19.37 19.38 19.40 19.41 19.43
3 10.13 9.55 9.28 9.12 9.01 8.94 8.89 8.85 8.81 8.79 8.74 8.70
4 7.71 6.94 6.59 6.39 6.26 6.16 6.09 6.04 6.00 5.96 5.91 5.86
5 6.61 5.79 5.41 5.19 5.05 4.95 4.88 4.82 4.77 4.74 4.68 4.62
6 5.99 5.14 4.76 4.53 4.39 4.28 4.21 4.15 4.10 4.06 4.00 3.94
7 5.59 4.74 4.35 4.12 3.97 3.87 3.79 3.73 3.68 3.64 3.57 3.51
8 5.32 4.46 4.07 3.84 3.69 3.58 3.50 3.44 3.39 3.35 3.28 3.22
9 5.12 4.26 3.86 3.63 3.48 3.37 3.29 3.23 3.18 3.14 3.07 3.01
10 4.96 4.10 3.71 3.48 3.33 3.22 3.14 3.07 3.02 2.98 2.91 2.85
11 4.84 3.98 3.59 3.36 3.20 3.09 3.01 2.95 2.90 2.85 2.79 2.72
12 4.75 3.89 3.49 3.26 3.11 3.00 2.91 2.85 2.80 2.75 2.69 2.62
13 4.67 3.81 3.41 3.18 3.03 2.92 2.83 2.77 2.71 2.67 2.60 2.53
14 4.60 3.74 3.34 3.11 2.96 2.85 2.76 2.70 2.65 2.60 2.53 2.46
15 4.54 3.68 3.29 3.06 2.90 2.79 2.71 2.64 2.59 2.54 2.48 2.40
ν1(DF)
ν2(DF) 20 30 40 50 60 4
20 2.12 2.04 1.99 1.96 1.95 1.84
30 1.93 1.84 1.79 1.76 1.74 1.62
40 1.84 1.74 1.69 1.66 1.64 1.51
50 1.78 1.69 1.63 1.60 1.58 1.44
60 1.75 1.65 1.59 1.56 1.53 1.39
4 1.57 1.46 1.39 1.35 1.32 1.00

CSSBB 2014
XII. APPENDIX
Table VIII - Distribution of F

F Table α = 0.025
ν1(DF)
ν2(DF) 1 2 3 4 5 6 7 8 9 10 12 15
1 647.8 799.5 864.2 899.6 921.8 937.1 948.2 956.7 963.3 968.6 976.7 984.9
2 38.51 39.00 39.17 39.25 39.30 39.33 39.36 39.37 39.39 39.40 39.41 39.43
3 17.44 16.04 15.44 15.10 14.88 14.73 14.62 14.54 14.47 14.42 14.34 14.25
4 12.22 10.65 9.98 9.60 9.36 9.20 9.07 8.98 8.90 8.84 8.75 8.66
5 10.01 8.43 7.76 7.39 7.15 6.98 6.85 6.76 6.68 6.62 6.52 6.43
6 8.81 7.26 6.60 6.23 5.99 5.82 5.70 5.60 5.52 5.46 5.37 5.27
7 8.07 6.54 5.89 5.52 5.29 5.12 4.99 4.90 4.82 4.76 4.67 4.57
8 7.57 6.06 5.42 5.05 4.82 4.65 4.53 4.43 4.36 4.30 4.20 4.10
9 7.21 5.71 5.08 4.72 4.48 4.32 4.20 4.10 4.03 3.96 3.87 3.77
10 6.94 5.46 4.83 4.47 4.24 4.07 3.95 3.85 3.78 3.72 3.62 3.52
11 6.72 5.26 4.63 4.28 4.04 3.88 3.76 3.66 3.59 3.53 3.43 3.33
12 6.55 5.10 4.47 4.12 3.89 3.73 3.61 3.51 3.44 3.37 3.28 3.18
13 6.41 4.97 4.35 4.00 3.77 3.60 3.48 3.39 3.31 3.25 3.15 3.05
14 6.30 4.86 4.24 3.89 3.66 3.50 3.38 3.29 3.21 3.15 3.05 2.95
15 6.20 4.77 4.15 3.80 3.58 3.41 3.29 3.20 3.12 3.06 2.96 2.86
ν1(DF)
ν2(DF) 20 30 40 50 60 4
20 2.46 2.35 2.29 2.25 2.22 2.09
30 2.20 2.07 2.01 1.97 1.94 1.79
40 2.07 1.94 1.88 1.83 1.80 1.64
50 1.99 1.87 1.80 1.76 1.72 1.55
60 1.94 1.82 1.74 1.70 1.67 1.48
4 1.71 1.57 1.48 1.43 1.39 1.00

CSSBB 2014
XII. APPENDIX
Table IX - Control Chart Factors

CHART FOR CHART FOR STANDARD
CHART FOR RANGES
AVERAGES DEVIATIONS
Sample Center Center

