Professional Documents
Culture Documents
Sigma Black
Belt
Section 1
• Organization Wide Planning and Deployment
Six Sigma Timeline
Johnson &
Control Allied Signal Johnson,
Charts Ford, Nissan,
Motorola General Electric Honeywell
Philosophy
• Generated sustained success
• Project selection tied to organizational strategy
• Customer focused
• Profits
Six Sigma • Project outcomes / benefits tied to financial
Benefits reporting system.
• Full-time Black Belts in a rigorous, project-
oriented method.
• Recognition and reward system established to
provide motivation.
Lean Reduce Waste
Reduced Inventory
Flexible Manufacturing
Improvement • DMAIC
• 8-D
• Kaizen
• Motorola:
• 5-Fold growth in Sales
• Profits climbing by 20% pa
• Cumulative savings of $14 billion over 11
Six Sigma years
• General Electric:
Benefits • $2 billion savings in just 3 years
• The no.1 company in the USA
• Bechtel Corporation:
• $200 million savings with investment of $30
million
• External Sources:
• Voice of Customer
Bulk of Fruit
Process Characterization
and Optimization
Ground Fruit
Logic and Intuition
• Three basic qualifications:
Owner, Managers
Suppliers, Customers
Employees, Partners
Local Community,
Associations, Media
Public
Stakeholders Analysis
Manage
Keep
closely
Satisfied LATENTS KEY PLAYERS
INTEREST
Monitor Keep
informed
APETHETICS DEFENDERS
POWER
Activities to
• set priorities,
• focus energy and resources,
• ensure that employees and other
Strategic stakeholders are working toward
Planning common goals,
• establish agreement around intended
outcomes/ results,
• assess and adjust the organization's
direction in response to a changing
environment.
Strategic plan provides the
direction
Weakness
Strength
that place the business
that gives it an
or project at a
advantage over Strength Weakness
others disadvantage relative
to others
Opportunity Threats
that the business or that could cause
project could exploit trouble for the
to Opportunity business or project
its
Threat
advantage
PEST Analysis
Political Social
Governmen Culural
t aspects
Interventio
n Economic Technological
How business Automation
operates and
innovation
A CONTINGENCY PLAN IS A PLAN
DEVISED FOR AN OUTCOME
OTHER THAN IN THE USUAL
(EXPECTED) PLAN.
Contingency
Planning
Process of creating goals, assigning them Ongoing process Creates a shared vision in the
measurable milestones, and assess organization
progress against those milestones
Portfolio • Selecting the right projects based on the
investment, expected gains and chance of
Analysis success. A
Roles
Champions
Black
Belts
Champions
Black
Belts
Identifies and removes organizational
and cultural barriers to Six Sigma Green Belts
success.
Rewards and recognizes Team Members /
Yellow Belts
Black Belt
Champions
Black
Belts
Champions
Black
Belts
Champions
Black
Belts
Champions
Black
Belts
Sigma •
•
Incorrect scope
Lack of training
Failures • Incorrect project selection
• No alignment to strategic goals
• No financial goals
Change - Lewin's Model
Business case for Monitor Effective Training and/or Counter Provide personal Monitoring of the
the change assumptions and communication skills resistance from counseling implementation
risks improvement the employees
Why Change?
Benefits of successful
implementation
Details of the change
(when? where? Who?
how much? etc.)
Stakeholder Analysis
Change
Management Readiness Assessment
Tools
Communication Plan
Stakeholders Analysis
Owner, Managers
Suppliers, Customers
Employees, Partners
Local Community,
Associations, Media
Public
Stakeholder Analysis
Manage
Keep
closely
Satisfied LATENTS KEY PLAYERS
INTEREST
Monitor Keep
informed
APETHETICS DEFENDERS
POWER
• To assess the level of preparedness for the
change.
• Is there a clear vision?
Readiness • Has that vision effectively communicated?
Manager Medium
Families High
Media Low
Impact on stakeholders
Owner, Managers
Suppliers, Customers
Employees, Partners
Local Community,
Associations, Media
Public
Benchmarking
The process of comparing
to
Features of
Performance products and
Benchmarking benchmarking
services e.g.
mileage, download
speed
External Benchmarking
Internal vs
• Just the revise of internal
External benchmarking
Benchmarking
• What function to benchmark (D)
• What is the current performance level (M)
• Select the Best-in-Class (M)
Benchmarking • Compare (A)
Process • Agree on actions to achieve or beat the Best-
in-Class and Implement (I)
• Monitor (C )
• Re-do
Management support
Lack of Resources
Benchmarking
Challenges Right team
Willingness to change
• Metrics to measure performance
• Commonly used performance measures for
outsiders are financial
Performance • Internal metrics include performance levels
Measures • Process yield
• Defect rate
• Average time to answer a call
• Schedule
• By Robert S.Kaplan and David P. Norton
## ##
Financial Customer
## ##
Return on investment =
Income x 100%
Cost
Payback Period
$2,000
$2,500
$2,500
Cross Functional
Team Types
Virtual
Self Directed
Leader
Facilitator
Team Roles
Coach
Members
Team Roles Provides direction to team members
Coach provides one to one R - Reality Current reality and challenges the
support after training. team has
Coach is the first person to
be contacted if the team has O - Obstacles What stops team from achieving
an issue and needs help. the goal.
GROW Model of Coaching
W – Way forward Steps needed to be taken to
achieve the goal
• Participate in team meetings
Team Roles - • Perform the tasks assigned to them
Members • Actively participate in brainstorming and idea
generation
Selection
Criteria Ability to influence
Openness to change
Team Member Required skillset
Selection
Subject Matter Expertise
Availability
Management support
Clear goals
Team Success
Factors Ground rules
Timeliness
• Abraham Maslow’s Hierarchy
Motivation • Herzberg’s Two Factory Theory
• Douglas McGregor – Theory X and Y
Maslow’s Hierarchy of Needs
Self Actualization
Esteem
Love and
Belongingness Safety
Physiological
Herzberg's two-factor theory
Hygiene Factors
Motivators
Douglas McGregor – Theory X andY
Theory X Theory Y
Dislikes work
Likes work
To be directed
Takes responsibility
Little creativity
Creative
Tuckman Model of Team Life Cycle
Strong dependence on
leader Forming
Simple ideas The forming stage
occurs when
topics
Minimum feedback
Leader
Storming
High Performing
creativity
Openness and When a team reaches
the performing stage
trust it is functioning as a
high performance
team.
Strong
relationships
High achievement
Leaders
Delegate
Adjourning
Encourage
communication
Communication Model
Noise
Sender Receiver
Encod Channel Decod
e e
Feedback
FACE TO FACE VIDEO CHAT TELEPHONE
Communication
Channels
Plan
Kickoff Set goal Face to face Once Team Champion Team Agenda
meeting Team alignment Sr Mgmt (optional) leader Minutes of
meeting
Team Review status Plan Face to face Weekly Team members Team Agenda
meetings for future Champion leader Minutes of
(Optional) meeting
Weekly Update Champion Email Weekly Champion Team Report
status report leader
Monthly Updates for senior Presentatio Monthly Champion Team Report
report management n Senior Management leader
Leader’s Role During Team LifeCycle
Direct Coach Facilitate Reassu
Delegate re
Supportin Delegatin
g g
Willingnes
s
Directin Coachin
g g
• when the followers have low willingness and
low ability for the task at hand.
Leadership • leader takes a highly directive role.
Supportin Delegatin
g g
Willingnes
s
Directin Coachin
g g
• Group Thinking
• Members avoid conflict and agree to a
Group point without critical evaluation.
• Group feels invulnerable
Behaviour • Unquestioned belief in group
• Group pressure on person opposing group
decision.
• Reluctant Participants
Group • Might answer only when specifically asked
a question.
Behaviour • Don’t embarrass them, encourage them to
participate
• Dominant Participants
• Create unbalanced participation in
discussion.
Group • If left unchecked all decisions will be biased
Behaviour based on the opinion of certain people
only.
• Use words such as “let’s listen to what
other opinion we have in the group.”
• Digressions
• Keep meeting on track.
Group • Project Charter might be used to keep the
Behaviour project within scope.
• Could suggest to deal with that issue after
the meeting.
Agenda
Time Management
Right resources
• Brainstorming
Consensus
Team Tools •
• Nominal Group Technique
• Multi-voting
• Brainstorming is a group or individual
creativity technique by which efforts are made
to find a conclusion for a specific problem by
Brainstorming gathering a list of ideas spontaneously
contributed by its member.
• Defer judgment,
• Reach for quantity
FOUR RULES: FOCUS ON WITHHOLD
QUANTITY CRITICISM
Brainstorming
Feedback surveys
Interviews / Meetings
Observations
What is the goal of the survey?
Unambiguous
Scale of 1 to 10
Clarity of questions Historical relevance (to compare year to
year change)
Open ended questions
Analyze
• A group of people are asked about their
perceptions, opinions, beliefs, and attitudes.
Focus Group • Generally 6-10 people having open discussion
with skilled moderator.
Engagement questions
• Start discussion to make
participants comfortable with the
process.
Focus Group Exploration questions
• Main questions
Exit question
• Anything else members want to add
Validity
• Data being recorded is what it was
intended to be?
• How closely do the results correspond to
the true state of the thing being
Data Validity measured?
• – False positive or negative
and Reliability
Reliability
• Method of data collection leads to
consistent results.
• Does the sample of customer surveyed
represent the customer base?
Customer Needsand Requirements
Data collected using Surveys,
Focus Groups, Interviews /
Meetings, Observations
Critical to X (CTX)
SIPOC
Kano Model
Voice of • Voice of the customer (VOC) is a term used to
describe the in-depth process of capturing a
Customer customer's expectations, preferences and
aversions. (stated
(VoC) and unstated)
• Here X can be:
Critical to X • Quality
• Safety
(CTX) • Cost
• Etc.
• VOC’s can be vague and difficult to define,
Critical to that’s where CTQ’s come in. The customer may
identify a requirement that is difficult to
Quality (CTQ) measure directly so it will be necessary to
break down what is meant by the customer
into identifiable and measurable terms
Quality
Function
Deployment
Elements of Charter
• Business Case
• Problem Statement
• Project Scope
• Goals and Objectives
• Project Performance Measures
Project • Charter Review
• Plan – DMAIC start/end
Charter • Team members commitment
Project Charter
• Just the right size:
• Not too big – “solving world hunger issue”
• Not too small
• Is it doable in 2-3 months?
Project Scope • Depth and width of scope:
• Depth is vertical – from purchase>receiving
inspection>production> assembly>dispatch
• Width is horizontal – production > machine
• 1, 2, 3, 4 … / Location a, b, c, d …
• If scope is too wide then you might consider
doing Pareto Analysis and select the “vital
fews”.
• It defines the project starting point and ending
point.
Project Scope • “From receiving of welding consumables - to
the
• testing of weld”
• Defines in and out of scope.
• In scope: Welding at weld shop
• Out of scope: Welding done during assembly
• First step in Six Sigma
• Elements of Charter
• Business Case
• Problem Statement:
Measurable
Achievable
SMART
Relevant
Time bound
Performance
Measures
Customer satisfaction
Nonmonetary
Employee satisfaction
Benefits
Reputation
• First step in Six Sigma
• Elements of Charter
• Business Case
Creates buy-in of
the key
stakeholders
1. Gantt charts
2. Toll-gate reviews
Project
Management
3. Work breakdown structure (WBS)
Tools
Review
MEASURE IMPROVE
Management
Tools 3. Work breakdown structure (WBS)
Management
Tools
Analytical Interrelationship
Diagram
2. Tree diagrams
Tools 5. Activity
3. Matrix
network
diagrams
diagrams
4.
Prioritization
matrices
After brainstorming
To organize large number
Affinity of ideas into natural
groups
Diagrams Interviews/surveys –
open ended questions
Also known as K-J
Method after Kawakita
Jiro who developed this
tool.
Affinity Diagrams
Small videos
Interactive
To the point Slides
More quizzes
Notes
Easy to
Flash
understan Detailed
cards
Practice d coverage
Cover the
BoK
To breakdown a goal or broad
category into fine level of details
Tree Diagram
Finer details of activities that
contributes to the issue.
Tree Diagrams
Org.
support
Motiv
ation
Finan
cial
Passing Binde
Handboo
r
CSSBB
k
exam Book
Youtub
s
e
Videos
Udemy
Quizzes
Matrix • Shows the relationship between two or more
Diagram groups.
Matrix
Diagrams – Section
1
Section
2
Section
3
Section
4
Section
5
L Shaped
Difficult 1 5 3 2 1
y Level
Conceptual 4 3 2 1 1
Knowledge
Statistical 1 1 1 1 5
Knowledge
Business 2 4 5 4 2
Application
Matrix Returns 5 3 1 1 3
Diagrams – Complaints 5 4 1 3 2
T Shaped
Sale 1 2 5 4 5
Efficiency 2 3 5 2 1
Pickup 1 3 4 5 3
Look 1 2 5 4 4
Comfort 2 1 4 2 3
Matrix Diagrams – X Shaped
Matrix
Diagrams –
Roof Shaped
Prioritization
Matrix It is used to
compare choices
Or to select a
project
Activity • To manage number of tasks in a sequence.
• List down tasks
Network • Time for each task
• Predecessor and successor tasks
Diagram • Identify bottlenecks
Activity Duration Depends
on
A 2
Activity Network Diagram B 4 A
C 1 A
D 2 B
CPM (Critical Path Method) is E 7 B, C
commonly used with all forms of
projects
It includes:
A list of all activities required to complete
the project
The time (duration) that each activity will
take to complete,
The dependencies between the activities
and,
Logical end points such as milestones or
deliverable items.
Activity Float
Network
• float or slack is the amount
Diagram of time that a task in a
project network can be
delayed without causing a
delay
Critical Path
• An activity on critical path
has "zero free float"
Activity Network Diagram
A 2
B 4 A
C 1 A
D 2 B
E 7 B, C
Program Evaluation and Review Techniques(PERT)
Metrics
Work in
Queue (WIQ)
Some of the Work
in Progress may not
be in the queue.
Touch Time
TIME THAT MATERIAL IS THIS EXCLUDES MOVING,
ACTUALLY BEING WORKED ON. WAITING TIME.
Takt Time
1 3.5 1
Hours Hours Hours
Total time a unit takes from the time the
work is requested to the time it is delivered.
• Example:
• WIP is 50 units
• Throughput is 10 units per day
Lead Time • Lead Time = 50/10 = 5 days
Throughput is the average
number of units processed per
time unit.
Process maps
Analysis Tools
Spaghetti diagrams
Circle diagrams
Gemba walk
Flow Chart • Shows the processes as boxes connected by
arrows and decision boxes.
BASIC SWIM LANE
FLOWCHARTS
Types of
Flowcharts
VALUE STREAM
MAPS
Flow chart and process map
are used interchangeably
Process
Mapping
Process mapping is the
process of creating a
diagram; the diagram itself
is called a flow chart.
• Type of Flow Chart showing how value flows
Value Stream through the organization
Maps • Flow of material
• Flow of information
Value
Stream
Maps
SigmaXL demonstration
Value Stream
Maps
Symbols
Value Stream Maps
Used to analyze waste
Venn Diagram
Pie Chart
Sales
Knowledge 9%
1st Qtr
10% 2nd Qtr
58% 3rd Qtr
23% 4th Qtr
Experience Attitude
Gemba Gemba = work area or shop floor
Walk
Originated from “gembutsu”, which
Data
Description Numbers
Measurements: Length, height, time
More information with less samples
Continuous
More sensitive
Data
Provide more information
More expensive to collect
Continuous vs
Discrete
Discrete
Count: Number of students, Number of heads
Data
Measurement Scales
Data
Data
Example:
Color: Blue, Green,
Red
Measurement Scales
Data
Data
Example
: Temperature:
Celsius
Measurement Scales
Data
Example
: Height, mass,
volume
Measurement
Scales
Nominal Ordinal Interval Ratio
Ordered N Y Y Y
Difference N N Y Y
Absolute Zero N N N Y
Data
Cluster Sampling
Accidental / Convenience
Probability
Sampling
Sampling
Judgemental Sampling
Non
Quota Sampling
Simple Random Sampling
Each item in the population has an
equal chance of being selected.
Examples: Using random tables,
Random draw of lot (lottery)
Sampling Stratified
Sampling
Systematic Random Sampling
Select elements at regular intervals
through that ordered list.
Example: Checking every 6th piece
produced by the machine.
Sampling Stratified
Random Sampling
Cluster Sampling
Stratified Random Sampling
Used to ensure that sub-groups within a
population are represented
proportionally in the sample.
Example: If 10 people are drawn to
represent a country, 5 of them are male
and 5 females to avoid the sex bias.
Simple Random Sampling
Systematic Random Sampling Stratified
Probability
Sampling
Judgemental Sampling
Non
Quota Sampling
Judgmental Sampling
The researcher chooses the sample
based on who they think would be
appropriate for the study.
Example: Auditor selects a sample
based on the concerns he/she had in
the earlier audit
Judgemental Sampling
Non
Quota Sampling
Quota Sampling
A quota is established and auditor are
free to choose any sample they wish
as long as the quota is met.
Example: 2% of the calibration records.
Judgemental Sampling
Non
Quota Sampling
Why you
need to Goal and Objective
collect data?
Collection Operational
Definition
Type of data – NOIR
Manual or Automatic
Plan
Past data vs Future
Is data
reliable?
Data Operational How is it Type of Sample Data Recording Comment
Collection
Measurement Who?
Definition measured? Data size Form s
Time to Time from picking up Using a stop Continuo Every 10th Operato Assembly Record
assemble the first piece to watch us piece r F-0156
placing the assembled
Plan
item in tray Ratio
• Adding, Subtracting
• Example: -95, -97, -98, -90
• Add 100 to each: 5, 3, 2, 10
– Missing
Data Delete the row
Measurement
System Analysis Process has variation.
(MSA)
• Accuracy
Measurement • Bias
• Linearity
System • Stability
Accuracy
Bias
Linearity
Stability
Precision
Repeatability
Reproducibility
Reference Measured
Bias Value (psi) Value (psi)
100 100
100 102
100 101
100 100
Average 101
Bia
s
Bias
Bias is the
systematic error.
Bias is addressed
by calibration.
Linearity
0 0 0
1.5
50 50.5 0.5
1
100 101 1
0.5
150 151.5 1.5
0
200 202 2 0 50 100 150 200 250
Stability
Stability measures the bias over
time. Also known as drift.
Time
Time 2
Bias 2
Time
1
Bias 1
Accuracy vs Precision
Accuracy
“Closeness” to the true value, or to an
accepted reference value.
Bias
Linearity
Stability
Precision
“Closeness” of repeated readings to each
other
Repeatability
Reproducibility
Repeatability
Variation in measurements obtained with one measuring
instrument when used several times by an appraiser.
Also called Equipment Variation (EV)
It’s the capability of the gauge to produce consistent results.
Reproducibility
Variation in the average of the measurements made by different
appraisers using the same gage
Also called Appraiser Variation (AV)
It’s the capability of the appraiser to produce consistent results.
Gage R&R (GRR)`
Combined estimate of repeatability
and reproducibility.
Gage R&R (GRR)
Combined estimate of repeatability
and reproducibility.
Precision to • How capable your measurement system is?
• Precision/Tolerance (P/T) is the ratio between
Tolerance the estimated measurement error (precision)
and the tolerance of the characteristic being
Ratio measured.
Precision to Tolerance Ratio
P/T ratio is the most common
estimate of measurement
precision
system Measurement
Toleran System
ce Variation
P/T = 10%
P/T = 100%
P/T = 200%
LSL USL
Precision to Tolerance Ratio
PTR
PTR = 5.15 σms
USL-LSL
Gage R&R
(GRR) – Two operators usually
measure 5 parts one time
Range each.
Method
Gage R&R (GRR) –
Range Method Part Appraise Appraise Rang
• Two operators usually measure 5 parts one # rA rB e
time each. 1 100 101 1
• GRR = R-bar/d2 = 1.6/1.19 = 1.34
• Process standard deviation = 2.23
2 99 102 3
• (From previous study) 3 97 99 2
• %GRR = 100 (GRR/process std dev) 4 103 103 0
• %GRR = 100 x 1.34 / 2.23 = 0.60
5 101 103 2
• Measurement variation consumes 60% of
the process variation. Measurement R-bar 1.6
system needs improvement. =
GRR –
Average and
Range
Method AVERAGE AND RANGE METHOD
PROVIDES THE ESTIMATE OF BOTH
INTERACTION BETWEEN APPRAISER AND
GAGE IS NOT COMPUTED (ANOVA
REPEATABILITY AND REPRODUCIBILITY. METHOD DOES THAT)
GRR – ANOVA Method
ANOVAis Analysis of Variance
Resolution:
Distinct
Categories
(NDC)
• Value of NDC
• Less than 2, the measurement system is of
Number of no value for controlling the process, since
one part cannot be distinguished from
Distinct another.
• Equal to 2, the data can be divided into
Categories two groups, say high and low.
• Equal to 3, the data can be divided into 3
(NDC) groups, say low, middle and high.
• More than 5, denotes an acceptable
measurement system.
Measurement System Analysis (MSA)
Crossed gage R&R study
A study in which each operator
measures each part. (Non-destructive)
Focus on those
All functions in the business measurements which ensure
have measurement systems the overall business success,
in place. not the one which are easy
to measure.
Measurement
Systems
• Number of customercomplaints
• Percent of itemsreturned
Measurement • Percent sales from returning customers
Systems - Sales •
•
Salesgrowth
Market share
• Customer satisfactionrating
Measurement Average number of
Systems - suppliers for key
Purchase components
Number of strategic
alliances
Rejection rate
Value added per Internal
Systems -
Production Work in Progress
Equipment
Utilization
• Ensure that your measurements are free from
Measurement bias
Units
metre m length L
kilogram kg mass M
second s time T
Units kilogram
second
The mass of the international prototype kilogram.
transition between the two hyperfine levels of the ground state of the
caesium-133 atom.
mole
The amount of substance of a system which contains as many elementary
entities as there are atoms in 0.012 kilogram of carbon-12
Calibration • Toensureconsistency
• Todetermine the accuracyreadings
• Canmeasurements be trusted
• “Property of a measurement result whereby
the result can be related to a reference
Traceability through a documented unbroken chain of
calibrations, each contributing to the
measurement uncertainty.”
Traceability
SI Units
Party Labs
Metrology
Department
Working Instruments
Basic Statistical Terms
Sample:
Population: Samplin Part of
Complete g population
collection Proces
to be s
studied
Inference
Parameter Statistic
Characteristic of Characteristic
a population of asample
N number of members n
μ mean standard x̅
σ deviation s
Notations Mean
POPULATION
PARAMETERS
μ
SAMPLE
STATISTICS
x̄
Standard Deviation σ s
Variance σ2 s2
Proportion of population P p
having an attribute
Proportion of population not Q (=1-P) q (=1-p)
having an attribute
Correlation coefficient ρ r
Number of elements N n
Central Limit Theorem
Central Limit • For almost all populations, the sampling
distribution of the mean can be approximated
Theorem closely by a normal distribution, provided the
sample size is sufficiently large.
Central Limit Theorem
Central Limit
Theorem
Central Limit
Theorem
Distribution of
Individual Observations
Distribution of
Sample Means
Standard Error of Means
Standard deviation of the
sampling distribution of the
sample mean
Called “standard error of the
mean”
Descriptive Statistics
Descriptiv
e
Statistics
Central
Variability
Tendenc
y
Mean Rang
e
Standard
Mod
Deviation
e
Interquartil
Median
e Range
Percentil
e
Mean
Also known asAverage Central
Tendenc
y
Mode = 10 Quartile
Median
Middle value when put in ascending Central
Tendenc
or descending order. y
Median = 10
descending order y
set
If i is whole number – Percentile
is
If i average
is “not” aof (i)th and
whole (i+1)th
number –
location
Percentile is located at (i+1)th
whole-num.
Example: 6,9,10,11, 11,14
Q1=9, Q2=10.5, Q3=11
Descriptive Statistics
Variability
Interquartile Standard
Rang
Range Deviation
e
Range
Difference between lowest and Variability
Range = 14-6 = 8
Interquartile Range
Range of middle 50%data Variability
deviation
Standard Deviation
x x-x̅ (x-x̅ )2
100 0 0 ∑(x-x̅
S2 = )2
101 1 1 n-1
99 -1 1
102 2 4 S 2 = 10/5 = 2
98 -2 4
S =√2=
100 0 0
1.414
x̅ =100 ∑(x-x̅ )=0 ∑(x-x̅ )2=10
• Box-and-whisker plots
• Scatterdiagrams
Graphical • Histograms
Methods •
•
Normal probability plots
Frequencydistributions
• Cumulative frequencydistributions.
Box and Whisker Plots
Also known as Box Plot
Shows the median
Shows Q1, Q3 and IQR
70
60
50 Median
25th
40
75th
30
20 Mean
70
60
50 Median
25th
40
75th
30
20 Mean
10
15
20
0
5
7.
1
11.
7
16.
2
20.
8
mo
25.
4
30.
0
34.
MS Excel
5
39.
1
43.
8
57.
4
Frequen
cy
10
15
20
25
30
0
5
7.1
12.5
8
18.0
6
23.5
n
Bi
4
29.0
2
34.5
39.9
Demonstration Using SigmaXL and
8
45.4
6
50.9
4
Frequency
56.4
2
More
• Hypothesis:
• A claim that we want to test
Valid • Null Hypothesis - H0
Statistical • Default position / Currently accepted
position / Assumed / Status Quo
Conclusions • Alternate Hypothesis – Ha
• Claim to be tested. Also known as
Research Hypothesis or the other option.
Valid Statistical Conclusions
Null Hypothesis and Alternate
Hypothesis are pair and cover all
possibilities.
Only one of these has to stand and
not both.
Types of Errors
True State of Nature
H0 Ha
Is true Is true
Support H0 /
Reject Ha Correct Type II Error
Conclusion
Conclusion Support Ha /
Reject H0 Type I Error Correct
Conclusion
Probability
Classic Model
Probability • Mean
z is the z-score,
X is the value of the element,
μ is the population mean,
σ is the standard deviation.
Z Table
Z Table in Excel
NORMSDIST(z) returns
p
NORMSINV(p) returns z.
Continuous • Normal probability distribution
Discrete Hypergeometric
Geometric
Probability Probability
Distribution
Distribution
Distributions
Negative Poisson
Geometric Probability
Distribution Distribution
Binomial Probability Distribution
A binomial experiment has the
following properties:
The experiment consists of n repeated
trials.
Each trial can result in just two possible
outcomes. We call one of these
outcomes a success and the other, a
failure.
The probability of success, denoted by
p, is the same on every trial.
The trials are independent; that is, the
outcome on one trial does not affect the
outcome on other trials.
Binomial Probability Distribution
A binomial experiment has the
following properties:
The experiment consists of n repeated
trials.
Each trial can result in just two possible
outcomes. We call one of these
outcomes a success and the other, a
failure.
The probability of success, denoted by
p, is the same on every trial.
The trials are independent; that is, the
outcome on one trial does not affect the
outcome on other trials.
Binomial Probability Distribution
x: The number of successes that result from the P(x) = nCx . px . (1 - p)n -
binomial experiment. x
n: The number of trials in the binomial
experiment.
p: The probability of success on an
individual trial.
q: The probability of failure on an
individual trial. (This is equal to 1 - p.)
n!: The factorial of n (also known as n
factorial).
P(x) : Binomial probability - the
probability that an n-trial binomial
experiment results in exactly x successes,
nwhen thenumber
Cx: The probability of successof on
of combinations n an
individual
things, trial xisat
taken p.a time.
Binomial Probability Distribution
The binomial probability refers to
the probability that a binomial
experiment results in exactly x
successes.
Suppose a binomial experiment
consists of n trials and results in x
successes. If the probability of
success on an individual trial is p,
then the binomial probability is:
P(x) = nCx . px . (1 - p)n - x
or
P(x) = { n! / [ x! (n - x)! ] } . px . (1 - p)n - x
Binomial Probability
Distribution
The mean of the distribution (μx) n: The number of trials in the
is binomial experiment.
n.p p: The probability of success on
an individual trial.
The variance (σ2x) is
n.p.(1-p)
and Ppk • Cpk is for short term and Ppk is for long term.
Taguchi Capability Index - Cpm
Process Capability
Select the process
Capability
Studies
Confirm normality of data
Frequen
15
10
cy
5
0
804
156
231
307
383
459
534
46
9
Cycle Time
(Minutes)
SigmaXL 1.83
NSCORE
Histogram
0.83
-0.17
-1.17
SigmaXL > Customer Data > Graphical -2.17
-4000.00
-2000.00
2000.0
4000.0
6000.0
8000.0
0.0
Tool> Basic Histogram
0
Cycle Time
0
(Minutes)
Anderson-Darling Normality Test 5.357
Defect vs
DISSATISFACTION.
Defective
• Circumstances in which CTQ can fail to meet.
• Number of defect opportunities relate to
complexity of unit.
Defect • Complex units – Greater opportunities of defect
than simple units
Opportunity • Examples:
• A units has 5 parts, and in each part there are 3
opportunities of defects – Total defect
opportunities are 5 x 3 = 15
• Number of defects divided by number of defect
Defects Per opportunities
• Examples:
Opportunity • In previous case (15 defect opportunities), if 10
units have 2 defects.
(DPO)
• DPO = 2 / (15 x 10) = 0.0133333
• DPO multiplies by one million
• Examples:
Defect Per • In previous case (15 defect opportunities), if 10
units have 2 defects.
Million
Opportunities • DPO = 2 / (15 x 10) = 0.0133333
(DPMO) • DPMO = 0.013333333 x 1,000,000 = 13,333
• DPU=(10+15+5)/3,000 =30/3,000
• =1/100 =0.01
Rolled Through Yield
Units entering a process = P
Defective Units = D
Yield = (P-D)/P
Y1 = 0.99, Y2 =0.95, Y3=0.98
RTY= Y1 . Y2 . Y3 = 0.99x0.95x0.98
=0.92169
Short Term vs Long Term Capability
Cp Cpk – Short term
z’ = .5[ln(1+r) – ln(1-r)]
• r = 0.88, r2 = 0.77
Regression • Quantifies the relationship between Y and X (Y
Analysis = a + bX)
Regression Analysis
Quantifies the relationship between
Y and X (Y = a + bX)
Hours Studied (X) Test Score %(Y) XY X2 Y2
20 40 800 400 1600
24 55 1320 576 3025
46 69 3174 2116 4761
62 83 5146 3844 6889
22 27 594 484 729
37 44 1628 1369 1936
45 61 2745 2025 3721
27 33 891 729 1089
65 71 4615 4225 5041
23 37 851 529 1369
SUM 371 520 21764 16297 30160
Regression Analysis
Quantifies the relationship between
Y and X (Y = 15.79 + 0.97.X)
Hours
Test
Studied XY X2 Y2
Score
(X)
% (Y)
20 40 800 400 1600
24 55 1320 576 3025
46 69 3174 2116 4761
62 83 5146 3844 6889
22 27 594 484 729
37 44 1628 1369 1936
45 61 2745 2025 3721
27 33 891 729 1089
65 71 4615 4225 5041
23 37 851 529 1369
SU 371 520 21764 16297 30160
Regression • For a student studying 50 hrs what is the
Analysis expected test score %?
Residual
Analysis
Y=15.79 +0.97.X
Residual Analysis – No pattern
Residual
20
15
10
0
0 20 30 40 50 60 70
-5 1
0
-10
-15
Residual
• Confidence interval
• 95% confidence interval, representing a
Confidence range of likely values for the mean
response.
Interval - • Prediction interval
Slope • 95% prediction interval, represents a range
of likely values for a single new
observation.
• Simple LinearRelation
• Y=a + bX
• Multiple LinearRegression
Multivariate • Y=a +b1X1+b2X2+ ……..+bnXn
• Multicollinearity
Tools • When two input variables(predictor
• variables - Xs) are correlated.
• Multivariate
• Two or more dependent variables(Ys)
Factor analysis / Principal
Component Analysis
• Cause-effect relationship
• Tools
• Classification of individuals
• Discriminant Analysis
Multivariate • Dimensionreduction
• Principal Component Analysis/Factor
• Analysis
• Cause-effectrelationship
• MANOVA
Discriminant Analysis
Explains how clusters are different
Principal Component
Analysis/ Factor Analysis
M A N O VA
Tests the effect of one or more
independent variables on two or more
dependent variables.
H0 Ha
Is true Is true
Support H0 /
Reject Ha Correct Type II Error
Conclusion
Conclusion Support Ha / Correct
Reject H0 Type I Error Conclusion
(Power)
Errors of StatisticalTests
Type I error (alpha) Type II error (beta)
Name Producer’s risk/ Consumer’s risk
Significance level
1 minus error is Confidence level Power of the test
called
Example of Fire False fire alarm leading Missed fire leading to
Alarm to inconvenience disaster
Effects on process Unnecessary cost Defects may be produced
increase due to frequent
changes
Control method Usually fixed at a pre- Usually controlled to < 10%
determined level, 1%, by appropriate sample size
5% or 10%
Simple definition Innocent declared as Guilty declared as innocent
guilty
• Level of Confidence / Confidence Interval: C =
0.90, 0.95, 0.99 (90%, 95%, 99%)
Significance
Level • Level of Significance:
• α = 1 – C (0.10, 0.05, 0.01)
• Power = 1 – β (or 1 - type II error)
• Type II Error: Failing to reject null hypothesis
when null hypothesis is false.
Z Critical
• p value is the lowest value of alpha for which
the null hypothesis can be rejected.
(Probability that the null hypothesis is correct)
p Value • If p = 0.01 you can reject the null hypothesis
at α = 0.05
• p is low the null must go / p is high the null fly.
n = ( zα/2 . σ / ME)2
n is sample size
σ is standard deviation
One sample p
test
Two sample z
test Two
Two
Tests Sample
s Paired t test
sample t test
Two sample
standard deviation
More than2
ANOVA
samples
• Calculated value
• z = [x̄ - μ ] / [σ / sqrt( n ) ]
• Example: Perfume bottle producing 150cc
with sd of 2 cc, 100 bottles are randomly
One Sample picked and the average volume was found to
be 152cc. Has mean volume changed? (95%
z Test confidence)
• zcalculated = (152-150)/[2 / sqrt( 100 ) ] = 2/0.2 =
10
• zcritical = ?
One
Sample z
Test
zcritical =1.96
One Sample z Test
Calculated value
z = [x̄ - μ ] / [σ / sqrt( n ) ]
Example: Perfume bottle producing
150cc with sd of 2 cc, 100
bottles are randomly picked and
the average volume was found to
be 152cc. Has mean volume
changed? (95% confidence)
zcalculated = (152-150)/[2 / sqrt( 100 ) ]
= 2/0.2 = 10
zcritical = 1.96 > Reject Ho
• Calculated value
• t = [x̄ - μ ] / [s / sqrt( n ) ]
• Example: Perfume bottle producing 150cc, 4
One Sample bottles are randomly picked and the average
volume was found to be 151cc and sd of
t Test sample was 2 cc. Has mean volume changed?
(95% confidence)
• tcal = (151-150)/[2 / sqrt( 4 ) ] = 1/1 = 1
• tcritical = ?
One SampletTest
• tcritical =3.182
One Sample t Test
Calculated value
t = [x̄ - μ ] / [s / sqrt( n ) ]
Example: Perfume bottle producing
150cc, 4 bottles are randomly picked
and the average volume was found to
be 151cc and sd of sample was 2 cc.
Has mean volume changed? (95%
confidence)
tcal = (151-150)/[2 / sqrt( 4 ) ] = 1/1 = 1
tcritical = 3.182 > Fail to reject Ho
One Sample p Test
H0: p = p0
Calculated value
One sample p
test
Two sample z
test Two
Two
Tests Sample
s Paired t test
sample t test
Two sample
standard deviation
More than2
ANOVA
samples
Two Samplez Test
Null hypothesis: H 0: μ 1 = μ 2
or H 0: μ 1 – μ 2= 0
Alternative hypothesis: H a : μ 1 ≠ μ
2
• Example: From two machines 100 samples
each were drawn.
Two Sample • Machine 1: Mean = 151.2 / sd = 2.1
H 0: μ 2 – μ 1 <= 0.3
H a: μ 2 – μ 1 > 0.3
Two Samplez Test
Example: From two machines 100
samples each were drawn.
Machine 1: Mean = 151.2 / sd = 2.1
Machine 2: Mean = 151.9 / sd = 2.2
Is there difference of more than 0.3 cc
in these two machines. Check at 95%
confidence level.
Zcal = (151.2 – 151.9) – (-0.3)/0.304
= -0.4 / 0.304 = -1.316
Zcritical = 1.64
Fail to reject Null Hypothesis.
• If two set of data are independent or
dependent.
• If the values in one sample reveal no
information about those of the other
• sample, then the samples are independent.
Two Sample • Example: Blood pressure of
male/female
t Test • If the values in one sample affect the
values in the other sample, then the
samples are dependent.
• Example: Blood pressure before and
after a specific medicine
Two Sample t Test
If two set of data are independent or
dependent.
If the values in one sample reveal no
information about those of the other
sample, then the samples are independent.
Example: Blood pressure of
male/female
Two sample t test
Two Sample
t Test
Two Sample • Assumptions: Normality, independent random
t Test samples, population variances are equal
Two Sample t Test
What if variance of two samples is
A C
qual?
not e 150 144
152 162
154 177
152 150
151 140
2 Sample t-Test
TestInformation
H0: Mean Difference =0
Ha: Mean Difference Not Equal To0
Assume Unequal Variance
Results: A C
Count 5 5
Mean 151.80 154.60
Standard Deviation 1.483 15.027
Results: A C
Count 5 5
Mean 151.80 154.60
Standard Deviation 1.483 15.027
Two SampletTest
• tcritical =2.776
Two Sample t Test
Minitab 17 output:
Two-sample T for A vs C
2 122 120 -2
3 143 141 -2
• Example: Before and after
medicine BP was
measured. Is there a 4 100 109 9
difference at 95%
confidence level? 5 109 109 0
tcal. = 1.4/2.04 = 0.69
t0.025, 4 =2.766
Fail to reject null hypothesis
Testsfor Mean, Variance & Proportion
One sample ztest
One sample p
test
Two sample z
test Two
Two
Tests Sample
s Paired t test
sample t test
Two sample
standard deviation
More than
ANOVA
2
samples
Two Sample p Test
Null hypothesis: H 0: p1 = p2
or H 0: p 1 – p 2= 0
Alternative hypothesis: H a : p 1 ≠ p
2
F critical =
4.1203
Two Sample Variance – F Test
Example: We took 8 samples from
machine A and the standard
deviation was 1.1. For machine B
we took 5 samples and the
variance was 11. Is there a
difference in variance at 90% F critical =
confidence level? 4.1203
n1 = 8, s1 = 1.1, s21 = 1.21, df = 7 (denominator)
n2 = 5, s22 = 11, df = 4 (numerator)
Fcalculated = 11/1.21 = 9.09 (higher value at top)
Reject H0
Two Sample Variance – F Test
Right tail F critical = 4.1203
Left tail F critical =?
Reverse degrees of freedom and
then take inverse.
F critical =
F (4,7) =4.1203
4.1203
F (7,4) =6.0942
Inverse of this is 1/6.0942 is F=0.164
• F-test
• for testing equality of two variances from
different populations
• for testing equality of several means with
technique of ANOVA.
Tests for • Chi-square test
Variance • For testing the population variance against
a specified value
• testing goodness of fit of some probability
distribution
• Fail to reject H0
One Sample Chi Square
SigmaXLOutput
• F-test
• for testing equality of two variances from
different populations
• for testing equality of several means with
technique of ANOVA.
ANOVA • Chi-square test
• For testing the population variance against
a specified value
• testing goodness of fit of some probability
distribution
• H0: σ2 = σ2
ANOVA • 1 2
• F calculated
4 x
3 x 3 vs 4
2 x 2 vs 3 2 vs 4
1 x 1 vs 2 1 vs 3 1 vs 4
1 2 3 4
• Why ANOVA?
• How many t Test we need to conduct if
• have to compare 4 samples? … 6
ANOVA • Each test is done with alpha = 0.05 or 95%
confidence.
• 6 tests will result in confidence level of
0.95x0.95x0.95x0.95x0.95x0.95 = 0.735
ANOVA
Comparing three machines:
150
148
Machine 1 Machine 2 Machine 3
146
ANOVA
Comparing three machines:
ANOVA • Ratio:
• Sbetween(or treatment) / Swithin(orerror)
ANOVA
150 153 156
151 152 154
152 148 155
152 151 156
Machine 1 x1 - x̄1 Sqr(x1 - x̄1) Machine 2 x2 - x̄2 Sqr(x2 - x̄2) Machine 3 x3 - x̄3 Sqr(x3 - x̄3)
150.00 -1.00 1.00 153.00 2.17 4.69 156.00 0.50 0.25
151.00 0.00 0.00 152.00 1.17 1.36 154.00 -1.50 2.25
152.00 1.00 1.00 148.00 -2.83 8.03 155.00 -0.50 0.25
152.00 1.00 1.00 151.00 0.17 0.03 156.00 0.50 0.25
151.00 0.00 0.00 149.00 -1.83 3.36 157.00 1.50 2.25
150.00 -1.00 1.00 152.00 1.17 1.36 155.00 -0.50 0.25
151.00 150.83 155.50 152.4
4
SSwithin = 4.00+18.83+5.50
4.0
0
= 28.33
18.8
3
5.50
Machine 1 Machine 2 Machine 3
ANOVA
150 153 156
151 152 154
152 148 155
152 151 156
Machine 1 x1 - x̄1 Sqr(x1 - x̄1) Machine 2 x2 - x̄2 Sqr(x2 - x̄2) Machine 3 x3 - x̄3 Sqr(x3 - x̄3)
150.00 -1.00 1.00 153.00 2.17 4.69 156.00 0.50 0.25
151.00 0.00 0.00 152.00 1.17 1.36 154.00 -1.50 2.25
152.00 1.00 1.00 148.00 -2.83 8.03 155.00 -0.50 0.25
152.00 1.00 1.00 151.00 0.17 0.03 156.00 0.50 0.25
151.00 0.00 0.00 149.00 -1.83 3.36 157.00 1.50 2.25
150.00 -1.00 1.00 152.00 1.17 1.36 155.00 -0.50 0.25
151.00 150.83 155.50 152.4
4
4.00 18.83 5.50
MSwithin = SSwithin(or error) /df within x̄1 = 151.00 x̄2 = 150.83 x̄3 = 155.50
DEMONSTRATE MS Excel
Machine 1 Machine 2 Machine 3
ANOVA
150 153 156
151 152 154
152 148 155
H0: Mean 1 = Mean 2 = ... = x̄1 = 151.00 x̄2 = 150.83 x̄3 = 155.50
Mean k
Ha: At least one pair Mean i ≠
Mean
j
ANOVA Table
Source SS DF MS F P-Value
Between 84.111 2 42.056 22.26 0.0000
5
Within 28.333 15 1.889
Total 112.44 17
ANOVA
Practice Exercise: Fill in the values
for ?A to ?E in this ANOVA Table:
ANOVATable
Source SS DF MS F
Between 84.111 ?C ?D ?E
Within ?A 15 1.889
Total ?B 17
• ?A=15x1.889=28.333
• ?B=84.111+28.33=112.44
• ?C=17-15=2
• ?D= 84.111/2 =42.056
• ?E=42.045/1.889 =22.265
• To test if the sample is coming from a
Goodness of population with specific distribution.
• Other goodness-of-fit tests are
Fit Test (Chi • Anderson-Darling
• Kolmogorov-Smirnov
Square) • Chi Square Goodness of Fit can be used for any
time of data: Continuous or Discrete.
H0: The data follow a specified
distribution.
X2 = 25.8
Goodness of Fit Test (Chi Square)
A coin is flipped 100 times. Number of
heads are noted. Is this coin biased?
X2cal= 25.8
X2(4,0.95)= 9.49
Goodness of Fit Test (Chi Square)
A coin is flipped 100 times. Number
of heads are noted. Is this coin
biased?
X2cal= 25.8
X2(4,0.95)= 9.49
Reject Null Hypothesis
Coin is biased
Contingency Tables
To find relationship between
two discrete variables.
Smoker Non
Smoker
Male 60 40 100
Female 35 40 75
95 80 175
EXPECTED
Operator 1 Operator 2 Operator 3
Shift 1 122x71/347 112x71/347 115x71/347 71
Shift 2 122x116/347 112x116/347 115x116/347 116
Shift 3 122x160/347 112x160/347 115x160/347 160
122 112 115 347
Contingency Tables
Calculate Chi square statistic.
EXPECTED
Operator 1 Operator 2 Operator 3
Shift 1 122x71/347 112x71/347 115x71/347 71
Shift 2 122x116/347 112x116/347 115x116/347 116
Shift 3 122x160/347 112x160/347 115x160/347 160
122 112 115 347
EXPECTED
Operator 1 Operator 2 Operator 3
Shift 1 24.96 22.91 23.53 71
Shift 2 40.78 37.44 38.44 116
Shift 3 56.25 51.64 53.02 160
122 112 115 347
Contingency Tables
Calculate Chi square statistic.
OBSERVED EXPECTED
Operator 1 Operator 2 Operator 3 Operator 1 Operator 2 Operator 3
Shift 1 22 26 23 71 Shift 1 24.96 22.91 23.53 71
Shift 2 28 62 26 116 Shift 2 40.78 37.44 38.44 116
Shift 3 72 22 66 160 Shift 3 56.25 51.64 53.02 160
122 112 115 347 122 112 115 347
Opera Opera
(O-E)2/E Operator 1
t or t or
2 3
Shift 1 (22-24.96)2/24.96 = 0.35 0.42 0.01 71
Shift 2 (28-40.78)2/40.78 = 4.00 16.11 4.03 116
Shift 3 (72-56.25)2/56.25 = 4.41 17.01 3.18 160 X2 = 49.52
122 112 115 347
Contingency Tables
Calculate Chi square statistic = 49.52
Degrees of freedom = (r-1)(c-1) = 4
Chi square critical = 9.49
Reject null hypothesis
There is a relationship between the
shift and the operator.
Contingency Tables
• Practice Exercise:
Smoker Non
• Calculate the Expected value for Non
Smoker Male?
Smoker
• What will be the degrees of freedom in Male 60 40 100
this example?
Female 35 40 75
95 80 175
Smoker Non
Smoker
Contingency
Tables
Male 60 40 100
Female 35 40 75
95 80 175
• Practice Exercise:
• Calculate the Expected value for Non Smoker Male? = 80x100/175 =
45.71
• What will be the degrees of freedom in this example? (2-1)(2-1)=1
• Parametric
• Assumes about the population from which
the sample has been drawn (e.g. Normally
distributed)
• Data is ratio or interval level
Parametric • Non Parametric
• Makes no assumption about the
vs Non population from which the sample has
been drawn
Parametric • Normally or small size data. No minimum
sample size.
• Data is ratio, interval, nominal or ordinal
level
• Less power (More likely to make Type II
error)
Parametric vs Non Parametric
Design Proces
FMEA s
FMEA
- System
Productio Assemb
- Subsyste n ly
m
FMEA FMEA
- Compone
nt - System - System
FMEA - Subsyste - Subsyste
m m
- Compone - Compone
nt nt
FMEA FMEA
• Failure Mode and Effect Analysis:
FMEA • Severity
• Severity 1 – No effect/ client might not even
notice it
• Severity 10 – Serious safety hazard without
warning
• Occurrence
• Occurrence 1 – Rare event, no data of such
type of failure in past
• Occurrence 10 – Failure almost inevitable
FMEA • Detection
• Detection 1 – Current system almost certainly
detect the problem (automation)
• Detection 10 – Current system can not detect
the problem
• Identify key processsteps
• Identify failuremode
• Identify failureeffects/severity
• Identify causes/occurrence
FMEA • Identify controls/detection
• Calculate RiskPriority Number(RPN)
• Prioritize by RPN–Higher RPNfirst
• Determine actionplan
• RecalculateRPN
• Update FMEA when there is plan to change /
actual change of :
• Design
FMEA • Application
• Material
• Process
Desire
Ga d
p State
Curre
nt
State
Gap Analysis
Defining Current State
Internal Measurements Strength Weakness
Gap Analysis
Benchmarking
• Bridging the gap
Policy of
Oil spill Leakage Gasket Sub
ordering to
on from damage standard
floor pump d gasket lowest
bidder
Cleaning Maintenanc
staff on e
leave overdue
Poor
Pump too old
Housekeepi
ng
Frequenc
y
10
0
1
2
3
4
5
6
7
8
9
Application opportunity
Concepts clear
Pareto Chart
Engaginginstructor
Complaint
Meeting expectatations
Instructor knoledgeable
Learning Valuable
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Fault Tree Analysis
OR AND
gate gate
Image
from
Cause
andEffect
Diagrams
Types of Wastes
Philosophy
• Waste exist in all processes at all
levels in the organization.
• Waste elimination is the key to
successful implementation of lean.
• Waste reduction is an effective way to
increase profitability.
Muda, Mura, Muri
Muda
an activity that is wasteful and
doesn't add value or is
unproductive
Mura
Any variation leading to
unbalanced situations.
Muri
Any activity asking
unreasonable stress or effort Wastes
from personnel, material or
equipment.
Muda
• Muda is a traditional Japanese term
for an activity that is wasteful and
doesn't add value or is unproductive
• Type I Muda: (Incidental Work)
• Non-value-added tasks which seam to
be essential. Business conditions
need to be changed to eliminate this
type of waste. Wastes
8 Types
of Wastes Defects
Motion
Unnecessary movement of Sorting, repetition or
people making scrap
or parts within a process.
8 Types
of Wastes
3. Motion
• Unnecessary movement of people or
parts within a process.
8 Types
of Wastes
4. Waiting time
• People or parts waiting for a work cycle
to finish.
8 Types
of Wastes
5. Over processing
• Processing beyond the demand from
the customers.
8 Types
of Wastes
6. Overproduction
• Producing too much, too early and/or
too fast.
8 Types
of Wastes
7. Defects
• Sorting, repetition or makingscrap
8 Types
of Wastes
8. Unexploited knowledge
• Failure when it comes to exploiting
the knowledge and talent of the
employees.
8 Types
of Wastes
Introduction Define Define: Identify the CTQs, Key Process Output
Variables (KPOVs) or Y for focus
to Improve
and Control Measure Measure: Collect Baseline data for Y, and also
understand what could be the Key Process Input
Variables (KPIVs) or X impacting Y
• Example: Let us assume our key output variable is Average Handle Time
(Y) during service calls, which impacts the financial fortunes of the
company.
Piloting • With analysis, we found that Hold Time is the key input variable
(X) for Average Handle Time. Controlling the Hold Time would
mean impacting Average Handle Time.
• The reasons could have been prioritized with the help of Pareto
Charts and, moving to the Improve Stage, we know what are the
main factors influencing changes in X.
• Example: Less training being provided to employee’s results in High Hold
Time.
• Design of Experiments can be used as a one-stop method for analyzing all influencing
factors, to arrive at a robust and a successful model.
• Designed experiments are preferred over OFAT (One Factor at a Time) experiments
because they don’t miss interactions (explained in this chapter).
• With techniques like Blocking, you can eliminate experimental mistakes. However, trials
should be randomized to avoid concluding Factor A is significant when time or sequence
may have influence over the response’s results.
• With techniques like Replication, you can test the variability given the same conditions
to improve the robustness of the model.
Basic Terms – 1
• Example: Determine factors related to consistent plastic part hardness, such as mold temperature
• Dependent Variables: Responses that vary as a result of changes made in the independent variable.
Example: plastic part’s hardness.
• Response: An outcome of an experimental treatment that varies as changes are made to levels
and factors.
• Independent Variables: Factors that are intentionally varied by the experimenter. In the above example,
mold temperature will be varied in the experiment.
• Factors: Factors (independent variables) are the items changed during an experiment in order
to see their impact on the output
• Factors may be quantitative or qualitative
• Levels: Levels are the values (or conditions) of the factors that are tested during the experiment
• most experiments test factors at 2 or 3 levels
•
•
Basic Terms – 2
Treatment: A certain combination of factor levels whose effect on the response
variable is of interest.
Example --- Output Hardness of the plastic compound (Hardness is the response)
Input Oven temperature and type of raw material (Temperature and type are
Factors)
Temperature can be varied at two levels: 700 Degrees and 900 Degrees. They are thus
quantitative factors. Raw material types are attributes, and used as plastic with filler and
plastic without fillers. They are qualitative factors.
Error: The variation in experimental units that have been exposed to the same
treatment is attributed to experimental error. This variability is due to the
uncontrollable factors.
Experimental unit: is the quantity of material (in manufacturing) or the number served
(in a service system) to which one trial of a single treatment is applied .
• Repetition: Running several samples during one experimental setup
without change in the setting (short-term variability)
3 • Example:
• Three (or three groups of) parts are manufactured during one
trial at 700 deg using plastic with fillers. This is repetition.
• After making parts using plastic without fillers, you come back
and make more parts at 700 deg using plastic with fillers. This is
replication.
• With combined analysis, experimental error can be determined
and will tell you if the differences in readings are statistically
significant or not.
DOE – A
Plastic
Molding
Example
A full factorial experiment is any experiment in which all possible combinations of factor
levels are tested.
This two-way heat treating experiment is a simple example of a full factorial design:
Note: This above simplified example is used to illustrate the concepts of main factor and
interaction effects.
Full
Factorial
Experiment
–Example
After conducting the experiment with two factors, two levels and two repetitions we get
the values as outlined in the boxes below for y1, y2 etc.
Main Effect
Interaction Effect
Results: The interaction plot shows that we should select low temperature and high
oven time to achieve the highest desired output of hardness. The parallel lines indicate
the output if no interactions occur between the main effects.
• Number of experiments in a DOE setting is known as Runs
Experimental
factors at all levels. you wish to save time. Response Surface for a very complex
Designs for a highly model.
optimized and a robust
model.
Method?
Design of Basics of Design of Experiments
Experiments
- Summary Factor, Level, and Response
# Pieces # Occurrences
# Defectives # Defects
n is subgroup size
Variable Measurements
Variable - Measurements
n>9 n =2 to 9 n=1
# Pieces # Occurrences
# Defectives # Defects
n is subgroup size
• Variables / Measurements
• I-MR or X-MR chart (Individual, Moving
Range)
• X bar - R chart (Average – Range)
Control Chart • X bar - s chart (Average – Standard
deviation)
Selection • Attributes / Counts
• np Chart (Number of defectives)
• p Chart (Proportion defectives)
• c Chart (Number of defects)
• u Chart (Number of defects per unit)
Data Type
# Pieces # Occurrences
# Defectives # Defects
n is subgroup size
np and p Chart
Variable Attribute
Total Defectives and Percent Defective
Binomial Distribution
Subgroup size is normally big compared to
variable charts
# Pieces
# Defectives
Constant Variable
np chart p chart
np Chart
Equal Subgroup size Control Limits
Control limits are straight lines
12 10.931
10
NP - Defectives
8
6 4.680
4
2
0.000
0
p Chart
Unequal Subgroup size Control Limits
Control limits change with the number of items in
the subgroup (subgroup size)
Larger Subgroup – narrow control limits
Smaller Subgroup – wider control limits
0.100
P - Unplanned Return
0.080
0.053
0.060
0.040
0.021
0.020
0.000
0.000
Attribute Measurements
Attribute - Counts
# Pieces # Occurrences
# Defectives # Defects
# Occurrences
# Defects
Constant Variable
c chart u chart
c Chart
Equal Subgroup size Control Limits
Average Defects
20.192
20
C - Defects
15
10.480
10
5
0.768
0
u Chart
Defects per Unit Unequal Subgroup size Control Limits
Control limits change with the number of items in
the subgroup (subgroup size)
0.200
0.150
U - Defects
0.105
0.100
0.050
0.004
0.000
Control ChartAnalysis
What is the problem with this process?
110.00
109.00
107.45
108.00
107.00
106.00
X-Bar: Shot 1 - Shot 3
105.00
104.00
103.00
102.00 100.91
101.00
100.00
99.00
98.00
97.00
96.00
94.36
95.00
94.00
Control ChartRules
Nelson Rules
Rule Pattern Probable Cause
1 1 point more than 3 Stdev from CL New person, wrong setup
2 7 points in a row on same side of CL Setup change, process change
3 7 points in a row all increasing or all decreasing Trend, Tool wear
4 14 points in a row alternating up and down Over control, tempering
5 2 out of 3 points more than 2 Stdev from CL (same side) New person, wrong setup
6 4 out of 5 points more than 1 Stdev from CL (same side) Small shift similar to Rule 1, 5
7 14 points in a row within 1 Stdev from CL (either side) Process change
8 8 points in a row more than 1 Stdev from CL (either side) Process change
Rule Pattern
A
107.00
106.00 5
X-Bar: Shot 1 - Shot 3
105.00 5
B
104.00
103.00
C
102.00
101.00
100.91
C
100.00 2
99.00
6 2
B
98.00
97.00
A
95.00
3
94.36
96.00
94.00
Rule Pattern
105.00 5
104.00
103.00
102.00 100.91
101.00
100.00 2
99.00
6 2
98.00
97.00
96.00 3
95.00 94.36
94.00
Rule Pattern
105.00 5
104.00
103.00
102.00 100.91
101.00
100.00 2
99.00
6 2
98.00
97.00
96.00 3
95.00 94.36
94.00
Rule Pattern
104.00
103.00 4
102.00
100.65
101.00
100.00
99.00
98.00
97.00
96.00
95.00 93.98
94.00
93.00
Rule Pattern
105.00 5
104.00
103.00
102.00 100.91
101.00
100.00 2
99.00
6 2
98.00
97.00
96.00 3
95.00 94.36
94.00
Rule Pattern
105.00 5
104.00
103.00
102.00 100.91
101.00
100.00 2
99.00
6 2
98.00
97.00
96.00 3
95.00 94.36
94.00
Rule Pattern
103.00
102.00 7
100.45
101.00
7
100.00
99.00
98.00
97.00
96.00
95.00 94.11
94.00
93.00
Rule Pattern
105.00 8
104.00
103.00
101.71
102.00
101.00
100.00
99.00
98.00
97.00 95.98
96.00
95.00
Rule Pattern
Probability of Rule 7
(0.68)14 = 0.0045
Probability of Rule 8
(1-0.68)8 = 0.0001
Total • OEE shows the efficiency of your production
system
Productive • Overall equipment effectiveness
Maintenance • (OEE) = Availability x Performance x Quality
Operator
Operator
I Use
TPM
Maintenance
Maintenance
We Maintain
I Maintain
Total Productive Maintenance (TPM)
5S
in the production processes.
Controls -
Types Instructional – How the task to be
performed?
Quality Manual
Procedures
Individual Value
504.188 Run B 500 4.1884 0
_
X=-0.15
506.879 Run B 500 6.8792
-10
506.189 Run B 500 6.1887
517.210 Run B 500 17.2103 -20 LCL=-21.45
479.511 Run B 500 -20.4891 1 3 5 7 9 11 13 15 17 19 21 23
495.170 Run B 500 -4.8304 Observation
ang
e
392.790 Run C 400 -7.2102 20
Mov ing
R
399.538 Run C 400 -0.4623 10 __
MR=8.01
392.192 Run C 400 -7.8076
403.812 Run C 400 3.8119 0 LCL=0