NNOVATION

1
S
S
IX
IGMA
I
NNOVATION
CUSTOMER COMPETITIVE INTELLIGENCE FOR PRODUCT,
PROCESS, SYSTEMS ENTERPRISE EXCELLENCE
DEPARTMENT OF STATISTICS
REDGEMAN_at_UIDAHO.EDU
OFFICE 1-208-885-4410
DR. RICK EDGEMAN, PROFESSOR CHAIR SIX SIGMA
BLACK BELT
2
S
S
IX
IGMA
a highly structured strategy for acquiring,
assessing, and applying customer, competitor, and
enterprise intelligence for the purposes of
product, system or enterprise innovation and
design.
DEPARTMENT OF STATISTICS
3
Six Sigma COPIS Model
Process Steps
Outputs
Inputs
Suppliers
Customers
How does Six Sigma Work?
The Voice of the Customer (VOC) is aggressively
sought and rigorously evaluated and used to
determine needed outputs and hence the optimal
process configuration needed to yield those
outputs and their necessary inputs for which the
best suppliers are identified and allied
with. From Concept to Market the Voice of the
Customer
4
Why is Six Sigma Important?
The Villain
Cost of Poorly Performing Processes (CP3)
s level DPMO
CP3 2 308,537 Not Applicable 3 66,807
25-40 of sales 4 6,210 15-25 of
sales 5 233 5-15 of sales 6 3.4 lt 1
of sales Each sigma shift
provides a 10 net income improvement
5
Define the problem and customer
requirements. Measure defect rates and document
the process in its current incarnation. Analyze
process data and determine the capability of
the process. Improve the process and remove
defect causes. Control process performance and
ensure that defects do not recur.
Define
Control
Measure
Improve
Analyze
Six Sigma Innovation
6
Six Sigma DMAICInnovation Approach

• Slides 4 Through 32 of the
• Project Definition Team Formation
• Powerpoint File Represent the Six Sigma
• DEFINE Stage

7
Define the problem and customer
requirements. Measure defect rates and document
the process in its current incarnation. Analyze
process data and determine the capability of
the process. Improve the process and remove
defect causes. Control process performance and
ensure that defects do not recur.
Define
Control
Measure
Improve
Analyze
Six Sigma Innovation
8

• Measure
• What Measurements are Important and
• What Tools Should be Used?
• Select Customer Critical to Quality (CTQ)
  Characteristics
• Define Performance Standards (Numbers Units)
• Establish the Data Collection Plan,
• Validate the Measurement System,
• and Collect the Necessary Data.

9
Measure 1. Select Customer Critical to
Quality (CTQ) Characteristics. Among
useful quality tools in the MEASURE phase
are Quality Function Deployment (QFD) which
relates CTQs to measurable internal
sub-processes or product characteristics. Process
Maps create a shared view of the process,
reveals redundant or Unnecessary steps, and
compares the actual process to the ideal
one. Fishbone Diagrams provide a structure for
revealing causes of the effect. Pareto Analysis
provides a useful quantitative means of
separating the vital few causes of the effect
from the trivial many, but require valid
historical data. Failure Modes and Effects
Analysis (FMEA) identifies ways that a
sub- process or product can fail and develops
plans to prevent those failures. FMEA is
especially useful with high-risk projects.
10
Kano Customer Need Model
Delighted
Degree of Execution
TIME
Fully Implemented
Absent
Disgusted
Stakeholder Satisfaction
11
Measure 1. Select Customer Critical to Quality
(CTQ) Characteristics. FAILURE
MODES AND EFFECTS ANALYSIS (FMEA) Failure Modes
and Effects Analysis (FMEA) Process is a
structured approach that has the goal of linking
the FAILURE MODES to an EFFECT over time for the
purpose of prevention. The structure of FMEA is
as follows Preparation ? FMEA Process ?
Improvement a. Select the team b. Develop
the process map and steps c. List key process
outputs to satisfy internal and external customer
requirements d. Define the
relationships between outputs and process
variables e. Rank inputs according to
importance.
12
Measure 1. Select Customer Critical to Quality
(CTQ) Characteristics. FAILURE
MODES AND EFFECTS ANALYSIS (FMEA)
Preparation ? FMEA Process ? Improvement a.
Identify the ways in which process inputs can
vary (causes) and identify associated
FAILURE MODES. These are ways that critical
customer requirements might not be met. b.
Assign severity, occurrence and detection ratings
to each cause and calculate the RISK
PRIORITY NUMBERS (RPNs). c. Determine
recommended actions to reduce RPNs. d.
Estimate time frames for corrective actions.
e. Take actions and put controls in place.
f. Recalculate all RPNs. FAILURE MODE How
a part or process can fail to meet
specifications. CAUSE A deficiency that
results in a failure mode ? sources of variation
EFFECT Impact on customer if the failure
mode is not prevented or corrected.
13
FMEA Standardized Rating System 1 lt RPN (Degree
of Severity)(Likelihood of Occurrence)(Ability
to Detect) lt 1000
14
(No Transcript)
15
Measure 1. Select Customer Critical to
Quality (CTQ) Characteristics. FAILURE MODES AND
EFFECTS ANALYSIS (FMEA) Preparation ? FMEA
Process ? Improvement Develop and
implement plans to reduce RPNs.
16
Measure 2. Define Performance Standards Numbers
Units At this stage customer needs are
translated into clearly defined measurable
traits. OPERATIONAL DEFINITION This is a
precise description that removes any ambiguity
about a process and provides a clear way to
measure that process. An operational definition
is a key step towards getting a value for the CTQ
that is being measured. TARGET PERFORMANCE
Where a process or product characteristic is
aimed If there were no variation in the product
/ process then this is the value that would
always occur. SPECIFICATION LIMIT The amount
of variation that the customer is willing to
tolerate in a process or product. This is
usually shown by the upper and lower
boundary which, if exceeded, will cause the
customer to reject the process or
product. DEFECT DEFINITION Any process or
product characteristic that deviates outside of
specification limits.
17
Measure 3. Establish Data Collection
Plan, Validate the Measurement System,
and Collect Data. A Good Data Collection
Plan a. Provides clearly documented strategy
for collecting reliable data b. Gives all
team members a common reference c. Helps to
ensure that resources are used effectively to
collect only critical data. The cost of
obtaining new data should be weighed vs. its
benefit. There may be viable historical data
available. We refer to actual process
variation and measure actual output a. what
is the measurement process used? b. describe
that procedure c. what is the precision of the
system? d. how was precision determined
e. what does the gage supplier state about
f. Do we have results of either a Accuracy
Precision Resolution
Test-Retest Study? Gage RR
Study?
18
Measure 3. Establish Data Collection Plan,
Validate the Measurement System, and Collect
Data. Note that our measurement process may also
have variation. a. Gage Variability
Precision Accuracy Both b.
Operator Variability Differences between
operators related to measurement. c. Other
Variability Many possible sources.
Repeatability Assess effects within ONE unit of
your measurement system, e.g., the variation in
the measurements of ONE device. Reproducibility
Assesses the effects across the measurement
process, e.g., the variation between different
operators. Resolution The incremental aspect of
the measurement device.
19
Measure 3. Establish Data Collection Plan,
Validate the Measurement System, Collect
Data. GAGE RR (Repeatability Reproducibility)
STUDY a. Operators at least 3
recommended b. Part the product or process
being measured. It is recommended that at least
10 representative (reflects the range of parts
possible) parts per study, with each operator
measuring the same parts. c. Trial each time
the item is measured. There should be at least 3
trials per part, per customer.
20
Define the problem and customer
requirements. Measure defect rates and document
the process in its current incarnation. Analyze
process data and determine the capability of
the process. Improve the process and remove
defect causes. Control process performance and
ensure that defects do not recur.
Define
Control
Measure
Improve
Analyze
Six Sigma Innovation
21
Analyze Where are we now? Where are we
going? What can prevent us from reaching our
goals? At this stage we determine the process
sigma level and regard variation as an enemy. We
must determine process capability, that is, the
ability of the process to meet customer
requirements. We require several z-scores to
make this evaluation. ZBENCH Zst ZLT
ZLSL ZUSL Where BENCH benchmark,
st short term, LT long term, LSL
lower specification limit, and USL upper
specification limit.
22
Analyze Where are we now? Where are we
going? What can prevent us from reaching
our goals? ZST best performance that can be
expected from a process ZLT allows for drift
through time (1 to 2 sigma drift is typical)
ZLSL (X LSL) / S then determine PLSL(d)
ZUSL (USL X) / S then determine PUSL(d)
P(d) PLSL(d) PUSL(d) then apply inverse use
of the Z-table to find ZBENCH
(long-term) P(d) 1,000,000 DPMO or PPM
ZBENCH
23
Analyze Where are we now? Where are
we going? What can prevent us from reaching
our goals? ZSHIFT ZST ZLT drift over
time (DPMO tables assume 1.5) ZST
(Specification Limit Target) / ?ST
process sigma is determined here indicates
potential process performance if only
common cause variation is present. ZLT
(Specification Limit - ?) / ?LT reveals
long-term process capability used to
estimate DPMO or PPM (parts
per million same as DPMO) includes
special cause variation