Control limit Control Limit Control Limit
Observations Line Line
Factors Factors Factors
Factors Factors
n A2 A3 c4 B3 B4 d2 D3 D4
2 1.880 2.659 0.7979 0 3.267 1.128 0 3.267
3 1.023 1.954 0.8862 0 2.568 1.693 0 2.574
4 0.729 1.628 0.9213 0 2.266 2.059 0 2.282
5 0.577 1.427 0.9400 0 2.089 2.326 0 2.114
6 0.483 1.287 0.9515 0.030 1.970 2.534 0 2.004
7 0.419 1.182 0.9594 0.118 1.882 2.704 0.076 1.924
8 0.373 1.099 0.9650 0.185 1.815 2.847 0.136 1.864
9 0.337 1.032 0.9693 0.239 1.761 2.970 0.184 1.816
10 0.308 0.975 0.9727 0.284 1.716 3.078 0.223 1.777
15 0.223 0.789 0.9823 0.428 1.572 3.472 0.347 1.653
20 0.180 0.680 0.9869 0.510 1.490 3.735 0.415 1.585
25 0.153 0.606 0.9896 0.565 1.435 3.931 0.459 1.541
X  R Charts X  S Charts
CL x  x  A 2 R CL x  x  A 3 S
UCLR  D4 R UCLR  B4 S
LCLR  D3 R LCLR  B3 S
Approximate capability Approximate capability
 R
  S

d2 c4

05 CSSBB Primer

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

05 CSSBB Primer

Uploaded by

Copyright:

Available Formats

© QUALITY COUNCIL OF INDIANA INTRO-1 (1)

© by Quality Council of Indiana - All rights reserved

Fourth Edition - September, 2014

Quality Council of Indiana

CSSBB Primer Contents

II. ENTERPRISE-WIDE DEPLOYMENT . . . . . . . . . . . II-1

III. PROCESS MANAGEMENT . . . . . . . . . . . . . . . . . III-1

IV. TEAM MANAGEMENT . . . . . . . . . . . . . . . . . . . . . IV-1

VI. MEASURE - DATA . . . . . . . . . . . . . . . . . . . . . . . . VI-1

VII. MEASURE - STATISTICS . . . . . . . . . . . . . . . . . VII-1

VIII. ANALYZE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII-1

IX. IMPROVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX-1

XI. DESIGN FOR SIX SIGMA (DFSS) . . . . . . . . . . . . XI-1

XII. APPENDIX/TABLES . . . . . . . . . . . . . . . . . . . . . XII-1

CSSBB Primer Question Contents

II. Organization-wide 8.0% 12 32 80

III. Process Management 6.7% 10 27 67

IV. Team Management 12.0% 18 48 120

V. Define 13.3% 20 53 133

VI. Measure - Data 7.3% 11 29 73

VII. Measure - Statistics 9.3% 14 37 93

VIII. Analyze 14.7% 22 59 147

IX. Improve 14.0% 21 56 140

X. Control 10.0% 15 40 100

XI. Design for Six Sigma 4.7% 7 19 47

Totals 100% 150 400 1000

Professionalizing Quality Education

To provide recognized six sigma fundamental training

Is a professional who can carry out, in a responsible

CSSBB participants must register with ASQ

A written test lasts 4 hours. Prometric exams last 4.5

Additional information can be obtained by calling ASQ

The CSSBB student should obtain the bibliography

Breyfogle, III, F.W. (2003). Implementing Six Sigma:

Harry, Mikel & Schroeder, R. (2000). Six Sigma: The

Pande, P.S., Neuman, R.P., & Cavanagh, R.R. (2000).

ASQ Certified Six Sigma Black Belt

I. Organization-wide Deployment [12 Q]

II. Process Management and Measures [10 Q]

III. Team Management [18 Q]

IV. Define [20 Q]

VI. Analyze [22 Q]

VII. Improve [21 Q]

VIII. Control [15 Q]

IX. Design for Six Sigma (DFSS) [7 Q]

II. ENTERPRISE-WIDE DEPLOYMENT I.A.1

SIX SIGMA HAS FOREVER CHANGED

II. ENTERPRISE-WIDE DEPLOYMENT I.A.1

Organization-wide considerations are presented in the

C Six sigma and lean fundamentals

II. ENTERPRISE-WIDE DEPLOYMENT I.A.1

Value of Six Sigma

Motorola®, under the direction of Chairman Bob Galvin,

II. ENTERPRISE-WIDE DEPLOYMENT I.A.1

Value of Six Sigma (Continued)

The ± 1.5 Sigma shift

II. ENTERPRISE-WIDE DEPLOYMENT I.A.1

Value of Six Sigma (Continued)

Note that Table II in the Appendix provides defect levels

It should be noted that the term “six sigma” has been

II. ENTERPRISE-WIDE DEPLOYMENT I.A.1

Value of Six Sigma (Continued)

Define: Select the appropriate responses (the “Ys”)

Measure: Data must be gathered to measure the