Six Sigma Overview

Objectives
To know what “Six Sigma” means
To be able to explain the meaning of Six Sigma as
a measure of a product’s ability to meet customer
requirements
To be able to explain the meaning of Six Sigma as
a change initiative in a business

1
 Comprehend Business Objectives/Priorities.
Objectives for this session
To gain a high-level understanding of Six Sigma and the
EMC approach to Six Sigma
To walk out with an understanding of what is required to write
a good project definition to assist your candidates to have a
clearly defined project before attending Week 1 training.
To ensure a common understanding of basic tools and
concepts used in Six Sigma training
To clarify what is expected from you as champions and
management to foster this cultural change in using these
tools in our day to day business.

2
 Does Your Company Need Six Sigma?

Does Your Company

Believe zero-defect goals are neither realistic nor achievable?
Have 10 times the number of suppliers required to run the
business?
Have 5 to 10% of its clients dissatisfied with the product, the
sales organization, or the service you’ve provided?
Have customers who will not recommend that others purchase
your goods or services?
Quantify profitability and growth?
Deliver new products to the market?

(continued)
3
 Does Your Company Need Six Sigma?
-cont.
Does Your Company -
Continually implement price reductions for current
products?
Have an increasing number of competitors?
Spend a high % of sales dollars on repairing or
reworking a product before it ships?

4
 Does Your Company Need Six Sigma?
-cont.
Does Your Company -

Have a magician in your

organization?

5
 Six Sigma Objectives

The Vision: Drive industries to design and produce

products/services to Six Sigma standards
The Goal: Produce goods and services at a Six Sigma level.
As your organization moves toward Six Sigma quality, you will:
Eliminate defects
Reduce production and development costs
Reduce cycle times and inventory levels
Increase profit margin
Improve customer satisfaction
The Strategy: Use a data-driven structured approach to attack
defects to improve the sigma level of your goods and services

6
 What Is Six Sigma?

Please grade your organization based on the following:

F Our organization uses only tribal knowledge.
We do not use data.
E Our organization collects data simply to say, “We collect data.”
D Our organization collects data and we sometimes look at the
numbers.
C Our organization logically groups the data. We form charts.
B Our organization uses sample data along with basic statistics.
A Our organization uses sample data along with inferential
statistics.
A+ Our organization quantifies processes via prediction equations.

Our Grade: _____________

7
 What Is Six Sigma?

A Vision of a Six Sigma Company

Organizational Issue Traditional Approach Six Sigma Approach
Problem resolution Fixing (symptoms) Preventing (causes)

Behavior Reactive Proactive

Decision making Experience-based Data-based

Process adjustment Tweaking Controlling
Supplier selection Cost (piece price) Capability
Planning Short-term Long-term

Design Performance Producibility

Employee training If time permits Mandated

Chain-of-command Hierarchy Empowered teams

Direction Seat-of-pants Benchmarking and metrics

Manpower Cost Asset
8
 What Is Six Sigma?

Sigma level: The business metric

used to indicate the performance
of a process to some specification
σ1

The number of standard

deviations that fit between the
mean and the nearest σ2
specification limit LSL USL

OR
A measure of the number of σ1
defects per opportunity produced
by a process
9
 What Is Six Sigma?

Is 99% yield good enough?

Five lost e-mail messages per month
No cable television for 3.5 hours each month
15,000 overnight carrier packages lost per week
25 incorrect car rental reservations per company
per day

Today’s Standard Automotive Standard Six Sigma Standard

3σ Capability 4σ Capability 6σ Capability

93.319 % 99.379 % 99.99966 %

Long-Term Yield
10
 Sigma level compared to defects
Short-term distribution shifted by 1.5
σ to obtain long-term PPM
Note: Industry standard has
defined a sigma level to imply
short term.
Sigma Short-Term Long-Term
Level PPM PPM
Defects per Million Opportunities

800000

700000
1 158655.3 691462.5
2 22750.1 308537.5
600000
3 1350.0 66807.2
500000
4 31.7 6209.7
400000
5 0.3 232.7
300000 6 0.0 3.4
200000
Short-Term PPM Long-Term PPM
100000

0
1 2 3 4 5 6
Short-Term PPM 158655.3 22750.1 1350.0 31.7 0.3 0.0
Long-Term PPM 691462.5 308537.5 66807.2 6209.7 232.7 3.4
Sigma Level

11
 Benchmarking Standards

Process Capability (Sigma Level) (Distribution

Shifted ± 1.5σ)

±3 Sigma ±4 Sigma ±5 Sigma ±6 Sigma

Rolled Throughput
1 93.319% 99.379% 99.977% 100.000%
Product Complexity
(# of Opportunities)

Yield (RTY)
10 50.086% 93.961% 99.768% 99.997%
80 0.396% 60.755% 98.156% 99.973%
100 0.099% 53.638% 97.700% 99.966%
150 0.003% 39.284% 96.570% 99.949%
300 0.000% 15.432% 93.257% 99.898%
1,200 0.000% 0.057% 75.636% 99.593%
3,000 0.000% 0.000% 49.753% 98.985%
150,000 0.000% 0.000% 0.000% 60.042%

4σ toothpick manufacturer (assume one opportunity for a defect): Has an RTY of 0.99379 (1) = 99.379%
4σ mechanical pencil manufacturer (assume 10 opportunities for a defect): Has an RTY of 0.99379 (10) = 93.961%

12
 Six Sigma Overview

What Is Cost of Poor Quality (COPQ)?

The cost of finding and fixing defects
Failing to meet customer expectations the first time
A missed opportunity for increased efficiency
The potential for higher profits
Loss in market share
Increase in cycle time
Labor associated with ordering replacement
material
Costs associated with disposing of defects
13
 Six Sigma Overview

Traditional Metrics

Equipment Util. vs. COPQ Test Yield vs. COPQ

$130,000 $130,000
$120,000 $120,000
$110,000 $110,000
COPQ

COPQ
$100,000 $100,000
$90,000 $90,000
$80,000 $80,000
$70,000 $70,000

$60,000 $60,000
0.825 0.85 0.875 0.9 0.925 0.95 0.975 1 80.0% 82.5% 85.0% 87.5% 90.0% 92.5%
Equipment Utilization Test Yield

14
 Six Sigma Overview

Six Sigma Metrics

Sigma Level vs. COPQ Cycle Time vs. COPQ

$450,000 $450,000

$375,000 $375,000

$300,000 $300,000

COPQ
COPQ

$225,000 $225,000

$150,000 $150,000

$75,000 $75,000

$- $-
1.00 1.75 2.50 3.25 4.00 4.75 5.50 0.00 30.00 60.00 90.00 120.00

Sigma Level Cycle Time (Minutes)

15
 Six Sigma Overview

For an average company,

the COPQ can be as high as

25%
of total sales!

16
17
 Six Sigma Overview

What Is Cost of Poor Quality (COPQ)?

The cost of finding and fixing defects
Failing to meet customer expectations the first time
A missed opportunity for increased efficiency
The potential for higher profits
Loss in market share
Increase in cycle time
Labor associated with ordering replacement
material
Costs associated with disposing of defects
18
19
 Six Sigma Overview

Indicators of COPQ

Low Yield Rate Process Downtime

High Customer Failure Rate High Operating Costs
(PPM) High Scrap/Rework Costs
Incoming Product Quality High Inventories (WIP)
Problems Long Cycle Times
Unpredictable Quality Unpredictable Product
Poor Process Capability (Cp, Cpk) Performance
Measured System Error Capacity Constrained
High Past Due to Customer High Product Volume
High Maintenance Costs Internal Perceived Poor Quality
Low Machine Utilization External Perceived Poor Quality

20
 Six Sigma Overview

Indicators of COPQ

Low Yield Rate Process Downtime

High Customer Failure Rate High Operating Costs
(PPM) High Scrap/Rework Costs
Incoming Product Quality High Inventories (WIP)
Problems Long Cycle Times
Unpredictable Quality Unpredictable Product
Poor Process Capability (Cp, Cpk) Performance
Measured System Error Capacity Constrained
High Past Due to Customer High Product Volume
High Maintenance Costs Internal Perceived Poor Quality
Low Machine Utilization External Perceived Poor
Quality

21
 EMC Six Sigma Approach

Objectives
To know the EMC approach to deploying Six Sigma
To know the components and importance of the
implementation strategy
To know the components and importance of the
application strategy

22
 EMC Six Sigma Approach

Implementation
Application
Strategy
Strategy
Strategic
Infrastructure 6σ Measure
Tactical Analyze
Infrastructure Improve
Operational Control
Infrastructure

23
 EMC Implementation Strategy

Traditional “Quality” program implementation

Need for quality improvement is recognized.
Top management “buys in” to quality improvement.
Implementation responsibility is turned over to the
Quality VP or leader.
The CEO makes a statement of support and
expectations.
All employees are trained in basic quality tools.
Improvement is expected.

24
 EMC Implementation Strategy
STRATEGIC LEVEL
Organizational Leadership Six Sigma Deployment Plan
Executives Executive Steering Committee
Master Black Belts

TACTICAL LEVEL
Tactical Management Six Sigma Project Teams
Operational Managers Champions
Support Managers Black Belts
Team Members
Stakeholders

OPERATIONAL LEVEL
Operational Work Six Sigma Institutionalization
Operators Green Belts
Support Staff Yellow Belts

Complements The Established Infrastructure;

It Does Not Replace It!
25
 EMC Implementation Strategy
Strategic Level
Steering committee
Defines Six Sigma Initiative objectives
Business analysis based on strategic objectives
Analysis/revision of existing business performance metrics
Establishes targets for deployment throughout the organization
Identifies and defines initial Six Sigma application
projects
Assigns Champions
Assigns Black Belts/Green Belts
Reviews Six Sigma projects
Reviews and revises strategic objectives and business
performance metrics

Champion Training Aligns Strategic Decisions

With Six Sigma Methodology.
26
 EMC Implementation Strategy
Champion Roles and Responsibilities
Roles:
Is responsible for coordination of the business roadmap to achieve 6σ
Selects projects, controls execution, and alleviates roadblocks for the 6σ
projects in his area of responsibility
Reporting Lines:
Is part of the functional organization and reports directly to the 6σ leader
Is a member of the organization’s 6σ leadership team
Responsibilities:
Selects projects, controls execution, and implements and realizes gains
(bottom line linkage)
Owns the “execution” portion of Black Belt/Green Belt certification
Obtains the needed project resources and eliminates roadblocks
Drives the cross-functional coordination of projects
Participates in all project reviews
Owns the Black Belt/Green Belt selection
Is the boss of the Black Belt/Green Belt (direct or dotted line)
Provides reward and recognition
Time Commitment:
Two days/week per 10 projects managed (20% to 80% based on
organization)

27
 EMC Implementation Strategy
Master Black Belt Roles and Responsibilities
Roles:
Is a mentor, trainer, and coach of Black Belts and others in the
organization
Brings broad organization up to the required 6σ competency level.
Reporting Lines:
Is generally a central resource; is cross functional.
Is a member of the 6σ leadership team and the steering committee
Responsibilities:
Mentors and coaches Black Belts
Develops and conducts several forms of training
Owns the 6σ technical development roadmap
Provides higher education of Black belts and Master Black Belts
Brings the entire organization to the 6σ level of competency
Is the custodian of the purity of the method  no compromising
Transfers lessons learned
Owns “knowledge” certification of Black Belts
Finds outside expertise/help when required
Networks with other 6σ organizations
Time Commitment:
Must be 100% dedicated

28
 EMC Implementation Strategy
Team Member Roles and Responsibilities
Roles:
Participates on the project teams and supports the goals of the project,
typically in the context of his existing responsibilities
Responsibilities:
Performs his normal job and supports the activity of the project as it relates
to that particular job
Learns the 6σ methodology as it applies to the particular project
Continues to learn and practice the 6σ methodology and tools after project
completion. Some may evolve to the Black Belt level of knowledge and
practice.
Time Commitment:
Is defined by the Six Sigma Belt, Champion, and Functional Manager
May be instructed to support any Six Sigma project as a high priority
Minimum Training Requirement:
Four-hour overview
Additional training from a Six Sigma Belt

29
 EMC Implementation Strategy
Stakeholder Roles and Responsibilities
Roles:
Agrees with the project selection and expected results
Supports the Black Belt/Green Belt in the organization and takes the
necessary actions to realize the gains
Responsibilities:
Approves the potential savings from the project (pre R0)
Participates in the project identification and selection process
Owns the development and implementation of actions to realize the
savings
Provides team resources
Ensures that 6σ training is implemented in the organization
Participates in 6σ reviews
Time Commitment:
As needed
Min. Training Requirement:
Four-hour overview

30
 EMC Six Sigma Approach

Implementation
Strategy Application
Strategic Strategy
Infrastructure 6σ Measure
Tactical
Infrastructure Analyze
Operational Improve
Infrastructure Control

31
 EMC Application Strategy

What Tool(s) Do You Need for Your Project?

6σ Tools:
Design for manufacturability
Design for Six Sigma, 6σ tolerancing, product scorecard
Tools:
5σ Process characterization (mapping, MSA, etc…)
Process optimization (DOE, etc…)
Tools:
Seven basic tools
4σ (paretos, fishbones, check sheets, histogram, flowcharts,
brainstorming, control charts)
Tools:
Common sense
3σ Tribal knowledge

1 to 2σ
32
 EMC Application Strategy

Black Belt/Green Belt Certification

Attend all weeks of classroom training
Complete a Six Sigma project successfully
Demonstrate knowledge and apply tools
Achieve the project objectives
Shows financially measurable impact
Complete the measure phase of a second project
(Black Belts only)

33
 EMC Application Strategy
Define project scope
Validate measurement systems
Measure Establish initial capability for Ys
Process exploration of all potential Xs

Characterize the response and analyze the raw data

Bimodal? Skewed? Is the problem with m or s2?
Analyze Use graphical analysis, multi-vari, hypothesis testing,
and basic statistical tools to identify the likely families
of variability

Identify the likely Xs

Use the design of experiments to find
Improve the critical few Xs
Move the distribution (shift m)
Shrink the spread (decrease s2)
Confirm the results
Mistake-proof the process
Control Tolerance the process
Measure the final capability
Place appropriate process controls on the critical Xs
Document the effort and results
34
 EMC Application Strategy

Project Reviews
Each phase has a list of potential project application
tools. Use these tools to move your project forward.
If the tool is not appropriate, please make sure you clearly
understand and can demonstrate its usage.
We are here to help you with both the application and the
underlying concept.
Local and Corporate Project Reviews
Present only the tools that are pertinent or of interest to the
audience during your reports.
Your presentation should take no longer than 15 minutes.
Please allow 10 minutes for the presentation itself and five minutes
for a brief question/answer session.
Your presentation should consist of approximately 8 to 10 slides. 35
 EMC Application Strategy
Phase 1: Product Measurement

Part

YTP

X1, … , XN Operation Verify Y 1, … , Y N

Process Capability Analysis for C1 Gage name:

Date of study:
Gage R&R (ANOVA) for Measure Reported by:
Tolerance:
Misc:
Lower Spec Upper Spec
Xbar Chart by Operator Operator*Part Interaction
Operator

S am ple Mean
1.1 1 2 3 1.1
1.0 1.0 1

Av erage
0.9 3.0SL=0.8796 0.9 2
0.8 X=0.8075 0.8 3
0.7 -3.0SL=0.7354
0.6 0.7
0.5 0.6
0.4 0.5
0.3 0.4
0 Part ID 1 2 3 4 5 6 7 8 9 10

R Chart by Operator By Operator

Sam ple Range

0.15 1 2 3 1.1
3.0SL=0.1252 1.0
0.10 0.9
0.8
0.7
0.05 R=0.03833 0.6
0.5
0.00 -3.0SL=0.000 0.4
7 8 9 10 11 12 13 14 0 Oper ID 1 2 3

Components of Variation By Part

Short-Term Capability 100 1.1
90 %Total Var 1.0
80
Perc ent

Cp 0.47 Targ * Mean 9.8997 %>USL Exp 15.11 PPM>USL Exp 151108 70 %Study Var 0.9
60 0.8
CPU 0.34 USL 11.000 Mean+3s 13.0992 Obs 14.00 Obs 140000 50 0.7
40
CPL 0.59 LSL 8.000 Mean-3s 6.7001 %<LSL Exp 3.74 PPM<LSL Exp 37442 30 0.6
20 0.5
Cpk 0.34 k 0.266 s 1.0665 Obs 2.00 Obs 20000 10 0.4
0
Cpm * n 100.000 Gage R&R Repeat Reprod Part-to-Part Part ID 1 2 3 4 5 6 7 8 9 10

36
 Session 1, Application Tools

Regardless of the tools used, you should address

Project Selection  each of these questions during your Session 1
*Problem statement and project status Review:
*Project assessment chart (Metric.doc) 1) What is your practical problem statement 
Process Exploration what are you trying to fix or avoid?
Process flow diagram 2) What is the business impact?
XY matrix, PFMEA, fishbones 3) Who is the customer (internal/external)?
Data collection system 4) What are the Ys?
Measurement System(s) Analysis (MSA) • How did you determine them?
Attribute/Variable gauge studies • How are you measuring them?
Capability Assessment (on each Y) • How good is the measurement system?
Capability (Cpk, σ level, DPU, RTY) • Have you done a Gage R&R?
Graphical tools (histograms, µ, σ) 5) Does this project have applications in other
areas? (tree of opportunity)
Progress Summary
*Conclusion(s) 6) Is this a technology or control problem?
*Issues and barriers 7) Did you develop a process flow chart?
*Next steps 8) Do you have adequate resources to
Completed “Local Project Review” complete the project?

* Note: Required reports 37

 EMC Application Strategy
Phase 2: Product Performance Analysis
Capability Analysis
Report 2: Process Capability for C5

Xbar and S Chart Capability Indices

600
UCL=557.2 ST LT
550
500 X=498.7 Mean 500.000 498.683

450 StDev 43.444 63.802

LCL=440.2
Z.USL 3.453 2.372
400
Z.LSL 3.453 2.330
Subgroup0 5 10 15 20
Z.Bench 3.261 2.080
90 UCL=85.58
80 Z.Shift 1.181 1.181
70
60 P.USL 0.0003 0.0089
50
40 S=40.96
30 P.LSL 0.0003 0.0099
20
10 P.Total 0.0006 0.0187
0 LCL=0.000
Yield 99.9445 98.1253
PPM 555 18747

Potential (ST) Capability Actual (LT) Capability Cp 1.15

Process Tolerance Process Tolerance Cpk 1.14

369.259 630.741 306.793 690.572 Pp 0.78

I I I I I I Ppk 0.77
I I I I I I
350 650 350 650 Data Source:
Specifications Specifications Time Span:
Data Trace:

Box Plot Analysis

0.06
Xs X2
0.05
b le
s si
0.04
o X1 X5
Impurity

0.03
l P X4
0.02 Al Xs
0.01
X3 X6 b le
0.00
oba X1
1 2 Pr X5
Piece

38
 Session 2, Application Tools

Project Status
*Problem statement and project status
Regardless of the tools used, you should address
*Project assessment chart (Metric.doc)
each of these questions during your Session 2
Process Capability Analysis
Review:
Distribution assessment
1) What is the statement of the statistical
Data transformations problem?
Capability analysis
2) Is the response discrete or continuous?
Graphical Analysis (X Search) • What does the distribution look like?
Boxplots/scatterplots/other graphs • Has this helped you reduce the potential Xs?
Identification of high-priority Xs 3) How much of the problem, as described in the
Capability (Cpk, s level, DPU, RTY) measure phase, are you going after?
Graphical tools (histograms, m, s) 4) Have you reduced the likely Xs to a number
Progress Summary that can be experimented with?
*Conclusion(s) 5) What are your next steps?
*Issues and barriers 6) Do you have adequate resources to complete
*Next steps the project?
Completed “Local Project Review”

*Note: Required reports 39

 EMC Application Strategy
Phase 3: Performance Improvement
Design of Experiments
Run Factor A Factor B Factor C Response 1
Main Effects for Response No. Feed Rate Direction Location ∆ Location
Centerpoint
1 .015IPR RIGHT HEADSTOCK
-1 1 -1 1 2 RAPID RIGHT HEADSTOCK
230 3 .015IPR LEFT HEADSTOCK
4 RAPID LEFT HEADSTOCK
215
5 .015IPR RIGHT TAILSTOCK
Response

200
6 RAPID RIGHT TAILSTOCK
7 .015IPR LEFT TAILSTOCK
185 8 RAPID LEFT TAILSTOCK

170

A B Interation Plot for Response

A
235 -1
225 1
215 -1
240 1
205
Centerpoint

Mean
220 195
RESPONSE

185
200
175
2
180 165
160 155
4
E

145
or

140
ct

22 -1 1
Fa

18 6
Factor 14 B
G

40
 Session 3, Application Tools

Regardless of the tools used, you

Project Selection See ProjPlan.xls
should address each of these
Problem statement & project status
questions during your Session 3
Project assessment chart (Metric.doc)
Review:
Remaining Session 1 and 2 1) Were all the potential Xs measurable
Deliverables and controllable for an experiment?
Design of Experiments 2) Are the vital few Xs statistically
significant?
DOE planning sheet 3) Are the effects of practical
DOE factorial experiments significance?
Y = F (X1, X2, X3, …) 4) How much of the problem have you
explained with these Xs?
Updated PFMEA 5) How much unexplained error exists?
Progress Summary 6) Are any new improvements
*Conclusion(s)
transferable across the business?
7) Is an action plan for spreading the
*Issues and barriers best practice in place and
*Next steps appropriate?
Completed “Local Project 8) Do you have adequate resources to
complete the project?
Review”
9) What are the next steps?

* Note: Required reports

41
 EMC Application Strategy
Phase 4: Process Control
Eliminate Xs 10

Cause System 9

Cause Effect Automate Xs 6s Spec 8

Range of Y 7
Control Xs 6

σ =
∑
N
i =1
(X i − μ) 2
N High Spec Low Spec
MHz
-1 1

Statistics Standards
Realistic Tolerance on X

EWMA Chart for Length

E W M A C h a r t fo r L e n g th
6 0 1.5
U C L = 6 0 1.2
U C L = 6 0 1.2

6 0 0 .5
EWMA

X = 6 0 0 .1
X = 6 0 0 .1

5 9 9 .5

L C L = 5 9 8 .9
L C L = 5 9 8 .9
5 9 8 .5

0 50 10 0

Sample Number

42
 EMC Application Strategy

Final project update: Before BB/GB can be

certified, they are required to submit in compressed
electronic format:
A final report
A copy of all presentations
All application tools used throughout the project
Supporting data files used throughout the project
Complete a local management review at their
facility before the corporate review.

43
 Basic Concepts

Objectives
To understand the fundamental equation that
drives Six Sigma
To know how the difference between First Time
Yield (FTY) and Rolled Throughput Yield (RTY) in
measuring process performance
To know the DPU Concept and how is related to
RTY.

44
 Fundamental Equation

The Fundamental Equation That Drives Six Sigma

Y = f(x)
The Output Is A Function Of The Inputs And The Process.

45
 Fundamental Equation

The Fundamental Equation

Y= f(x)
What is Y? What are X . . . X ?
1 N

Output ■ Input process

variables
Dependent
■ Independent
Effect ■ Cause
Symptom ■ Problem
Monitor ■ Control
46
 Fundamental Equation

Output Variation Y = f(x)

Sources of Variability

Incoming Parts Process

Design and Materials Capability

Process Process Process

characterization and characterization and characterization
optimization optimization and optimization
Design for Logic and intuition Logic and intuition
manufacturability
Seven basic tools Seven basic tools
Design for Six Sigma (paretos, fishbones, (paretos,
maps, etc.) fishbones, maps,
etc.)
47
 Six Sigma Metrics
Individual Product

Z-score or Rolled
Parts per
“sigma throughput
million,
value” yield,
ppm
RTY

Process
capability
Cp,Cpk,
Pp,Ppk
Defects per
Defects per million
Yield unit, opportunities
DPU ,
DPMO

48
 Understanding Yield

First Time Yield (FTY) is a common output metric

(Y metric) used to identify and target problem areas.
First Time (End of Line) Yield by Week

FTY = P
* 100%
100 U
Where:
98
FTY = First Time Yield
Weekly Yield (%)

(test yield)
96
P= Number of units
that pass the test
94
U= Number of units
tested
92

90
Wk 7
Wk 1

Wk 4

Wk 6

Wk 9
Wk 2
Wk 3

Wk 5

Wk 8

Wk 10

Wk 12
Wk 13

Wk 15
Wk 11

Wk 14

49
 Understanding Yield

FTY is not a good predictor for improving profits

and/or decreasing scrap.
Expected Relationships
9
25
8
20 7
6

Scrap
Profit

15 5
4
10
3
2
5
1
0 0
80 90 100 80 90 100
Test Yield Test Yield

50
 Understanding Yield

“Hidden” costs in the “Real Factory”

More manpower
Extra floor space
Longer cycle time _______
Good Units
More raw material Total units tested
More $$$$

FTY

Test

Operation 1 Test Operation 2 Test Product

51
 Understanding Yield

“Hidden” costs in the “Real Factory”

More manpower
Extra floor space
Longer cycle time Hidden _______
Good Units
More raw material Factory Total units tested
More $$$$
Re-Work Re-Work
or Scrap or Scrap
FTY

Failure Failure
Analysis Analysis
Test

Operation 1 Test Operation 2 Test Product

52
 Defects vs. Defectives

Defects:
Countable failures associated with a single unit.
A single unit can be found to be defective, but it may have
more than one defect.
Defectives:
Completed units that are classified as bad. The whole unit is
said to be defective regardless of the number of defects it has.

First time yield =

non-defectives / total units.

53
 Defects per Unit (DPU)

Unit: The entity that is transformed by value-added

activities. Typically, it is defined as the “product”
that is sold to the customer.

Defects
DPU =
Units Produced

DPU can be applied at both the individual process-

step level and the product level.

54
 Example of DPU
At the part level:
Pedal assemblies arrive at our plant weekly to support
production needs. The following defect data is collected on a
sample basis. It was collected over the previous 12- month
period for 500 total samples (n = 500).
Reflector missing 25
Threads marred 15
Pedal bent 10
Total 50

The average number of defects per unit (pedal) is:

Total Defects 50
DPU = = = 0.1
Units 500 (continued)
55
 Example of DPU
-cont.
At the product level:
DPUs of sub-assemblies can be summed to obtain the total
number of defects found in the finished unit.
n
DPUtotal = ∑ DPUsub −assembly ( i )
i =1

Here are defect rates for four sub-assemblies that make up

the final product:
A 0.10 DPU
B 0.15 DPU
C 0.05 DPU
D 0.10 DPU
Product 0.40 DPU

56
 Rolled Throughput Yield (RTY)

RTY: Measuring the probability of obtaining

a defect-free unit, instead of first time yield.
Two calculations:
RTY = Yield process a × Yield process b × ⋅ ⋅ ⋅ × Yield process n
or converting from DPU
RTY = e −DPU

The second method is derived from a Poisson

distribution model. It is a valid approximation for
defect rates of 10% or lower. It shows there is a
relationship between DPU and RTY.

57
 Rolled Throughput Yield Example

The pedal sub-assembly process is outlined below.

We have identified the DPU associated with each
process step.
Select R Hand- Hand- Torque L Torque R
and L part tighten L tighten R pedal to 5 ft pedal to 5 ft
#s from bin pedal pedal lbs lbs
location
DPU = 0.02 DPU = 0.01 DPU = 0.01 DPU = 0.03 DPU = 0.03

Process Step
Select Part
DPU Step Yield
0.02 98%
RTY = e − dpu = e −0.1 = 90% ***
Hand Tighten L 0.01 99% or
Hand Tighten R 0.01 99%
Torque L 0.03 97% RTY = 0.98 × 0.99 × 0.99 × 0.97 × 0.97
Torque R 0.03 97%
Total 0.10 = 90%
*** Good for DPU < 0.1 , otherwise use product of step yields
58
 Understanding First Time Yield and RTY

Rolled Throughput Yield (RTY) is a measure that is

based on defects that occur throughout the
process.
Process RTY: Yield prior to
Capability inspection and test Outgoing
Quality

Operation: Verify: Good FTY: Yield after

RTY = 37% 9/10 Good FTY = 90%
inspection and test
DPU = 1
based on based on non-defectives
defects
Re-Work Bad Probability of a defect free unit (RTY)
Defects as a function of DPU’s.
DPU Probability %
1 37
Scrap Defects .5 60
.1 90
.01 99
59
 Defects per Million Opportunities
(DPMO)
It is sometimes helpful to compare products, parts, and
processes of differing complexities. We would expect
products/processes of higher complexity to have a lower
RTY (and “sigma value”).
RTY = Yield process a × Yield process b × ⋅ ⋅ ⋅ × Yield process n
n

−DPUtotal
− ∑ DPUprocess ( i )
RTY = e =e i =1

The higher n (the complexity) is, the lower the RTY.

DPMO is for Benchmarking and project selection and DPU
for Six Sigma Project Metric.

60
 The Calculation

To compare products/processes of differing complexity, we

must start with a measure of complexity. We call this
measure an “opportunity.”
Opportunities can be defined as the number of parts in the
product, or the number of process steps needed to assemble
a product. Both would be indicators of overall
product/process complexity.
Other definitions of “opportunity” can be applied as long as
the “measure” of complexity is used uniformly.

Total Defects
DPMO = × 1,000,000
Total Opportunities

61
 Understanding Opportunities

Which product is performing better, the pencil or the

blender?

Pentel Pencil Blender

Defects per Unit (DPU) 0.030 0.189

Rolled Throughput
Yield (RTY)
97.04 % 82.78 %

62
 Understanding Opportunities

Which product is performing better, the pencil or the

blender?
Pentel Pencil Blender
RTY = e - dpu
Defects per Unit (DPU) 0.030 0.189 ln(RTY) = - dpu
97.04 % dpu = - ln(RTY)
Rolled Throughput
Yield (RTY)
82.78 %
Opportunities 15 97
DPU/Opp 0.00200 0.00195
DPMO 2000 1948
Product Sigma Level 4.405 4.405
Note: Assumes the defects were collected in the long term.
63
 Measurement Systems Analysis (MSA)
Attribute Gage Repeatability and Reproducibility
(Attribute Gage R&R)

Task: You have 60 seconds to document the

number of times the 6th letter of the alphabet
appears in the following text.

64
 Inspection Exercise

Task: You have 60 seconds to document the number of times

the 6th letter of the alphabet appears in the following text.

The necessity of training farmhands for first-class

farms in the fatherly handling of farm livestock is
foremost in the eyes of farm owners. Since the
forefathers of the farm owners trained the farmhands
for first-class farms in the fatherly handling of farm
livestock, the farm owners feel they should carry on
with the family tradition of training farmhands of first
class farmers in the fatherly handling of farm
livestock because they believe it is the basis of good
fundamental farm management.
65
 Attribute R&R

An attribute R&R is used to:

Determine if operators across all shifts, all machines, etc., use
the same criteria to determine “good” from “bad”
Assess your inspection or workmanship standards against your
customer’s requirements
Identify how well these operators are conforming to themselves
Identify how well these operators are conforming to a “known
master,” which includes:
How often operators decide to ship truly defective product
How often operators do not ship truly acceptable product
Discover areas where:
Training is needed
Procedures are lacking
Standards are not defined
66
 Increasing Yield through Inspection

DPUin Inspect DPUescaping

DPU removed
DPUremoved Efficiency =
DPU in
Starting Defects per Unit, DPU
5 2 1 0.5
99% 3 3 3 3
Inspection
Efficiency

95% 5 4 4 4
90% 6 6 5 5
85% 7 7 7 6
80% 9 8 8 7
70% 12 11 10 10
50% 20 19 18 17
Number of consecutive inspectors required to achieve an
escaping defect rate of 3.4 ppm (6σ, short term)
67
 How to Run an Attribute R&R

Step 1: Select about 30 parts from the process.

50% of the parts in your study should have
defects.
50% of the parts should be defect free.
If possible, select borderline (or marginal) good
and bad samples.
Step 2: Identify the operators who should be
qualified.
Step 3: Have each operator independently and in
random order assess these parts and
determine whether or not they pass or fail
(continued)
(judgment of good or bad). 68
 How to Run an Attribute R&R
-cont.
Step 4: Repeat Step 3 for a second trial.
Step 5: Use the AttrR&R2.xls spreadsheet to report
the effectiveness and efficiency of the
attribute measurement system (operators and
the inspection process).
Step 6: Document and implement appropriate actions
to improve the inspection process (if
necessary).
Step 7: Re-run the study to verify the improvement.

69
 Example

Attribute Gage R&R Effectiveness

Blank Form: AttrR&R2.xls
Enter Pass/Fail,
0/1, etc. SCORING REPORT Title
5
Attribute Legend (used in computations)
DATE: Today's Date
NAME: Green Belt
Block
1 pass PRODUCT: ABC 123
2 fail BUSINESS:Division A

Known
Known Population Operator #1 Operator #2 Operator #3
Workmanship
Sample # Attribute
or Try #1 Try #2 Try #1 Try #2 Try #1 Try #2
Customer
1 Result
pass pass pass pass pass fail fail
2 pass pass pass pass pass fail fail
3 fail fail fail fail pass fail Operatorfail
4 fail fail fail fail fail fail fail
5 fail fail fail pass fail fail
Resultsfail
6 pass pass pass pass pass pass pass
7 pass fail fail fail fail fail fail
8 pass pass pass pass pass pass pass
9 fail pass pass pass pass pass pass
10 fail pass pass fail fail fail fail
Data Entry 70
 Spreadsheet Results with
Calculated Confidence Intervals
Statistical Report - Attribute Gage R&R Study

DATE: Today's Date % of time each

% of time trial 1 NAME: Black Belt
operator agrees
agrees with trial 2 PRODUCT: ABC 123
BUSINESS: Division A with the standard
for each operator
% Appraiser %Score vs Attribute
Source Operator #1 Operator #2 Operator #3 Operator #1 Operator #2 Operator #3
Total Inspected 10 10 10 10 10 10
# Matched 10 8 10 7 6 6
False Negative (operator rejected good product) 1 1 3
False Positive (operator accepted bad product) 2 1 1
Mixed 0 2 0
95% UCL 100.0% 97.5% 100.0% 93.3% 87.8% 87.8%
Calculated Score 100.0% 80.0% 100.0% 70.0% 60.0% 60.0%
95% LCL 69.2% 44.4% 69.2% 34.8% 26.2% 26.2%
Statistical Report

71
 Summary

An attribute Gage R&R must be performed to

ensure the integrity of attribute data.

Operators must inspect both known “good”,

“borderline” and “bad” parts:

Attribute measurement systems can be improved

by establishing standards and by operator training.

72
 Measurement Systems Analysis
(MSA)

Variable Gage
Repeatability and Reproducibility
(Variable Gage R&R)

73
 Why Study Measurement Systems?

Before you spend time and effort on a Green Belt project, we

must validate the integrity of the data we are going to use in the
decision-making process.
The study of measurement systems will provide information as
to the % of variation in your process data that comes from error
in the measurement.
It is also a great tool for comparing two or more measurement
devices or two or more operators against one another.
Measurement Systems Analysis should be used as part of the
criteria required to accept and release a new piece of
measurement equipment to manufacturing.
It should be the basis for evaluating a measurement system
that is suspect of being deficient.
74
 Possible Sources of Variation
Observed process or product
variation

Actual process or Measurement

product variation variation

Due to
Long-term Short-term Due to
measurement
variation variation operators
device (Gage)

σ 2 total = σ 2 product + σ 2 measurement Repeatability Reproducibility

Variable R&R study

Linearity
Calibration
Stability
program and
Gage
selection Accuracy
75
 Repeatability Defined

Repeatability of the instrument is a measure of the variation obtained when one

operator uses the same device to “repeatedly” measure the identical
characteristic on the same part.
part When no operator is present, repeatability
accounts for repeat measurements taken on an automated piece of test
equipment.
True value for one
part Quantifies the
repeatability of the
measurement system

Repeatability
Performance Characteristic

A variable R&R study will quantify the repeatability of the measurement system.

σ 2 total = σ 2 product + σ 2 measurement system

σ 2 total = σ 2 product + σ 2 repeatability + σ 2 reproducibility
76
 Reproducibility Defined

Reproducibility is the variation in the averages of measurements made by

different operators using the same device when measuring identical
characteristics of the same parts. Reproducibility may also be used to quantify
differences caused by different measuring devices (substitute measuring device
for operator)
Reproducibility
Quantifies
Operator B Operator A differences between
Device B Device A the operators
(devices)

Performance Characteristic
A variable R&R study will quantify the reproducibility of the measurement system.

σ 2 total = σ 2 product + σ 2 measurement system

σ 2 total = σ 2 product + σ 2 repeatability + σ 2 reproducibility
77
 The Methodology

Step 1: Collect 10 samples that represent the full range of long-

term process variation. In addition, identify the operators
who perform measurements on these parts daily.
Step 2: Calibrate the Gage or verify that the last calibration date
is valid.
Step 3: Set up the Minitab data-collection sheet for the R&R
study.
Step 4: Ask the first operator to measure all the samples once in
random order. Blind sampling, in which the operator does
not know the identity of each part, should be used to
reduce human bias.
Step 5: Have the second and then the third operators measure
all the samples once in random order. All operators have
now measured the samples once (this is Trial 1).
(continued)
78
 The Methodology
-cont.
Step 6: Repeat Steps 4 and 5 for the required number of trials.
Step 7: Enter the data into Minitab.
Step 8: Use Minitab to analyze the data by assessing the quality
of the measurement system. Determine follow-up actions.
Step 8: Analyze the Minitab output.

79
 Variable Gage R&R Example

Situation: In our bicycle factory, the quality department

measures the length of the tube stock used to form the
handlebars on a sample basis. They use a steel rule
with an end stop to make this measurement. The rule
measures to the nearest 0.01".There are three
operators who cut tube stock and record this data.

Task: Determine the adequacy of the measurement

system.

80
 Example

Step 1: Collect 10 samples that represent the full range of long-term

process variation. In addition, identify the operators who use this
instrument daily.
To determine the full range, let’s look at the capability analysis for these
parts.
Process Capability Analysis for Bikebar Length
The parts have a mean of
44.1" and a standard LSL USL
deviation of 0.10”, so we USL 44.25
would expect 95% of the LSL 43.75
parts produced to be Mean 44.10
between 43.9" and 44.3" StDev 0.10
(± 2σ). We should use
parts in this range for the
study. 43.8 43.9 44.0 44.1 44.2 44.3 44.4

Mary, Pat, and Joe are the operators who measure this parameter and so
should be selected as part of the measurement study.
81
 Example, Gage R&R Using Minitab

Step 2: Calibrate the Gage or verify that the last calibration date is valid.
In this case, we could measure some standard lengths to ensure that the steel
rule is not biased.
Step 3: Set up the Minitab data-collection sheet for the R&R study. Create the
R&R data-collection sheet for 10 parts, each measured two times by three
operators.
Column Headings:
Column 1: PartID (1 to 10)
Column 2: Operator (1 to 3)
Column 3: Operator Name
Column 4: Trial (1 to 2)
Column 5: Measure

82
 Example, Gage R&R Using Minitab

Step 4: Ask the first operator to measure all the samples once in
random order. The operator should not know the identity of each
part, to reduce human bias.
Mary measures all of the parts in random order.
Step 5: Have the second and then the third operator measure all
the samples once in random order. All operators have now
measured the samples once (this is Trial 1).
Pat and then Joe measure all of the parts in random order.
Step 6: Repeat Steps 4 and 5 for the required number of trials.
Mary then Pat and then Joe measure the parts a second time in
random order.

83
 Minitab Worksheet

Step 7: Enter the data into Minitab.

84
 Run the Analysis

Step 8: Use Minitab to analyze the data.

Stat>Quality Tools>Gage R&R Study>Options

Enter PartID,
Operator, and
Measure.

5.15 standard deviations

represent 99% of the
normal curve.
0.5 comes from ±0.25 Enter 5.15 and
the process tolerance 0.5.

85
 Gage R&R Output: Graphical
Step 9: Analyze the Minitab output.
Ga g e n a m e :
Da te o f s tu d y :
Gage R&R (AN OVA) for Meas ure Re p o rte d b y :
To l e ra n c e :
M is c :

Components of Variation By Part ID

4 4 .2 5
100 % Co n tri b u ti o n
% Stu d y Va r
% To l e ra n c e 4 4 .1 5
Percent

50
4 4 .0 5

0 4 3 .9 5
Ga g e R&R Re p e a t Re p ro d Pa rt-to -Pa rt Pa rt ID 1 2 3 4 5 6 7 8 9 10

R Chart by Name By Name

Joe M ar y Pat 4 4 .2 5
0 .0 4 UCL =0 .0 4 1 8 2
Sample Range

0 .0 3 4 4 .1 5
0 .0 2
R=0 .0 1 2 8 0 4 4 .0 5
0 .0 1
0 .0 0 L CL =0 4 3 .9 5
0 Na m e J oe M a ry Pa t

Xbar Chart by Name Name* Part ID Interaction

Joe M ar y Pat Na m e
4 4 .2 4 4 .2 J oe
Sample Mean

M a ry
UCL =4 4 .1 4
M e a n = 4 4 .1 1 Pa t
Average

4 4 .1 L CL =4 4 .0 9 4 4 .1

4 4 .0
4 4 .0
4 3 .9
0 Pa rt ID 1 2 3 4 5 6 7 8 9 10

Graphs! But what do they mean? Let’s investigate each section one at a time.
86
The Bar Chart

87
The Xbar and Range Chart

88
Gage R&R, Xbar, and R

89
Operator Bias

90
A Part-by-Part Look

91
 Another Graph: Gage Run Chart

It allows us to visualize repeatability and reproducibility within and

between operator and part
The center line is the overall average of the parts.
Gage name:
Runchart of Measure by Part ID, Name Date of study:
Reported by:
Tolerance:
Misc:
44.24
Joe
Mary
Measure

44.14
Pat
44.04

43.94
Part ID 1 2 3 4 5

44.24
Measure

44.14

44.04

43.94

6 7 8 9 10
Part ID

92
 Gage Results: Minitab Session Window

%Contribution
Source VarComp (of VarComp)

Total Gage R&R 1.90E-04 2.86

Repeatability 8.73E-05 1.32
Reproducibility 1.02E-04 1.54
Name 1.02E-04 1.54
Part-To-Part 6.45E-03 97.14
Total Variation 6.64E-03 100.00

StdDev Study Var %Study Var %Tolerance

Source (SD) (5.15*SD) (%SV) (SV/Toler)

Total Gage R&R 1.38E-02 0.070923 16.90 14.18

Repeatability 9.35E-03 0.048129 11.47 9.63
Reproducibility 1.01E-02 0.052093 12.41 10.42
Name 1.01E-02 0.052093 12.41 10.42
Part-To-Part 8.03E-02 0.413572 98.56 82.71
Total Variation 8.15E-02 0.419609 100.00 83.92

93
 Gage R&R, Distinct Categories

Number of Distinct Categories = Eight The part is in one of these eight zones.
This is the number of distinct categories this
measurement system can distinguish.
The number of groups within your process
data that your measurement system
can discern.
Source VarComp
Total Gage R&R 1.90E-04
Repeatability 8.73E-05
Reproducibility 1.02E-04
Name 1.02E-04
Part-To-Part 6.45E-03
Total Variation 6.64E-03

 σ 2 
Dist Categories = RoundDown Total
* 2
 σ MS
2


94
 Handling Poor Gage Capability

If a dominant source of variation is repeatability (equipment), you

need to replace, repair, or otherwise adjust the equipment.
If, in consultation with the equipment vendor or upon searches of
industry literature you find that the Gage technology you are using
is “state of the art” and it is performing to its specifications, you
should still fix the Gage.
If a dominant source of variation is operator (reproducibility), you
must address this via training and definition of the standard
operating procedure. You should look for differences among
operators to give you some indication as to whether it is a training,
skill, and/or procedure problem.
Evaluate the specifications. Are they reasonable?
If the Gage capability is marginal (as high as 30% of study
variation) and the process is operating at a high capability (Ppk
greater than 2), then the Gage is probably not hindering you and 95
you can continue to use it.
 Summary

Before you spend time and effort in a Green Belt

project, we must validate the integrity of the data we
are going to use in the decision-making process.
Measurement error is included with the process
variation in any observed “Y.”
σ 2 total = σ 2 product + σ 2 repeatability + σ 2 reproducibility

When conducting a Gage study, we need parts that

are representative of the entire range produced by the
process.
We will use Minitab output, both graphical and
numeric, to assess the capability of the Gage.
96
 Project Selection I

Objectives
To understand the importance of project selection
to Six Sigma success
To understand the difference between traditional
project selection and Six Sigma project selection
To be able to establish a business case

97
 Importance of Project Selection

Dissemination of the Six Sigma culture

depends on news of successful projects
having significant business impact.

Poor project selection is the most

common root cause of delays in
completion of Six Sigma projects.

98
 Six Sigma vs. Traditional Projects

Traditional Project Selection

Selected to optimize performance of one part of the business
Implementation of a pre-determined solution
Managing the exceptions vs. the norms
Lack clarity for project expectations
Examples:
Increase the number of sales Create a new Reporting
System
Reduce OT in one department Reduce the number of non-
Reduce the cost of one operation expensable expenses Paid
Improve on-time delivery Increase total revenue
Reduce the cycle time in one Install new equipment,
hardware/software
sub-process

(continued)
99
 Six Sigma vs. Traditional Projects -cont.

Traditionally, Projects Show Little or No Business Impact

Optimize part of the business at the expense of another
Decreasing cycle time in a non-bottleneck process
Reducing the cost in one area by increasing the cost of
another
Do not address the root causes of existing problems
Automating a bad process nets producing defects quicker
Create more incremental costs than savings
Increasing the number of products/features sold but not
generating additional revenue from sales
Reducing the total paid out in expenses by implementing an
audit process that costs more than the overage in expenses
paid (continued)
100
Six Sigma vs. Traditional Projects -cont.

Has too large a scope

Too many insignificant things are
distracting the attention of the team
Not enough attention is being
focused on things with the most
impact

101
 Six Sigma vs. Traditional Projects

Six Sigma Project Selection

Establishing a business case for a project
Avoids selecting projects with little or no business
impact
Narrowing the project focus based on a business
case
Avoids scope problems
Identifies the most significant areas to impact the
business case
Defining a project
Quantifies the problem and objectives, and outlines
the metrics used to determine project success
102
 Six Sigma vs. Traditional Projects

Six Sigma Project Selection

Establishing a business case for a
project
Avoids selecting projects with little or no
business impact
Narrowing the project focus based on a business case
Avoids scope problems
Identifies the most significant areas to impact the business
case
Defining a project
Quantifies the problem and objectives, and outlines the
metrics used to determine project success

103
 Establishing a Business Case

Business Case
The business case establishes the importance of the
project to the business in terms of meeting business
objectives
Components
The output unit (product/service) for external customer
The primary business measure of the output unit for the
project
The baseline performance of the primary business
measure
A gap in the baseline performance of the primary
business measure from the business objective
(continued)
104
 Establishing a Business Case -cont.

A business case establishes the need for a project

in terms of business objectives
Six Sigma business objectives
Reduce the cost/unit of a product
Decrease defects of a product
Increase product yield
Decrease the total cycle time of a product

(continued)
105
 Project Selection II
Narrowing Project Focus
Objectives
To know how to identify a narrow project focus that
will provide the largest impact to the problem
outlined in the business case
To be able to evaluate several potential projects
objectively using the project desirability matrix

106
 Narrowing Project Focus

Comments on Narrowing a Project’s Focus

When selecting an initial training project, it is
important that we look for high-leverage projects
where the return justifies the investment in time and
effort, and where the need for improvement is
substantial.
Please keep in mind that decisions based on
factual data are always better than those based
upon intuition, hearsay, or folklore.

(continued)
107
 Narrowing Project Focus -cont.

Narrowing Project Focus:

Narrowing of the focus must be consistent with the
primary business measure in the business case
The following data is used in narrowing the project
focus:
COPQ (re-work, scrap, etc.)
Narrows projects focused on reducing cost/unit
Defect counts (actual defects, RTY, FTY)
Narrows projects focused on increasing quality or
yield
Non-value-added time (re-work, delay, inspection)
Narrows projects focused on decreasing cycle time 108
 Narrowing Project Focus
Reducing Cost
Analysis to Reduce Project Focus:

SUB SUB SUB

INPUTS OUTPUT
PROCESS PROCESS PROCESS

Cost to
Cost + Cost + Cost + Cost = Produce
Step 1: High-level process map analysis of the
COPQ included in the cost to produce
Identify the re-work and scrap throughout the
process.
Attach the cost to business for each re-work
and scrap point in the process.
VALIDATE THESE COSTS WITH THE(continued)
COMPTROLLER. 109
 Narrowing Project Focus
Reducing Cost -cont.
Analysis to Reduce Project Focus
Step 2: Pareto analysis of the COPQ
COPQ of re-work vs. All Defects by Category,
xx/xx/99 to xx/xx/99
scrap costs 100
n=1298

COPQ for each sub 1000 80

process

Percent
60

Count
500 40
COPQ of each re-work
20
and scrap step
0 0
1 2 3 4 5 6 7 8 rs
o ry ory o ry ory ory ory ory ory he
teg teg teg teg teg teg teg teg Ot )
Ca Ca Ca Ca Ca Ca Ca Ca (1 2

Count 800 200 75 60 55 20 15 14 59

Percent 61.6 15.4 5.8 4.6 4.2 1.5 1.2 1.1 4.5
Cum % 61.6 77.0 82.8 87.4 91.7 93.2 94.4 95.5 100.0
Defect

110
 Narrowing Project Focus
Improving Quality
Analysis to Reduce Project Focus

SUB SUB SUB

INPUTS OUTPUT
PROCESS PROCESS PROCESS

Defects + Defects + Defects + Defects = Defects

RTY X RTY X RTY X RTY = RTY

Step 1: A high-level process map determines where

defects occur throughout the process
The volume of re-worked units within sub process,
rejected at start of next process
The volume of scrapped units within sub process,
(continued)
rejected at start of next process 111
 Narrowing Project Focus
Improving Quality -cont.
Analysis to Reduce Project Focus
Step 2: Pareto analysis of defects
All Defects By Category,
Defects for each model xx/xx/99 to xx/xx/99
100
Type of defect n=1298

1000 80

Defects by machine

Percent
60

Count
Defects by shift 500 40

20

Defects by production 0 0

line Ca
teg
ory
1

Ca
teg
ory
2

Ca
teg
ory
3

Ca
teg
o ry
4

Ca
teg
ory
5

Ca
teg
ory
6

Ca
teg
ory
7

Ca
teg
ory
8
he
Ot )
( 1 2
rs

Count 800 200 75 60 55 20 15 14 59

Defects by plant Percent

Cum %
61.6
61.6
15.4
77.0
5.8
82.8
4.6
87.4
4.2
91.7
1.5
93.2
1.2
94.4
1.1 4.5
95.5 100.0
Defect

(continued)
112
 Narrowing Project Focus
Improving Quality -cont.
Analysis to Reduce Project Focus
Step 3: Assess COPQ for the defects in the
narrowed project focus area
How much raw material is scrapped due to defects?
Re-worked? Re-cycled?
What is the cost of the time in labor, machinery, and raw
materials for the scrapped materials due to defects?
What is the cost of the time and labor spent re-working
defects?
What is the cost of the time and labor lost on recycled
defects? How do recycled materials impact the final product?
VALIDATE THESE COSTS WITH THE COMPTROLLER.

113
 Narrowing Project Focus
Reducing Cycle Time
Analysis to Reduce Project Focus:

SUB SUB SUB

INPUTS OUTPUT
PROCESS PROCESS PROCESS

Cycle
Time + Time + Time + Time = Time

Step 1: A high-level process map analysis of

non-value-added time throughout the process
Sub process input and output rates
Time spent in delay
Time spent in re-work sub processes
(continued)
Time spent in test/inspection sub processes 114
 Narrowing Project Focus
Reducing Cycle Time -cont.
Analysis to Reduce Project Focus
Step 2: Pareto analysis of non-value-added time
Volume of backlog at All Defects by Category,
xx/xx/99 to xx/xx/99
each delay step 100
n=1298
Time spent in delay for 1000 80

each backlog step

Percent
60

Count
Total non-value-added 500 40

time after the last 20

bottleneck 0
ory
1
ory
2
ory
3
o ry
4
ory
5
o ry
6
ory
7
ory
8
he
rs
0

teg teg teg teg teg teg teg teg Ot )

Ca Ca Ca Ca Ca Ca Ca Ca 2

Time spent in each

( 1

Count 800 200 75 60 55 20 15 14 59

Percent 61.6 15.4 5.8 4.6 4.2 1.5 1.2 1.1 4.5

re-work step Cum % 61.6 77.0 82.8 87.4

Defect
91.7 93.2 94.4 95.5 100.0

Time spent in each test/ (continued)

inspection step 115
 Narrowing Project Focus
Reducing Cycle Time -cont.
Analysis to Reduce Project Focus
Step 3: Assess COPQ for the non-value-added
time to be eliminated by the project focus area
What is the inventory cost of the time in delay?
How much do re-work steps cost in terms of labor?
Equipment?
How much do test/inspection steps cost in terms of
labor? Equipment?
What will be the impact on units sold as a result of
the lower cycle time?
VALIDATE THESE COSTS WITH THE
COMPTROLLER. 116
 Narrowing Project Focus
Additional Considerations
In addition to the impact to the stated problem in
the business case, there may be:
Business constraints on capital investment in the
project
Business constraints on resource investment in the
project
For certification only, one additional factor must be
considered when selecting a project:
The project must serve as a learning opportunity.
IT MUST BE FEASIBLE TO COMPLETE THIS
PROJECT WITHIN FOUR MONTHS.
The project should offer the opportunity to use as
many tools as possible. 117
 Project Selection III Project Definition
Problem Statement
What is a problem statement?
A problem statement describes in specific, concrete
terms what the data have revealed. It describes the
present undesirable situation with clarity and
objectivity while avoiding “hidden” solutions.
What are the purposes of a problem statement?
To focus the team on a process deficiency
To communicate the significance to others

118
 Project Definition
Problem Statement
Required Criteria for a Good Problem Statement
It states the effect.
It states what is wrong, not why it is wrong. Avoid
“lack of” and “due to” statements. These always
imply solutions.
It is measurable.
It states how often, how much, and when.
It is specific.
It avoids broad and ambiguous categories such as
“morale”, “productivity,” “communication,” and
“training.”
119
 Project Definition
Problem Statement
Additional Criteria for a Good Problem Statement
It focuses on the gap between what is and what
should be.
The gap may be a change or deviation from the
norm, standard, or the customer’s valid requirement
or expectation.
It is stated in an objective manner.
It is not stated as a question, which may tend to
imply a solution. It does not imply blame on any
person or department.
It focuses on the pain, either explicitly or implicitly.
The problem statement highlights “how” customers
are affected and the areas of discomfort, hurt, or
120
annoyance.
 Project Definition
Problem Statement
How are problem statements developed?
You should analyze and discuss all data collected
through narrowing the project focus.
A problem statement should be concise and answer
these questions:
Who is impacted by this problem?
What is the impact of this problem?
When has the problem occurred?
How do you know the problem occurs?
How many times does the problem occur?

121
 Project statement:

Fill In the Blanks For Your Project:

During ________________________ , the ____________________ for

(Period of time for Baseline Performance) (Primary Business Measure)

_____________ was _____________ . This gap of ________________

(Output Unit) (Baseline Performance) (Bus Obj Target vs. Baseline)

from ___________ represents ____________ of cost savings. This

(Business Objective) (Cost Impact of Gap)

project will _______________________________________________.

(Project’s Expected Impact on Performance of Primary Business Measure)

by _______________________.
(Project’s Expected Completion date)

122
 Project Definition
Project Objective
The project objective states the goal of the project.
It must:
Address the issue in the problem statement
Quantify the expected performance improvement
Identify the expected timing

123
 Project Definition
Project Objective
Why are objectives useful?
Objectives are set to give the team, as well as
others, a measure of the effectiveness of
performance,
To see whether improvement efforts are successful
in addressing the problem and, therefore, having an
impact on the problem stated in the business case.

124
 Project Definition
Project Objective
How to set objectives:
Objectives should be set to be challenging but
achievable during a reasonable amount of time.
They should be based on logic, not just pulled out
of the air.

125
 Project Definition
Primary Metric
The yardstick that will be used to measure your
success:
Must be consistent with the problem statement and
objective
Must be plotted on a time series graph with the
following lines:
Baseline performance (average over the past 12
months, if possible)
Actual performance
Target performance

126
 Project Definition
Primary Metric
Project Metric Line Graph Example
Product Returns
7%
Return $ as a Pct of Sales $

6%

5%

4%

3%
Baseline
2%
Actual
1% Target

0%
Feb97

Jun97
Mar97
Oct96
Nov96

May97

Oct97
Dec96

Dec97
Sep96

Jul97

Sep97

Nov97
Aug96

Jan97

Aug97
Apr97

127
 Project Definition
Secondary Metric
The conscience that will “keep you honest”
Tracks potential negative consequences
More than one may be required

128
 Project Selection Summary

Careful project selection is critical to the success of

the Six Sigma Quality Initiative.
The more desirable projects you have, the higher
the business impact, less effort is required, and
you’ll have a higher probability of success than
others.
Good problem statements and objectives clearly
communicate the scope, significance, and goals of
a project.
Primary and secondary metrics will be used to
measure the success of a project.
129
 6 Sigma Project Scope
Candidate Name_____Chris Reynolds_____________________ Champion Name: ____Lavon Baxter_____________________________

Division: ____White- Rodgers__________________________________ Location: _Harrison, Arkansas_________________________

_
Problem Statement: __In fiscal year 1999, 25M gas valve reject rate for noise was 125,500 PPM. This PPM level equates to $105,000
in identified non-value added cost. This is the highest non-value added cost activity and requires over 6,000 hours of rework labor
Each year. In addition, the scrap generated from this reject level is over $15,000.

Baseline Time Frame: ____FY 99______________ ( The time period for data collection)

Project Primary Metric: (example - Cycle Time, PPM, Productivity) ___PPM’s at Noise Test.__________________________________

Project Baseline: __125,000 PPM’s_________ Project Goal: ____50,000 PPM’s____________ Gap: _75,000 PPM’s________________

Project Secondary Metric: (example - Cycle Time, PPM, Productivity) ____Rework Labor_____________
Project Third Metric:_________________________________

This project begins with the Tube Assembly process operation and ends with the _____Noise Test______ process operation.

This project will be focused in the ___Harrisburg_________________ Facility, ____at the 25M line____________________
(Product Name, Line or Processes etc.)

Project Objective: _The objective of this project is to reduce the 25M Gas Valve noise test PPM by 60% by Jul 00.
This reduction would yield a cost savings of $63,000. _______________________________________________________________
(Project expected impact on performance of project primary business metric, dollars & completion date)

130
 Candidate Selection

Black Belt / Green Belt Selection Criteria

Desire to Drive Change
Someone who is not afraid to take data that might
not tell people what they want to hear
Effective Communication Skills
Someone who can communicate across functional
and hierarchical boundaries
Demonstrated Leadership Abilities
Someone who can motivate others to participate in
the work ahead

131
 EMC Implementation Strategy
Six Sigma Belt Roles and
Roles: Responsibilities
Leads the teams as they implement the 6σ methodology on projects
Introduces the methodology and tools to Team Members and the broader
organization
Reporting Lines:
Can report directly to the Champion or to the function
Must feel a part of the functional organization
Responsibilities:
Applies the methodology completely to the project
Acts as both a technical and cultural change agent for quality
Spreads the methodology to the project teams
Supports the efforts of the function by spreading the use of the
methodology when called on to assist on other business issues
Has dual membership: in the functional and 6σ teams
Must have technical competencies required to effectively execute the
Six Sigma tools
Time Commitment:
Must be substantially dedicated (25% – 100%)
Min. Training Requirement:
Attend Six Sigma training class
Complete all project-related requirements

132
Mentoring

133
 Mentoring
Candidate: Mr or Ms Candidate Champion: Mr or Ms Champion Location: Anywhere USA
Tel / e-mail: [email protected] Tel / e-mail: [email protected]
Project Title: To significantly improve my process Training Phase: Measure
Date of this report: 01/01/2001 Support Person: Mr or Ms SSQI Support

Problem Statement:
Enter your problem

Objective / Goal Statement:

Enter your goal

A 0 rating indicates No action has been taken

Project Deliverables - update each visit Weighting Score (0-4) Weighted score 2 3 4

Complete - does not

Complete
have all elements
Problem Statement 3 4 12
Vague - No dollars or Perfectly clear -
Objective / Goal Statement 3 4 12 no goal. Goals set
COPQ Definition 3 4 12 Estimate Signed Off
Contains 3 elements;
Baseline, Goal & up to date
Metric Chart Actual
Primary ~ Defects - DPU, PPM, RTY, etc. 3 4 12
Secondary ~ Potential negatives - COPQ, etc. 3 4 12
Project Time Line (Projplan.XLS) 3 4 12 not up to date up to date
Team Meetings 3 4 12 < 1 per week >= 1 per week
Project Notebook 3 4 12 not up to date up to date
Project Complexity Information Only Very Very Complex Complex Slam Dunk
% Time dedicated to Six Sigma since last review Information Only 34% to 66% <33% 34% to 66% >66%
Fair. Somewhat
Poor. Having difficult Good. Leader,
understand the
time understanding aggressive,
tools and their
Candidate Self Assessment Information Only Good how tools are easily gets
application,
selected and applied tasks
getting by with
to the project accomplished
help from SSQ.
Project Status Summary
reviewed with:
Champion 3 4 12 Not per Project Plan Per Project
Key Stakeholders 3 4 12

134
 Weighted
Mentoring
Measure Phase Deliverables score A 0 rating indicates No action has been taken
Weighting Score (0-4) 1 2 3 4
Process Map 3 3 9 Started First Pass Being Revised Completed
Multi Level Pareto Charts 3 3 9 Data being collected One Level Two Level Three Level
Fishbone/5-Why 1 3 3 Planned In process Completed Analyzed
Decision(s)
XY Matrix (XYMatrix.XLS) 2 3 6 Data being collected In Process Completed made
Action plan
FMEA (pfmea_cp.XLS) 3 3 9 Started Completed RPN's Ranked developed
Gage R&R 3 3 9 Scheduled / Planned In Process Completed Analyzed
Data collection Capability
Process Capability 3 3 9 Scheduled / Planned In Process completed established
Analyze Phase Deliverables - Graphical Analysis (x Search)
Demonstrated appropriate tool use Tool use not Appropriate tool Exceptional
in completing the Measure Phase appropriate use Application
List of possible x's 3 0 0 Scheduled / Planned Being Compiled Being Finalized Completed
Identification of Critical X's 3 0 0 Scheduled / Planned <=50% Complete <=75% Complete Completed
Multi-Vari Analysis 3 0 0 Understands proper Understands how Has applied tool Has applied
t - Tests / Non-parametric hypothesis 3 0 0
Test of Equal Variance 3 0 0
Proportions Tests / Chi - Square 3 0 0
ANOVA 3 0 0

Correlation and Regression 3 0 0

Improve Phase Deliverables
Demonstrated appropriate tool use Tool use not Appropriate tool Exceptional
in completing the Analyze Phase appropriate use Application
Analysis of Analysis of
DOE 4 0 0 Scheduled / Planned In Process simulation project DOE
Sample Size Selection 3 0 0 Scheduled / Planned Applied
Second Project (Measure Phase) 3 0 0 Started <=50% Complete <=75% Complete Completed
Control Phase Deliverables
Demonstrated appropriate tool use Tool use not Appropriate tool Exceptional
in completing the Improve Phase appropriate use Application
Second Project (Analyze Phase) 3 0 0 Started <=50% Complete <=75% Complete Completed
Control Plan 4 0 0 Scheduled / Planned In Process Completed Revised
Lean Fundamentals 2 0 0 5 S's Mistake Proofing Both In Control Plan
1st pass
Draft Draft Completed
One Page Summary Report 3 0 0 submitted
Final Report 4 0 0

Measure Phase Total Score #

Tool Knowledge Appropriateness of Tool Use
green N/A

See Guidelines tab for rating explanation.

135
 Change Management

Objectives
To know how a team leader can set up a team for
success in the change effort
To know the elements of making the team’s change
efforts successful in the organization

136
 Successful Change Project

100% THE TARGET

FOR A SUCCESSFUL BLACKBELT PROJECT

Six Sigma Team Support,
BELTS PLANT STAFF
INVOLVEMENT

MEASURE ANALYZE IMPROVE CONTROL

TIME
137
 Elements of Successful Change
Identifying the Change Leader
Successful change initiatives require strong
committed leadership throughout the entire project
lifecycle. The leadership offers:
Visible, active, and public commitment and support
A willingness to take personal initiative and to
challenge the status quo
A high-level of attention to the project
Giving time to the team
Talking about the project to others
Establishing priorities for the project against other
demands in the organization

138
 Elements of Successful Change
Building Mutual Need
Creating a shared need involves framing the need
to appeal to the interest of all those to be impacted
by the change:
A shared recognition by the team and the
stakeholders for the need and logic for the change
Appealing to the unhappiness with the current
situation
The ability to take perceived threats of the change
and turn them into opportunities for the future

139
 Elements of Successful Change
Forming a Vision
Building a vision provides an organization with the
direction and motivation to make the change:
A view of what the future will look like with this
change
Appeals to logic and intuition while explaining why
the change is needed
Can help to establish milestones to mark the
progress of the change

140
 Elements of Successful Change
Mobilizing Commitment

Mobilizing commitment positions the team for

support when it is faced with obstacles.
Create an alliance of committed supporters.
Identify the potential sources of resistance.
Convert all key influencers.

141
 Elements of Successful Change
Sustaining Change

Change initiatives must be composed of

commitments rather than “assignments:”
Having consistent, visible, and tangible
reinforcement of the changed behavior is needed.
Integrating change into on-going work behaviors is
needed.
Aligning systems and structures helps make the
change a part of individual and team behavior.

142
 Elements of Successful Change
Monitoring Change

Good measurement systems need to be

established early in the project.
What are the metrics that will be used to determine
the team’s success in making a sustained change?
Tracking these metrics and sharing them across all
those impacted by the team
Winning support of doubters through results

143
 Elements of Successful Change
Aligning Systems and Structures

Assessing the existing system of measures and

rewards will highlight existing practices that are
inconsistent or unproductive.
Identifying critical system and structure areas that
must be addressed
Assessing the risk of slipping back into “old habits”
Aligning systems and structures with desired
behaviors

144
 Elements of Successful Change
Aligning Systems and Structures
System and Structures Integration Template -cont.
Evaluate rewards and recognition
1. List the behavioral changes.
2. Evaluate the organization’s likely reaction to
displaying behaviors that are desired to move to and
behaviors desired to move from:
A= Reward or approval
B= Punishment or disapproval
C= No reaction
D= Impossible to predict
(continued)
145
 Elements of Successful Change
Aligning Systems and Structures
System and Structures Integration Template -cont.
3. List the existing rewards available to employees.
4. Evaluate the existing rewards using the
measurement checklist.
5. Develop an action plan to reward the desired
behaviors not included in performance metrics or
included in the likely to receive reward or approval.
6. Develop an action plan to eliminate rewarding
behaviors that are not desired.

146
Wrap-Up

147
148
149
150
151
152