24
Analyze An Alternative Means of Approximating
the Sigma Capability for Your Process

• Step Action Equations Your Calculations
• What process do you want to consider? N/A
  Billing Charging
• How many units were put through the N/A
  2,000
• process?
• Of the units that went into the process, N/A
  1,800
• how many were OK?
• 4 Compute process yield (step 3)/(step 2)
  0.9000
• 5 Compute defect rate 1.0 (step 4)
  0.1000
• Determine the number of potential N number
  of 16
• things that could create a defect
  critical-to-quality
• characteristics
• Compute the defect rate per CTQ (step
  5)/(step 6) 0.00625
• characteristic
• 8 Compute DPMO (step 7)(1 million)
  6,250
• 9 Convert DPMO to s value conversion chart
  About 4.0
• 10 Draw conclusions JUST ABOUT INDUSTRY
  AVERAGE

25
Analyze Where are we now? Where are we
going? What can prevent us from reaching our
goals? Yield Rates A yield rate is a pass rate
and can be characterized in various
ways Classical Yield Rate YC (total defect
free parts) / (total parts) First Time Yield
Rate YFT (parts defect free on the first
pass)/(total parts) Throughput Yield YT
eDPU where DPU defects per unit and is
calculated as DPU (number of defects at any
stage) / (total inspected). Note that due to
rework some items may be inspected multiple times
with each inspection adding to the total.
26
Where are we now? Where are we going? What can
prevent us from reaching our goals? Yield
Rates Classical Yield Rate (total defect free
parts) / (total parts) ¾ 75 First Time
Yield Rate (parts defect free on the first
pass) / (total parts) ¼ 25 Throughput Yield
eDPU where DPU defects per unit and is
e18/8 .1054 The rework that it
takes to raise throughput yield to the
classical yield level is called Hidden Factory.
X
Analyze
27
Analyze Setting Performance Objectives
Critical to the Setting of Performance
Objectives are the Concepts of Baseline,
Process Entitlement, Benchmark and
Benchmarking BASELINE This is the process
performance level at the start of the Six Sigma
Project. PROCESS ENTITLEMENT This is our best
expectation for process performance (e.g., the
sigma level) with the current technology
that is, without substantial reengineering or
investment. This can be estimated from
Zst. BENCHMARK This is the current best in
class performance level. BENCHMARKING The
process of finding the benchmark
performance level and then matching or exceeding
that performance.
28

• Analyze
• Sources of Variation
• This is the search for the Vital Xs
• the factors that drive the customer CTQs.
• Various statistical and quality methods are
  useful in this effort.
• Among these are
• HYPOTHESIS TESTING, which can
• Reveal Significant Differences in Performance
  Between Processes
• Validate Process Improvements
• Identify Factors that Impact the Process Mean
  and Variation.
• FISHBONE or CAUSE-AND-EFFECT DIAGRAMS

29

• Analyze
• Sources of Variation
• The Hypothesis Testing Algorithm
• Formulate the Null and Alternative Hypotheses, H0
  and HA.
• Specify the Sample Size and Significance Level of
  the Test, n and ?
• Determine Which Type of Test Should be Employed.
• State the Critical Value(s) the Test Statistic
  Specify the Decision Rule.
• Collect and Validate Process Data.
• Determine the Calculated Value of the Test
  Statistic (Data Based)
• As Appropriate, Construct and Interpret
  Confidence Intervals.
• Determine and Pursue a Course of Action.
• Key Vocabulary Type I and II Errors, ? and ?

30
Define the problem and customer
requirements. Measure defect rates and document
the process in its current incarnation. Analyze
process data and determine the capability of
the process. Improve the process and remove
defect causes. Control process performance and
ensure that defects do not recur.
Define
Control
Measure
Improve
Analyze
Six Sigma Innovation
31
Improve The goal of the improve phase is to
test sources of variation to determine which of
these actually cause process variation in the
customer CTQ. 7. Screen / Identify Causes of
Variation. 8. Discover Variable
Relationships. 9. Estimate Operating Tolerances
Pilot Solutions.
32
Improve 7. Screen / Identify Causes of
Variation. At this stage we determine which
factors will be changed to improve the CTQs. In
step 6 (MEASURE) we selected the vital few xs
for each CTQ (little y). In step 7 we select an
appropriate improvement strategy based
upon characterizing xs as either operating
parameters or critical elements. Operating
Parameters are xs that change in amount, rather
than being replaced with another type / kind.
Operating Parameters can be set to several
levels to see how they affect the process
Y. Critical Elements are xs that are typically
changed in type or kind, rather than in amount.
These xs are not necessarily measurable on a
specific scale.
33
Improve 7. Screen / Identify Causes of
Variation. Having identified the pertinent
operating parameters and / or critical
elements, we would then review whether Design of
Experiments (DOE) would be appropriate and, if
so, develop the appropriate design, called a
screening design. The screening design is used
to validate or eliminate factors (i.e. xs), but
is not ordinarily able to determine the optimal
settings of the xs. Important considerations
include the number of factors, number of
levels of each, the range of settings for each
factor, replication, randomization whether to
use blocking variables.
34
Improve 8. Discover Variable Relationships.
GOAL to determine the precise changes needed
It is common to apply Optimizing DOE at this
point, to determine the best settings of the
xs. It is common to use fractional
factorial designs or central composite designs to
accomplish this goal. It is common to include
baseline conditions among the factor settings.
We desire to determine the transfer function
(the regression equation). In combination these
are intended to yield a proposed solution to
achieve project objectives. Important
considerations include the testing budget,
available personnel, and time allotted for the
study.
35
Improve 9. Estimate Operating Tolerances
Pilot Solutions. PURPOSE to estimate the range
of values for each vital x that will satisfy
customer requirements. CONCEPT IF we can
characterize the x-Y relationship AND we know the
required specifications of Y, THEN the tolerances
can be set for each x factor. Specifications
flow down from customer requirements and we
adjust tolerances accounting for variation,
unless variation is small enough to be
ignored. STATISTICAL TOLERANCING
36
1

Adjust for Variation in Y, THEN
YUSL
2
This graph indicates an indirect relationship
between x and Y
YUSL
3
YTarget
Statistical Tolerancing
YLSL
3
2
Adjust further for variation in X.

YLSL
3
3
1
Original Y Specifications


xU
xL
Improve 9. Estimate Operating Tolerances
Pilot Solutions.
37
Define the problem and customer
requirements. