18-19.

Quality Management

By Prof. N. Narayanan
In
“OPERATIONS MANAGEMENT – I”
(WMP, Term I, Part B, July – Sep, 2012)

Text Book Reference:

Ch 5: Quality Management (pp 149 – 179),
Sup 5: Statistical Process Control (pp 181 – 211)
In: J. Heizer, B. Render and J. Rajashekhar, Operations
Management, 9th Edition, Pearson Education, 2009
 Sessions Plan – Part B (Continued)

Sn. Text
Nos. Date Topic
Reading
Core Agenda for Session
18 – Quality Ch 5: Quality 1. Case: TQM improves copier service (pp 164)
19 Management Management – Describes role played by TQM in improving
(pp 149 – service quality
179); 2. Case: Richey International’s Spies (pp 171)
3. Problem: Determining ‘Process capability
index’ at a shoe company (pp 196)
Sup 5:
Statistical 4. Problem: Control Limits for weights of oats
Process flakes boxes (pp 186 – 7)
Control (pp 5. Problem: range limits at Clinton
181 – 211) manufacturing (pp 188)
6. Case: Unisys Corp.’s costly experiment in
health care services (pp 192)

• Case: Hank Kolb – Director (Quality

Assurance) (Handout)
• Case: Singapore General Hospital: Delivering
Service Quality (Handout)
• Case: Turnaround at the Preston Plant
(Handout)
Quiz – II (Covering session topics from 16 to 19)
 Overview
Chapter 6:
1. What is Quality? (pp 152 – 156)
2. Total Quality Management (TQM) as philosophy
(pp 157 – 63)

3. Tools of TQM (pp 163 – 67)

4. The Role of Inspection (pp 167 – 169)
5. TQM in Services (pp 169 – 171)
Supplement 6:
6. Process Capability (195 – 197)
7. Statistical Process Control - Introduction
(pp 182 – 85)

8. Control Charts for Variables (185 – 89)

9. Control Charts for Attributes (190 – 193)

10. Acceptance Sampling (197 – 99)

 Part A

Basics in
Quality Management
 Global Company Profile:
Managing Quality Provides a Competitive Advantage
Arnold Palmer Hospital (p 150 – 51)
 Delivered over 10,500 babies in 2004, and
over 13,000 in 2006.
 Virtually every type of quality tool is employed
 Continuous improvement

 Employee empowerment

 Benchmarking

 Just-in-time

 Quality tools
1. What is Quality? (pp 152 – 56)
 Different Views on ‘What is Quality?’

1. User-based – better Quality definition

performance, more features must start with
user, but must
2. Product-based – specific and also be taken
measurable attributes of the backward to the
product earlier stages, to
help solving
3. Manufacturing-based – problems at root
conformance to standards, cause levels.
making it right the first time
 Key Dimensions of Quality

User Product Manufacturer

-based -based -based
 Performance  Durability  Conformance
 Features  Aesthetics  Making it right
 Reliability  Perceived the first time
(process
quality
 Serviceability quality)
 Value

Need is to approach Quality from the Customer/User-backwards

 Defining Quality

The totality of features and characteristics of a

product or service, which bears on its ability to
satisfy stated or implied needs of the customers.

– American Society for Quality

 Takumi – The Japanese view of
the ‘how?’ of quality
A Japanese character that
symbolizes:
 a broader dimension than
quality,
 a more perfect method
than persistence, and
 a deeper process than
education.
This view leads, in turn, to a new ‘precedence sequence-wise
priority’ statement:

People before Quality before Productivity

 Costs of Quality – In four different areas:

1. Prevention costs - reducing These can be purely

the potential for defects one-time costs!

2. Appraisal costs - evaluating

products, parts, and services These occur as
recurring
costs, but are
3. Internal failure - producing capable of being
defective parts or service reduced, as a
result of one-
before delivery time investments
in the
4. External costs - defects ‘Prevention
Costs’.
discovered after delivery
 Two Views on ‘Cost of Quality’ (Not in the book)
The traditional View The TQM View
Costs Costs
Prevention
and Total
appraisal costs
Total costs
costs

Failure
costs Prevention and
Failure appraisal costs
costs

100% Targeted 0% Targeted

100% 0%
Defectives levels of Defectives levels of
Defectives Defectives
quality quality

• In the TQM view, the quality costs are to be not only optimized
through right trade-offs in the short term, but also brought down
to near zero levels in the long term.
 Costs of Quality

Total Total Cost

Cost
External Failure

Internal Failure

Prevention

Appraisal

Quality Improvement
 Caselet:
The Very High Costs of Quality at Mercedes (pp 154)
1. The first efforts at revival of the Company’s performance
resulted in the diversion of attention from quality, leading, by
2003, to severe loss in quality (particularly in the suppliers’),
and hence in high quality costs:
• Series of recalls from its $50,000 E-Class sedan.
• Spate of problems in 2004 with brake control systems—680,000 cars
were recalled.
• The biggest recall in history in 2005—of 1.3 million cars with faulty
fuel pumps made by supplier, Robert Bosch.
• Many other defects arising from software problems and failed
interfaces in the complex electronic systems.
• A total of $600 million in warranty costs in a single year!
2. The costs of quality also resulted in a toll in sales, and loss of
the number one position, as luxury car maker, to BMW!
 Quality and Strategy

 Managing quality supports

differentiation, low cost, and response
strategies
 Quality helps firms increase sales and
reduce costs
 Building a quality organization is a
demanding task
2. “Total Quality Management
(TQM)”, etc.,
as Quality Philosophies
(pp 157 – 162)
 TQM

• Encompasses entire organization, from

supplier to customer.

• Stresses a commitment by management

to have a continuing, companywide drive
toward excellence, in all aspects of
products and services that are important
to the customer.
 Continuous Improvement

1. Represents continual improvement of

all processes
2. Involves all operations and work
centers including suppliers and
customers
 People, Equipment, Materials, Procedures
 Just-in-Time (JIT)

Relationship to quality:

1. JIT cuts the cost of quality

2. JIT improves quality—through
faster feedback.
3. Better quality means less
inventory and better, easier-to-
employ JIT system
Shewhart’s PDCA Model

1.Plan:
4. Act:
Identify the
Implement the improvement and
plan (revised make
as and when a plan.
necessary).

3. Check: 2. Do:
Is the plan Test the
working? plan.
 Taguchi Concepts
 Three Taguchi Concepts
1. Quality loss function The purpose is to promote
quality by minimizing variances
from the target values.
2. Target-oriented quality
The purpose is to promote quality
3. Quality robustness through ‘Parameter design’ by
minimizing the effects of variances
from the target values.

 Engineering and experimental design methods

to improve product and process design
 Identify key component and process variables
affecting product variations.
 Quality Loss Function
L = D 2C
where
High loss
L= loss to
Unacceptable society, which includes
Loss the firm, customers
(to producing Poor and other members of
organization, society (i.e., all the
customer, Fair losses incurred from
and society) loss in quality)
Good D= distance
from target value
Low loss Best
C= cost factor
related to deviation
Target-oriented quality
yields more product in the
“best” category
Target-oriented quality
Frequency brings product toward the
target value
Specification limits-oriented
quality (which Taguchi
termed ‘Goal Post mentality’)
Lower Target Upper keeps products within 3
standard deviations.
Specification
 ‘Quality Loss Function’ and
‘Target-oriented Quality’
1. Shows that costs increase as the
product moves away from what the
customer wants1.
Target-
2. Costs include customer oriented
dissatisfaction, warranty quality
and service, internal
scrap and repair, and costs to society.
3. Traditional conformance specifications
are too simplistic.
1
Any tolerances specified around target value are merely out of consideration
for manufacturing’s available capabilities, not customers’ needs.
 Quality Robustness (refers to ‘design quality’)
 Ability to produce products uniformly, even
under adverse manufacturing and environmental
conditions, based on the following principles:
 Small variations in materials and processes need not
destroy product quality, if we can remove the effects1 of
the adverse conditions.
 Through right ‘Parameter Design’ for these materials and
processes, it is possible to minimize the effects, if not
the cause, of the adverse conditions.
 This can lead to less need for emphasizing ‘Tolerance
Design’, and hence have the ability to get quality even
under adverse conditions at the downstream/ customer
end.
1
The additional adverse effects arising from the interactions between the
variances in the different parameters of the product/ process design.
 Six Sigma
• ‘Six Sigma’ is essentially aimed at ‘target-oriented
quality’.
• It reflects the goal of having the ‘natural variation
(±3σ)’ of the process outputs to be half the
customer specification limits of the items
produced by a process (which, by definition will
then be ±6σ).
• With such a process, the probability of producing
a part outside the specification limits is about 2
parts per billion.
• However, Motorola’s definition of ‘Six Sigma’ is a
limit of defects of 3.4 per million, as from a
process with specification limits at ±6σ, but with a
bias of 1.5σ.
 Six Sigma
 Two broad meanings:
1. Statistical definition of a process that is
99.9997% capable, 3.4 defects per
million opportunities (DPMO)
Note: This definition is related to ‘conformance
to specification limits’, even though our ideal
is, “Target-oriented quality”

2. A motivation-oriented program designed

to reduce defects, lower costs, and
improve customer satisfaction
 A 3σ Process
– is a process with its ±3σ spread within the specification limits.

A 3σ process A 3σ process, with its

leads to a defect of natural variation, ±3σ lying
2,700 defects/million within specification limits.
(when the process centre
has zero bias with respect to
the target value) Upper
Lower
specification specification
limit limit

Mean (Target
Value)
±3

The natural variation of ±3 lies within the

specification limits (when there is zero bias).
 Motorola’ definition of ‘Six Sigma Process’
A Six Sigma process, is one
with the variation of ±3σ lying
within specification limits, (even)
when having a bias of 1.5σ.
A ‘Six Sigma’ Or, other processes with
process as defined different bias and σ, but
above can yielding 3.4 or less defects/
lead to a defect of million
3.4 defects/million 1.5
or less. Lower σ Upper
specification specification
limit limit

Mean Mean
(Target
Value)

±3

Note: In a ‘six sigma process’, its ±6σ spread need not be within the
specification limits.
 Process Capability & Defects
Process Capability Total Products outside two-sided
Index (Cpk) specification limits
0.25 453,255 ppm

0.50 133,614 ppm

0.60 71,861 ppm

0.80 16,395 ppm

1.00 2,700 ppm

1.20 318 ppm

1.50 7 ppm

1.70 0.34 ppm

2.00 0.0018 ppm

Source: Quality Planning & Analysis, Juran & Gryna, Chapter 17, 3e
A ‘Six Sigma Process’, as compared to a ‘3σ Process’
A Six Sigma process, with the
variation of ±6σ lying within
A 3σ process specification limits, but with a
bias of 1.5σ
leads to a defect of
A 3σ process, with its
2,700 defects/million natural variation, ±3σ lying
(when the process centre within specification limits.
has zero bias with respect to
1.5
the target value) Upper
Lower σ
specification specification
limit limit
A ‘Six Sigma’
process can
lead to a defect of Mean Mean (Target
3.4 defects/million Value)
(as when a 6σ process ±3
centre has a bias of 1.5σ
with respect to the
target value) ±6

From a statistical viewpoint, the ‘Six Sigma’s defect level of 3.4 defects per
million’ represents a process that can have a ‘bias’ of up to 1.5σ, while
its spread of ±6σ still remains within the specification limits.
(Not in the book)
 Six Sigma
– a structured approach to process improvement

1. Define critical outputs

and identify gaps for DMAIC Approach
improvement
2. Measure the work and
collect process data
3. Analyze the data
4. Improve the process
5. Control the new process to
make sure new performance
is maintained
3. Seven Tools of “TQM”

(Detailed coverage of this topic

is provided at the end.)
 Seven Tools of TQM
1. Tools for Generating Ideas
1. Check sheets
2. Scatter diagrams
3. Cause-and-effect diagrams
2. Tools to Organize the Data
4. Pareto charts
5. Flowcharts
3. Tools for Identifying Problems
6. Histogram
7. Statistical process control chart
4. The Role of Inspection
 Inspection
1. Many problems
1. Worker fatigue
2. Measurement error
3. Process variability
2. Cannot inspect quality into a product. It is
more a ‘last resort’, after the defects are
produced1.
3. Robust design, empowered employees, and
sound processes—as ‘Quality at Source’
(based on ‘self-inspection)—are better
solutions
1
In “Poka yoke”, ‘source inspection’ (inspection at/of the process from
which the product is produced) assures defect free process, and hence
defect free product. This kind of inspection is ‘value-adding’
 Inspection (Continued)
4. Involves examining items to see if an
item is good or defective
5. Detect a defective product
1. Does not correct deficiencies in process or
product
2. It is expensive
6. Issues
1. When to inspect
2. Where in process to inspect
 Inspection (Continued)
7. Self-inspection (and hence ‘self-
assessment’) can be far more accurate
and insightful than an external inspection
that is done by another person.
 When and Where to Inspect
1. At the supplier’s plant while the supplier is
producing
2. At your facility upon receipt of goods from the
supplier
3. Before costly or irreversible processes
4. During the step-by-step production process
5. When production or service is complete
6. Before delivery to your customer
7. At the point of customer contact
 Case:

Richey International’s Spies

(pp 171)
(raises issues on the role of ‘Inspection’)
 Case: “Richey International’s Spies” (p 171)

The Case Situation:

1. Hotels maintain their one-on-one service quality (largely
dependent on personal behavior) through inspection.
2. They hire Richey International to do quality evaluations via
spying.
3. Before such inspection, management must have established
(as standards) what the customer expects, and specific
services that yield customer satisfaction.
4. Aggressive training and objective inspections reinforce true
service behaviors on the part of the hotel employees.
5. More than 50 different standards are evaluated, before the
inspectors even check in at a luxury hotel. Then there are
further checks over the next 24 hours on a large number of
standards (using a variety of ‘QC Tools’)
6. Quality happens as a result of the established standards,
aggressive training and objective inspections.
 Case: “Richey International’s Spies” (p 171)

The Possible Problems to Consider:

1. Are Richey International, as the spying service provider, do this
service as specialists in hotel industry, or do they provide this service
for other very different types of industries and customers as well?
2. If they are specialists for the hotel industry, how their relevant
experience can be best taken advantage of, in the assessment of the
quality process, and how the process may be further improved?
3. If they are not specialists for the hotel industry, how do they adapt
their approach to spy with respect to the requirements of the hotel
industry?
4. The employees would obviously know of the spies coming in as
customers. Are they able to get some clues of when some customer
is actually a spy, and accordingly behave differently? How do
employees actually take to the idea of being watched by spies? Does
this awareness affect their performance for better or for worse?
 Case: “Richey International’s Spies” (p 171)

The Possible Problems to Consider (continued):

5. Can the inspection carried out by “Richey International” be

organized in a different way whereby it is carried out with full
knowledge of the service worker concerned, and the information
from the inspection be provided also to the worker as feedback,
to enable and empower the worker in his efforts at self
improvement? What may be effect of such a different
arrangement of the inspection process?
5. TQM in Services
 TQM In Services

1. Service quality is more difficult to

measure than the quality of goods
2. Service quality perceptions depend on
1. Intangible differences between
products
2. Intangible expectations customers have
of those products
 Determinants of Service Quality

1. Reliability 7. Credibility
2. Responsiveness 8. Security
3. Competence 9. Understanding/
4. Access knowing the
customer, all in
5. Courtesy addition to the
6. Communication 10.Tangibles
 Service Recovery Strategy

 Managers should have a plan for when

services fail
 Marriott’s LEARN routine
1. Listen
2. Empathize
3. Apologize
4. React
5. Notify
 Case:

TQM Improves Copier Service

(pp 164)
– Describes role played by TQM
in raising service quality
Case: Ricoh Corp – TQM Improves Copier Service (p 204)
The Case Situation:
• Distinction in products offered in the copier industry blurred
by the sameness of technology used by all the players.
• At Ricoh Corp., Japan, competitive advantage is sought
through superior service to the customers.
• Two of the ways the service quality were improved, while
also reducing cost:
– Found, using “tools of TQM”, that significant time on service calls
were wasted when engineers had to go back to their trucks for
spare parts.
 The firm assembled a “call kit”, which allowed engineers to carry onto customer
premises all parts with high probability of use. Service calls became faster, and
costed less.
– The “callback” problem (as a result of errors in the job by the service
engineers) was tackled using the Pareto principle.
 Retraining confined to only 11% of the customer engineers with the most
callbacks, led to a 19% drop in return visits.
Case: Ricoh Corp – TQM Improves Copier Service (p 204)
Possible Problems to Consider:
• More ways to enhance service by identifying newer areas of
possible problems in the future.
• Design of call kits:
– Need for having different call kits, for different kinds of equipment,
which are to be provided with service.
– Determining for best way to standardize call kits applicable to a range
of products being serviced
– Way to modify standardization in call kits, as conditions change over
time because of aging of equipment, changes in customer usage
patterns, etc.
• Consider the practice of ‘retraining only the 11% customer
engineers with most call backs’ in order to reduce the
revisits.
– Should not amount to ‘rewarding’ only the laggards
– What ways are possible to reward the high performers, while the
laggards are being retrained.
 Case – 4:

The Culture of Quality at

Arnold Palmer Hospital (APH) (pp 177)
 The Culture of Quality at
Arnold Palmer Hospital (APH) (pp 171)
1. Broad range of medical services including neonatal and
pediatric intensive care, pediatric oncology and cardiology,
care for high-risk pregnancies, and maternal intensive care
2. Have developed comprehensive and scientific means of
asking customers to judge quality of care they receive.
3. Employees empowered to provide gifts in value up to $200 to
patients who find reason to complain about any hospital
service such as food, courtesy, responsiveness or
cleanliness.
4. Quality measured not only through customer service but also
through a variety of internal measures.
5. The result of all these efforts has been a quality culture, as
manifested in APH’s high ranking in patient satisfaction, and
one of the highest survival rates of critically ill babies.
 Case: “Arnold Palmer Hospital” (p 217)
The Case Situation:
1. Provides a broad range of medical services including
– Neonatal and pediatric intensive care, pediatric oncology and cardiology,
care for high-risk pregnancies, and maternal intensive care.
2. Have developed comprehensive and scientific means of asking
customers to judge the quality of care they receive.
– Customer survey questionnaire mailed one week after patient discharge
– Executive Director, Swanson, follows the results daily.
3. A variety of internal measures, starting at grassroots level,
where the staff sees a problem and develops ways to track
performance
– Morbidity,
– Infection rates,
– Readmission rates,
– Costs per case,
– Length of stays.
– Swanson uses these measures to monitor for continuous improvement.
 Case: “Arnold Palmer Hospital” (p 217)
The Case Situation (Continued):
4. Tools used daily by Swanson:
– Pareto charts
– Flow- and process charts
– Benchmarking against hospitals both nationally, and in the southeastern
region
5. Employees are empowered to provide gifts in value up to $ 200/-
to patients who find reason to complain about any hospital
service such as food, courtesy, responsiveness or cleanliness.
6. The deeply underlying philosophy and concept at the hospital is
that
– Each patient is important, respected as a person, and is entitled to
comprehensive, compassionate, family-centered health-care provided by a
knowledgeable physician-directed team.
7. Result: a quality culture manifesting in
– High ranking in patient satisfaction
– One of the highest survival rates of critically ill babies.
 Case: “Arnold Palmer Hospital” (p 217)
Discussion Questions:
1. Does the patient have the expertise to judge the health care she receives?
Yes, as far as the basic care is concerned. She can judge the general attitude. She
can be right in believing that if the right attitude is there, the rest will follow.
2. How would you build a culture of quality?
See pp192 – Global company profile
3. What techniques does APH practice in its drive for quality and continuous
improvement?
•The hospital’s longstanding philosophy of care for the patients.
•Customer surveys, a variety of internal measures to track performance.
4. Fishbone diagram:

Material Machinery Attitude of care for patient

Quality
Health
care
Methods Manpower Environment symbolizing
warmth and friendliness
Note: This fish bone diagram depicts ‘good’ effects arising from ‘good’ causes.
 Quality Inspection at APH
What is
Organization Standard
Inspected
Arnold Billing Accurate, timely, and
Palmer correct format
Hospital
Pharmacy Prescription accuracy,
inventory accuracy
Lab Audit for lab-test accuracy
Nurses Charts immediately
updated
Admissions Data entered correctly and
completely
 Case – 5:

Quality at the Ritz Carlton Hotel

Company (pp 178)
Quality at the Ritz Carlton Hotel Company (pp 178)
1. Treats quality as if it is the heart beat of the company—means a
daily commitment to meeting customer expectations and making
sure that each hotel is freee of any deficiency.
2. The need and the challenge for every employee is to create the
right combinations of elements (ranging from ‘maintenance’ to
‘management’) that make the ‘experience’.
3. Company efforts (in the process of pursuit of Baldrige award) at
rigorous self examination through measuring and quantifying
quality of 19 selected processes (ranging from room service
delivery times and reservations to valet parking and
housekeeping efficiency).
4. The next focus was on ‘continuous improvement’.
5. Organization of ‘self-directed’ work teams.
6. To support the above, employees provided with opportunities to
take additional training in hotel operations.
 Case: Ritz-Carlton (p 178)
The Case Situation:
1. Quality in hotel industry has to be about the customer’s
‘experience’.
– This experience must happen as a result of every employee, from
maintenance to management, working to create the right
combination of elements.
2. The first hotel company to win the Malcolm Baldrige
National Quality Award.
– Rigorous self-examination of operations, based on 19 processes,
such as:
 Room service delivery,
 Guest reservation and registration,
 Message delivery, and
 Breakfast service
to measure and quantify quality, before applying for the Baldrige
Award.
 Case: Ritz-Carlton (p 217) – Continued
The Case Situation (Continued):

– The study included statistical measurement of process work

flows and cycle times for areas such as:
 Room service delivery times,
 Reservations,
 Valet parking,
 Housekeeping efficiency
– Results used to develop performance benchmarks, against
which future activity could be measured.
3. With specific quantifiable targets in place, the
employees now focus on
– Continuous improvement
– The goal is 100% customer satisfaction.
 Case: Ritz-Carlton (p 217) – Continued

The Case Situation (Continued):

4. Employees organized into “self-directed” work teams.

Employee teams determine
– Work scheduling,
– What work needs to be done, and
– What to do about quality problems in their own work areas.
5. Employees given opportunity to take additional training in
hotel operations,
– To enable them see the relationship of their specific area to the
overall goals
– This is based on the company’s belief that a more educated
and informed employee is in a better position to make decisions
in the best interest of the company
 Case: Ritz-Carlton (p 217)
Discussion Questions:
1. In what ways could the Ritz-Carlton monitor its success in achieving
quality?
Nineteen processes important for quality (ranging from room service delivery
times to valet parking and housekeeping efficiency) were studied using
statistical measurements of process workflows and cycle times — These
measures usually start at the grassroots level (where the staff sees a problem)
and develops ways to track performance.
2. What actions might you expect from a company that intends quality to
be more than a slogan or buzzword?
Measure and reward quality in action
3. Why might it cost the Ritz-Carlton less to “do things right” the first
time?
4. How could control charts, Pareto diagrams, and cause-and-effect
diagram be used to identify quality problems at a hotel?
5. What are some of the non-financial measures of customer satisfaction
that might be used by the Ritz-Carlton?
 Quality Inspection at Hard Rock Hotel

Standard

Hard Rock Hotel Reception Use customer’s name

desk

Doorman Greet guest in less than

30 seconds
Room All lights working, spotless
bathroom
Minibar Restocked and charges
accurately posted to bill
Quality Inspection at Olive Garden Restaurant

What is
Organization Standard
Inspected
Olive Garden Busboy Serves water and
Restaurant bread within 1 minute
Busboy Clears all entrée items
and crumbs prior to
dessert
Waiter Knows and suggest
specials, desserts
 Part B

Statistical Quality Control

 Caselets
– Statistical Quality Control
a. Fritto-Lay Uses SPC to Keep Its Ruffles
Tasty (p 190)
b. Unisys Corporation’s Costly Experiment in
Health Care Services (p 192)
c. Bayfield Mud Company (p 208)
d. Alabama Airlines’s On-Time Schedule (p 209)

e. Farm to Fork: Quality at Darden (p 210)

6. Process Capability
 Process Capability

Two measures of ‘Process Capability’:

1. Process Capability Ratio (Cp)

2. Process Capability Index (Cpk )

 Process Capability
1. The natural variation of a process should
be small enough to produce products
that meet the standards required.
2. Process capability is a measure of the
relationship between the natural variation
of the process and the design
specifications.
Note: The measure is in relation to specification
limits, even though we talked of ‘target-
oriented quality’.
 Predictive capability for processes
Which process is better?

Process B

Process A

Spread of a
LSL process is
USL
indicative of
its capability

LCLa LCLb Centers of the two UCLb UCLa

processes
 Predictive capability of processes
Which process is better?
Process A

LCLa UCLa
Process B
A process that is
aligned closer to
the desired
target is likely to
LCL UCL be more capable
b
Offset of the process center
from the target value

LSL USL
Thus, any measure of ‘process capability’ must be based on both the
‘process spread’, and any ‘offset’ that may be present in the process center.
 Process Capability Ratio
Upper Specification Limit – Lower Specification Limit
Cp = 6

1. A capable process must have a Cp of at least 1.0

2. Does not look at how well the process is centered

in the specification range
3. Often a target value of Cp = 1.33 is used to allow
for off-center processes
4. Six Sigma quality requires a Cp = 1.5 when the
process centre has no bias with respect to target
value, and requires a Cp = 2.0 when the process
has a bias of 1.5σ. (Not as in the book)
Process Capability Ratio – An Example (pp 195)
Insurance claims process

Process mean x = 210.0 minutes

Process standard deviation  = .516
minutes
Design specification = 210 ± 3
minutes

Upper Specification – Lower Specification

Cp =
6
 Process Capability Ratio
GE’s Insurance claims process (pp 195)
Process mean x = 210.0 minutes
Process standard deviation  = .516 minutes
Design specification = 210 ± 3 minutes

Upper Specification - Lower Specification

Cp =
6

213 – 207
= = 1.938
6(.516)
 Process Capability Ratio – An Example
(Continued)

GE’s Insurance claims process (pp 195)

Process mean x = 210.0 minutes
Process standard deviation  = .516 minutes
Design specification = 210 ± 3 minutes

Upper Specification - Lower Specification

Cp =
6
213 - 207
= = 1.938
6(.516)

Process is capable (because cp » 1.33)

 Process Capability Ratio, Index
• Process Capability is defined by both the ‘spread of the
process’ and any ‘offset’ of the process mean from the
target value.
• Process Capability Ratio (Cp) is defined as the ratio of
the difference in specification limits to the process spread
Cp = Specification Range (USL  LSL)

Pr ocess Capability 6
• Process Capability Index (Cpk) takes into consideration
the extent to which the process has deviated from the
desired target
  Pr ocess Centre  LSL  USL  Pr ocess Centre  
Cpk = Min    ,  
 3   3 
• In general, Cpk ≤ Cp
If, for a process, Cp = Cpk , then the process is exactly centered
at its target value, with no ‘offset’ or bias in the process mean.
 Process Capability Index1
 (tolerance )
bias of the b
data mean
Target values Range of
rejected units

LSL USL
3 3

Upper Lower
Cpk = minimum of Specification – x , x – Specification
Limit Limit
 
1. A capable process must have a Cpk of at least 1.0
2. A ‘capable process’ is not necessarily in the center
of the specification2, but it falls within the
specification limit at both extremes.
1
‘Process Capability Index’ as a measure also takes into account (unlike ‘Process Capability’)
any bias in the process mean (i.e., process being not in the centre of the specifications).
2
A process with Cpk value above 1.0 will be ‘capable’, in spite of the bias in the process.
 Process Capability Index
of an ‘Insole cutting machine’ (pp 196)

New Cutting Machine

New process mean x = .250 inches

Process standard deviation  = .0005 inches
Upper Specification Limit = .251 inches
Lower Specification Limit = .249 inches
 Process Capability Index

New Cutting Machine

New process mean x = .250 inches

Process standard deviation  = .0005 inches
Upper Specification Limit = .251 inches
Lower Specification Limit = .249 inches

(.251) – .250
Cpk = minimum of ,
(3).0005
 Process Capability Index

New Cutting Machine

New process mean x = .250 inches
Process standard deviation  = .0005 inches
Upper Specification Limit = .251 inches
Lower Specification Limit = .249 inches

Cpk = minimum of (.251) - .250 , .250 - (.249)

(3).0005 (3).0005

Both calculations result in

.001
New machine is NOT capable,
Cpk =
.0015
= 0.67 because Cpk < 1.0
 Interpreting Cpk

Cpk = negative number

This process
Cpk = zero has bias, but
in spite of it,
it is just
‘capable’
because of
Cpk = between 0 and 1 low
standard
deviation, σ.
or
Cpk = 1

Cpk > 1
LSL USL
 Process Capability & Defects
Process Capability Total Products outside two-sided
Index (Cpk) specification limits
0.25 453,255 ppm

0.50 133,614 ppm

0.60 71,861 ppm

0.80 16,395 ppm

1.00 2,700 ppm

1.20 318 ppm

1.50 7 ppm

1.70 0.34 ppm

2.00 0.0018 ppm

Source: Quality Planning & Analysis, Juran & Gryna, Chapter 17, 3e
 A 3σ Process
– is a process with its ±3σ spread within the specification limits.

A 3σ process A 3σ process, with its

leads to a defect of natural variation, ±3σ lying
2,700 defects/million within specification limits.
(when the process centre
has zero bias with respect to
the target value) Upper
Lower
specification specification
limit limit

Mean (Target
Value)
±3

The natural variation of ±3 lies within the

specification limits (when there is zero bias).
 Motorola’ definition of ‘Six Sigma Process’
A Six Sigma process, is one
with the variation of ±3σ lying
within specification limits, (even)
when having a bias of 1.5σ.
A ‘Six Sigma’ Or, other processes with
process as defined different bias and σ, but
above can yielding 3.4 or less defects/
lead to a defect of million
3.4 defects/million 1.5
or less. Lower σ Upper
specification specification
limit limit

Mean Mean
(Target
Value)

±3

Note: In a ‘six sigma process’, its ±6σ spread need not be within the
specification limits.
A ‘Six Sigma Process’, as compared to a ‘3σ Process’
A Six Sigma process, with the
variation of ±6σ lying within
A 3σ process specification limits, but with a
bias of 1.5σ
leads to a defect of
A 3σ process, with its
2,700 defects/million natural variation, ±3σ lying
(when the process centre within specification limits.
has zero bias with respect to
1.5
the target value) Upper
Lower σ
specification specification
limit limit
A ‘Six Sigma’
process can
lead to a defect of Mean Mean (Target
3.4 defects/million Value)
(as when a 6σ process ±3
centre has a bias of 1.5σ
with respect to the
target value) ±6

From a statistical viewpoint, the ‘Six Sigma’s defect level of 3.4 defects per
million’ represents a process that can have a ‘bias’ of up to 1.5σ, while
its spread of ±6σ still remains within the specification limits.
(Not in the book)
Six Sigma Quality – Process Requirements
(Not in the book)

1. When the process is centered:

Then, minimum Cp = Cpk = 1.5, and
σ ≤ (USL – LSL)/9………….How?
2. When the process is biased, with bias =
1.5σ:
Then, minimum Cp = 2, and minimum Cpk = 1.5,
and
σ ≤ (USL – LSL)/12…………How?
3. Or, all the positions between the above two
extremes.
7. Statistical Process Control (SPC),
Using ‘Control Charts’

1. Control Charts for Variables

2. The Central Limit Theorem
3. Setting Mean Chart Limits (x-Charts)
4. Setting Range Chart Limits (R-Charts)
5. Using Mean and Range Charts
6. Control Charts for Attributes
7. Managerial Issues and Control Charts
 Statistical Process Control (SPC)
1. Variability is inherent
in every process
1. Natural or common
causes
2. Special or assignable causes
2. Provides a statistical signal when
assignable causes are present
3. Detect and eliminate assignable causes
of variation
Statistical process control only measures the performance of
a process; it does not help to identify a particular specimen
produced as being “good” or “bad,” in or out of tolerance.
 Natural Cause Variations
1. Also called common cause variations (usually
numerous and small)
2. Affect virtually all production processes
3. Expected amount of variation
4. Output measures follow a probability
distribution
5. For any distribution there is a measure of
central tendency and dispersion
6. If the distribution of outputs falls within
acceptable limits, the process is said to be “in
statistical control”
 Assignable Cause Variations

1. Also called special causes of variation

1. Generally this arises from some change in
the process (usually few, but large)
2. Variations that can be traced to a specific
reason
3. The objective is to discover when
assignable causes are present
1. Eliminate the bad causes.
2. Incorporate the good causes.
 Control Charts

• Constructed from historical data, the purpose of

control charts is to help distinguish between natural
variations and variations due to assignable causes.
• The ‘control limits’ would need to be much closer
than the ‘specification limits’, in order for the
‘capability’ of a process to approach the ‘Six Sigma’.
Control Chart – A generalized representation
(Not in the book)
Upper Control Limit (UCL) = Process Average + zσ
Plot of sample
data

Process Average

Lower Control Limit (LCL) = Process Average – zσ

where , z = number of normal standard deviations

= 2 for 95.45 % confidence
= 3 for 99.73 % confidence

The UCL and the UCL reflect the capability of the process
in attaining the ‘target value’
When a Process is “Out of Statistical Control”
• A process is said to be “out of statistical control” (with
respect to a particular ‘control chart’ drawn for the
process), when it has some ‘assignable cause/s’ of variation
in the process, making the sample statistic from the
process, plotted on the control chart, give a signal of such a
possible presence.
• This is not the same as the process being ‘out of control’,
since,
• Even if a process is ‘capable’ in adhering to the specification
limits, it can still be ‘out of statistical control’ with respect to
the control chart in use’, and
• A process in ‘statistical control’ (as indicated by ‘control
charts’ in use) may not necessarily meet the design
specifications, and hence may be ‘not capable’.
 Taking Advantage of
‘Central Limit Theorem’
In Developing Control Charts
• Regardless of the distribution of the population, the
distribution of sample means, drawn from the
population will tend to follow a normal curve
• In case of ‘Control Chart for Variables’, the sample
statistic, ‘mean’ can be taken to follow normal
distribution with:
1. The mean of the sampling distribution x=
(x) will be the same as the population
mean 
2. The standard deviation of the
sampling distribution (x) will equal 
the population standard deviation x =
n
() divided by the square root of
the sample size, n
 Sampling Distribution
Sampling
distribution of
the sample
means

Process
distribution of
the population

x=
(mean)

The ‘sample distribution’ will show lesser variance than of the

population. Why?
Steps In Creating and Using Control Charts

1. Take samples from the population and compute

the appropriate sample statistic
2. Use the sample statistic to calculate control limits,
and draw the control chart
3. Plot sample results on the control chart, and
determine the state of the process (‘in’ or ‘out of
control’)
4. Investigate possible assignable causes, and take
any indicated actions
5. Continue sampling from the process, and reset
the control limits when necessary
 Steps in setting up a ‘process control system’
1. Choose the characteristic
for process control

2. Choose the 3. Choose an appropriate

Measurement method Sampling procedure

4. Choose the type of

Control Chart 5. Calculate
control limits

6. Plot the data & Analyze

 Inspection of
Attributes Versus Variables
 Attributes
 Items are either good or bad, acceptable or
unacceptable
 Does not address degree of failure

 Variables
 Measures dimensions such as weight, speed,
height, or strength
 Falls within an acceptable range

 Use different statistical techniques

 Types of Data

Variables Attributes
 Characteristics that  Defect-related
can take any real characteristics
value
 Classify products as
 May be in whole or either good or bad,
in fractional or count defects
numbers
 Categorical or
 Continuous random discrete random
variables variables
 Measurement Methods
• Variable Based
– Detailed observation of the characteristic (such as
length, diameter, weight)
– measurement will be expensive and more time
consuming but will provide a wealth of information about
the process
• Attribute Based
– simple clustering of the characteristic into a few
categories (such as good or bad)
– Two frequently used attribute measures are:
• Proportion of defects (denoted as p)
• Number of defects (denoted as c)
– measurements are easy to make, quick & less
expensive, but reveal very little information about the
process
8. Control Charts for Variables
 Control Charts for Variables

1. For variables that have

continuous dimensions
• Weight, speed, length,
strength, etc.
2. x-charts are to control
the central tendency of the process
3. R-charts are to control the dispersion of
the process
4. These two charts must be used together,
when monitoring variables.
 Setting Chart Limits
For x-Charts when we know population standard deviation)

Upper control limit (UCL) = x + zx

Lower control limit (LCL) = x – zx

where x = mean of the sample means or

a target value set for the process
z = number of normal standard
deviations
x = standard deviation of the
sample means
= / n
 = population standard deviation
 Setting ‘Mean Limits’ fro Variables,
using samples (Example S1 in page 186 -87)
Weights of boxes of Oat Flakes, sampled each hour:
Hour 1: Sample size n = 9 The 12 hourly samples:
Sample Weight of Hour Mean Hour Mean
Number Oat Flakes 1 16.1 7 15.2
1 17 2 16.8 8 16.4
2 13 3 15.5 9 16.3
3 16 4 16.5 10 14.8
4 18 5 16.5 11 14.2
n=9 5 17 6 16.4 12 17.3
6 16
7 15 For 99.73% control limits, z = 3
8 17
9 16 UCLx = x + zx = 16 + 3(1/3) = 17 ozs
Mean 16.1
 = 1 LCLx = x – zx = 16 – 3(1/3) = 15 ozs
 Setting Control Limits

Control Chart
for sample of 9 Variation due
Out of to assignable
boxes control causes
17 = UCL

16 = Mean Variation due to

natural causes

15 = LCL

| | | | | | | | | | | | Variation due
1 2 3 4 5 6 7 8 9 10 11 12 to assignable
Out of causes
Sample number control
 Setting Chart Limits
For x-Charts when we don’t know 

Upper control limit (UCL) = x + A2R

Lower control limit (LCL) = x – A2R

where R = average range of the samples

A2 = control chart factor found in Table
S6.1 (p 187)
x = mean of the sample means
 Control Chart Factors1 (p 187)
for Computing Control Chart Limits (3 sigma)
Sample Size Mean Factor Upper Range Lower Range
n A2 D4 D3
2 1.880 3.268 0
3 1.023 2.574 0
4 .729 2.282 0
5 .577 2.115 0
6 .483 2.004 0
7 .419 1.924 0.076
8 .373 1.864 0.136
9 .337 1.816 0.184
10 .308 1.777 0.223
12 .266 1.716 0.284

1
American Society for Testing Materials, 1951
Setting Control Limits – An Example S2 (pp 187)

Process average x = 12 ounces

Average range R = .25
Sample size n = 5
 Setting Control Limits

Process average x = 12 ounces

Average range R = .25
Sample size n = 5

UCLx = x + A2R
= 12 + (.577)(.25)
= 12 + .144
= 12.144 ounces
From Table
S6.1
 Setting Control Limits
Process average x = 12 ounces
Average range R = .25
Sample size n = 5

UCLx = x + A 2R UCL = 12.144

= 12 + (.577)(.25)
= 12 + .144 Mean = 12
= 12.144 ounces

LCLx = x – A 2R LCL = 11.857

= 12 – .144
= 11.856 ounces
 R – Chart

1. Type of variables control chart

2. Shows sample ranges over time
• Difference between smallest and
largest values in sample

3. Monitors process variability

4. Independent from process mean
 Setting Chart Limits
For R-Charts

Upper control limit (UCLR) = D4R

Lower control limit (LCLR) = D3R

where
R = average range of the samples
D3 and D4 = control chart factors
from Table S6.1
Setting ‘Range Limits’ for Variables (p188)
Average range R = 5.3 pounds
Sample size n = 5
From Table S6.1 D4 = 2.115, D3 = 0

UCLR = D 4R UCL = 11.2

= (2.115)(5.3)
= 11.2 pounds Mean = 5.3

LCLR = D3R LCL = 0

= (0)(5.3)
= 0 pounds
If the plot of a sample R value falls below the LCLR, the process would still
need to be examined, but for discovery of a ‘good cause’ that has enabled
the process range to fall so much.
 Using Mean and Range Charts

Mean and Range charts complement each other,

by showing the Mean and the dispersion
of the Normal Distribution for the samples.
 Mean and Range Charts Sample
Sample
4
(a) Sample 3
Sample 2
These 1 (Sampling mean is
sampling shifting upward but
distributions range is consistent)
result in the
charts below

UCL
(x-chart detects
x-chart shift in central
tendency)
LCL

UCL
(R-chart does not
R-chart detect change in
mean)
LCL
 Mean and Range Charts
Sample Sample
(b) Sample 3 4
Sample 2
These 1
sampling (Sampling mean is
distributions constant but
result in the dispersion is
charts below increasing)

UCL
(x-chart does not
x-chart detect the increase
in dispersion)
LCL

UCL
(R-chart detects
R-chart increase in
dispersion)
LCL
9. Control Charts for Attributes
 Control Charts for Attributes
1. For variables that are categorical
• Good/bad, yes/no, acceptable/unacceptable
2. Measurement is typically counting defectives
3. Charts may measure
• Percent defective (p-chart): Control charts for
fraction defectives are effective when a piece
considered defective even if there is a single defect.
• Number of defects in each piece (c-chart):
Control charts for defectives are effective when a
large number of defectives can occur in each piece,
but the actual number that do occur is relatively
small.
 Control Limits for p-Charts
Proportion defectives p, of the population (as
representing fraction defectives), will have a
binomial distribution, but applying the Central
Limit Theorem allows us to assume a normal
distribution for the ‘sample statistics’.
p(1 – p)
UCLp = p + z^p ^p =
n
LCLp = p – z^p
where p = mean of the ‘fraction defectives’, based
on all the samples taken
z= number of standard deviations
^ p = estimate of the standard deviation of the
‘fraction defectives’, based on all the samples taken
n = sample size
(Note: The calculation of the standard deviation reflects the binomial distribution of the population.)
Insurance Records at Mosier data Systems (pp 191 – 192)
p-Chart for Data Entry
Sample Number Fraction Sample Number Fraction
Number of Errors Defective Number of Errors Defective
1 6 .06 11 6 .06
2 5 .05 12 1 .01
3 0 .00 13 8 .08
4 1 .01 14 7 .07
5 4 .04 15 5 .05
6 2 .02 16 4 .04
7 5 .05 17 11 .11
8 3 .03 18 3 .03
9 3 .03 19 0 .00
10 2 .02 20 4 .04
Total = 80

80 (.04)(1 - .04)
p = (100)(20) = .04 p^ = = .02
100
 p-Chart for Data Entry
UCLp = p + z^p = .04 + 3(.02) = .10

LCLp = p – z
^ p = .04 – 3(.02) = 01

.11 –
.10 – UCLp = 0.10
.09 –
.08 –
Fraction defective

.07 –
.06 –
.05 –
.04 –
.03 – p = 0.04
.02 –
.01 –
.00 –
| | | | | | | | | | LCLp = 0.00
2 4 6 8 10 12 14 16 18 20
Sample number
1
LCL taken as zero, when the calculation yields negative value.
 p-Chart for Data Entry
UCLp = p + z^p = .04 + 3(.02) = .10
Possible
LCLp = p - z^p = .04 – 3(.02) = 0 causes
assignable
present
.11 –
.10 – UCLp = 0.10
.09 –
.08 –
Fraction defective

.07 –
.06 –
.05 –
.04 –
.03 – p = 0.04
.02 –
.01 –
.00 –
| | | | | | | | | | LCLp = 0.00
2 4 6 8 10 12 14 16 18 20
Sample number
 Control Limits for c-Charts
The number of defects per unit of the
population, c, will have a Poisson distribution 1,
but applying the Central Limit Theorem allows
us to assume a normal distribution for the
sample statistics
UCLc = c + 3 c LCLc = c – 3 c ^
^
^

where c = estimate of the mean number defects per unit,

^ an
c = estimate of the standard deviation of the number
1 of defects
The standard deviation of the per unit distribution
of Poisson = c of the population equals the
square root of the mean
 c-Chart for Cab Company

c = 54 complaints/9 days = 6 complaints/day

14 – UCLc = 13.35

UCLc = c + 3 c 12 –

Number defective
10 –
=6+3 6 8 –
= 13.35 6 –
4 – c= 6
2 –
LCLc = c – 3 c 0 –
LCLc = 0
=6–3 6 | | | | | | | | |
= 01 1 2 3 4 5 6 7 8 9
Day
1
LCL taken as zero when the calculation yields negative value.
Managerial Use of Control Charts
 Which Control Chart to Use
Variables Data

 Using an x-chart and R-chart:

1. Observations are variables
2. Collect 20 – 25 samples of n = 4, or n = 5, or
more, each from a stable process and
compute the mean for the x-chart and range
for the R-chart
3. Track further samples of n observations each
 Which Control Chart to Use
Attribute Data

 Using the p-chart:

1. Observations are attributes that can be
categorized in two states
2. We deal with fraction, proportion, or
percent defectives
3. Have several samples, each with many
observations in each.
 Which Control Chart to Use
Attribute Data
 Using a c-Chart:
1. Observations are attributes whose
defects per unit of output can be
counted.
2. The number counted is a small part of
the possible occurrences.
3. Defects such as number of blemishes
on a desk, number of typos in a page of
text, flaws in a bolt of cloth.
 Patterns in Control Charts
Upper control limit

Target

Lower control limit

Normal behavior. Process is

“in control.”
 Patterns in Control Charts

Upper control limit

Target

Lower control limit

One plot out above (or
below). Investigate for cause.
Process is “out of control.”
 Patterns in Control Charts
Upper control limit

Target

Lower control limit

Trends in either direction, 5
plots. Investigate for cause of
progressive change.
 Patterns in Control Charts
Upper control limit

Target

Lower control limit

Two plots very near lower (or
upper) control. Investigate for
cause.
 Patterns in Control Charts
Upper control limit

Target

Lower control limit

Run of 5 above (or below)

central line. Investigate for
cause.
 Patterns in Control Charts
Upper control limit

Target

Lower control limit

Erratic behavior1. Investigate.

1 Even an erratic behaviour may show a pattern, and if so
may mean a non-random behaviour, calling for
 10. Acceptance Sampling
Strictly speaking, ‘Acceptance Sampling’ is a part of
“Statistical Quality Control”,
but not a part of
“Statistical Process Control”.

SPC – based on view of quality as a producer responsibility

AS – based on view of quality as a user responsibility
Acceptance Sampling

1. Operating Characteristic Curve

2. Average Outgoing Quality
 Acceptance Sampling1
1. Form of quality testing used for incoming
materials or finished goods
1. Take samples at random from a lot
(shipment) of items
2. Inspect each of the items in the sample
3. Decide whether to reject the whole lot based
on the inspection results (based on an
‘acceptable quality level (AQL) and Lot
Tolerance Percent Defective (LTPD)).
2. Only screens lots; does not drive quality
improvement efforts
1
Note: Acceptance sampling relates to the aggregate, not the individual unit.
 Acceptance Sampling1
1. Form of quality testing used for incoming
materials or finished goods
1. Take samples at random from a lot
(shipment) of items
2. Inspect each of the items inlots
Rejected thecan
sample
be:
3. Decide whether to1.reject the whole
Returned to thelot based
on the inspection results (based
supplier, or on an
‘acceptable quality level (AQL) and Lot
Tolerance Percent2.Defective
Culled for defectives
(LTPD)).
(100% inspection)
2. Only screens lots; does not drive quality
improvement efforts
1
Note: Acceptance sampling relates to the aggregate, not the individual unit.
 Operating Characteristic Curve

1. Shows how well a sampling plan

discriminates between good and bad
lots (shipments)
2. Shows the relationship between the
probability of accepting a lot and its
quality level, and similarly, also
3. Shows the relationship between the
probability of rejecting a lot and its
quality level
 The “Perfect” OC Curve
Keep whole shipment
P(Accept Whole Shipment) 100 –

75 –

50 –
Return whole shipment

25 – Cut-Off (based on a specified

‘Acceptance Quality Limit’, AQL)
0 –
| | | | | | | | | | |

0 10 20 30 40 50 60 70 80 90 100
% Defective in Lot

With the ‘Perfect OC curve’,

The probabilities of both rejecting a good lot and
accepting a bad lot (based on a criterion of
‘Acceptance Quality Limit’ or AQL) will be zero.
 AQL and LTPD

 Acceptable Quality Level (AQL)

 Poorest level of quality we are willing to
accept
 Lot Tolerance Percent Defective (LTPD)
 Quality level we consider bad
 Consumer (buyer) does not want to
accept lots with more defects than LTPD
 A Typical OC Curve
100 –  = 0.05 producer’s risk of rejection of good
95 – lot, with Percent defectives below AQL

75 –
As sample size of a
sampling plan increases,
Probability of its OCC will more and
Acceptance more be closer to the
50 –
‘perfect’ OC curve, but
such a plan will become
25 – costlier bec

10 –
 = 0.10 | | | | | | | | | Percent
0 0– 1 2 3 4 5 6 7 8 defective
AQL LTPD
Consumer’s risk
of acceptance of Good Indifference
bad lot with lots zone Bad lots
percent defects
above LTPD
 Producer’s and Consumer’s Risks
(as related to an ‘Operating Characteristics Curve’)
 Producer's risk ()
 Probability of rejecting a good lot (also
known as type I error – "false positive": the error
of rejecting a null hypothesis when it is actually true )

 Probability of rejecting a lot when the

fraction defective is at or above the AQL
 Consumer's risk (β)
 Probability of accepting a bad lot (also
known as type II error – "false negative": the error
of failing to reject a null hypothesis when in fact we should
have rejected it. In )

 Probability of accepting a lot when fraction

defective is below the LTPD
 OC Curves for Different Sampling Plans
Suitable sampling plans can be developed, once
managers specify the desired values for AQL, LTPD,
α, and β

Two possible acceptance sampling plans.

n = 50, c = 1

n = 100, c = 2
 Average Outgoing Quality

1. If a sampling plan replaces all defectives

out of the rejected lots (which will improve
the ‘Average Outgoing Quality’ in terms of
percentage), and

2. If we know the incoming percent

defective for the lot, then

We can compute the average outgoing

quality (AOQ) in percent defective, as
depicted in the next slide.
 Average Outgoing Quality
(Pd)(Pa)(N – n)
AOQ =
Where N
Pd = true percent defective of the lot
Pa = probability of accepting the lot for a given sample
size and quantity defective
N = number of items in the lot
n = number of items in the sample
AOQ varies with Pd—initially rising with increasing Pd,
and later falling with increasing Pd.
The maximum of AOQ (i.e., its worst value)
corresponds to the highest percent defective or the
lowest average quality for the sampling plan, and is
called the Average Outgoing Quality Limit (AOQL).
 Relationships between
SPC (with Control Charts),
Cpk, and Acceptance Sampling
 Acceptance Sampling, SPC and Process Variability
Lower Upper
specification specification
limit limit
(a) Acceptance sampling
(Some bad units
accepted)

(b) Statistical process

control (with ‘Control
Charts’—Keep the
process in control)

(c) Cpk >1 (Design a

process, or improve
it, to be in control)
Process mean, 
Progress in Cpk
over time1
1
With such progress, a stage will be reached when, with low values of p, etc.,
the required large sample sizes will lead towards 100% inspection!. That is,
SPC will no longer be enough. Poka Yoke is based on 100% inspection!
 Case:
Unisys Corp.’s Costly Experiment in Health Care
Services (p 192)
 Case: Unisys Corp. (p 192)
The Case situation:
1. Unisys Corp.’s entry into the ‘computerized health care service
business’:
– To handle the 215,000 Florida state’s employees’ health insurance claims—a
seemingly simple and lucrative growth area for an old-line computer company
like Unisys.
– One year later, the customer abrogated the contract, and also fined Unisys
more than $500,000 for ‘not meeting quality standards’.
2. The firm was out of control on the following two measures of quality:
1. Percent of claims processed with errors: Errors in 8.5% of claims processed, as
against the industry standard of 3.5% defectives.
2. Percent of claims processed within 30 days: In one month’s sample, 13% of the
claims exceeded the 30-day limit, as against the 5% allowed by the State of
Florida
3. CEO of Unisys pulled the plug on future ambitions in health care.
4. State of Florida’s spokesperson said, “We really need somebody
that’s in the insurance business.”
 Case: Unisys Corp. (p 192)
Possible Problems to Consider:
1. Was there a satisfactory monitoring system at Unisys for
tracking the error percent, which happened to be way above the
industry standard?
The error % of 8.5 as against 3.5% standard of the industry over a 3-month
period was found by Coopers and Lybrand, and not by Unisys themselves.
Why?
2. Was there a system to investigate the root causes of such high
error %, and to attend to them at least when they went way
above industry standard?
3. Could the company’s new business have been successful, with
adequate quality control systems at the company? Explain, with
reasons for your answer.
4. What lessons should Unisys draw out of the expansion/
diversification move in the ‘computerized health care service’
business that the company had ventured into?
 Case:

Hank Kolb: Director – Quality

Assurance (Handout)
 Case:
Hank Kolb: Director – Quality Assurance
(Handout)

1. Develop CED for the ‘bad quality’ in the

‘Greasex line.
2. However, an unbalance emphasis on
defects reduction or cost-reduction alone,
can be counterproductive.
3. Develop another CED for possible ‘good
quality’ in the Greasex line, and in the
Company as a whole.
 Case: Hank Kolb, Director: Quality Assurance
The Cause–Effect Diagram as drawn by one participant:
Machines Use of Manpower
* Inappropriate
nonstandard
modifications
machine
Lack of training
*
Inadequate
maintenance Faulty work assignment
No scheduled PM for practices High
the equipment
proportion
Burred Nozzle Heads of Greasex
Inadequate defects.
No use of ‘Control
supplier
response Charts’, etc., to
help detect
Untested problems and root This is, essentially, a ‘bad causes-
can design causes bad effect’ diagram. However, a
Designer not brain-storming, with this diagram,
interest ed in can help draw a ’good causes-good
manufacturing effect’ diagram’ as well, which can
capabilities be help develop improvement plan
Materials Methods with a more positive mindset
The End: Any Questions?
Seven Tools of “TQM”
May be skipped
 Seven Tools of TQM
1. Tools for Generating Ideas
1. Check sheets
2. Scatter diagrams
3. Cause-and-effect diagrams
2. Tools to Organize the Data
4. Pareto charts
5. Flowcharts
3. Tools for Identifying Problems
6. Histogram
7. Statistical process control chart
 Seven Tools of TQM

1) Check Sheet: An organized method of

recording data

Hour
Defect 1 2 3 4 5 6 7 8
A /// / / / / /// /
B // / / / // ///
C / // // ////

The goal is to capture data, at source,

in all its richness (i.e., suitably stratified).
 Seven Tools of TQM

2) Scatter Diagram: A graph of the value

of one variable vs. another variable
Productivity

Absenteeism
 Seven Tools of TQM
3) Cause-and-Effect Diagram: A tool that identifies
process elements (causes) that might effect an
outcome
Causes
Materials Methods

Effect

Manpower Machinery
 Cause-and-Effect Diagrams
Material Method
(ball) (shooting process)
Grain/Feel Aiming point
(grip)
Size of ball
Air pressure Bend knees
Hand position
Balance
Lopsidedness
Follow-through

Missed
free-throws
Training Rim size

Conditioning Motivation Rim height

Consistency Rim alignment Backboard

stability
Concentration

Machine
Manpower (hoop &
(shooter) backboard)
 Seven Tools of TQM

4) Pareto Chart: A graph to identify and plot

problems or defects in descending order of
frequency
Frequency

Percent
A B C D E
 Pareto Charts
Data for October

70 – – 100
– 93
60 – – 88
50 – 54
– 72
Frequency (number)

Cumulative percent
40 –
30 –
Number of
20 –
occurrences
10 –
0 – 12
4 3 2

Room svc Check-in Pool hours Minibar Misc.

72% 16% 5% 4% 3%

Causes and percent of the total

 Seven Tools of TQM

5) Flowchart (Process Diagram): A chart that

describes the steps in a process
 Flow Charts

MRI Flowchart
1. Physician schedules MRI 7. If unsatisfactory, repeat
2. Patient taken to MRI 8. Patient taken back to room
3. Patient signs in 9. MRI read by radiologist
4. Patient is prepped 10. MRI report transferred to
5. Technician carries out MRI physician
6. Technician inspects film 11. Patient and physician discuss

8
80%
1 2 3 4 5 6 7 11
9 10
20%
 Seven Tools of TQM
6) Histogram: A distribution showing the frequency
of occurrences of a variable

Distribution
Frequency

Repair time (minutes)

 Seven Tools of TQM
7) Statistical Process Control Chart: A chart with
time on the horizontal axis to plot values of a
statistic

Upper control limit

Target value

Lower control limit

Time
 An SPC Chart

Plots the percent of free throws missed

20%
Upper control limit

10%

Coach’s target value

0%

| | | | | | | | |
1 2 3 4 5 6 7 8 9
Lower control limit

Game number
Statistical Process Control (SPC) Chart
 Uses statistics and control charts to tell
when to take corrective action
 Drives process improvement
 Four key steps
 Measure the process
 When a change is indicated, find the
assignable cause
 Eliminate or incorporate the cause
 Restart the revised process
The End: Any Questions?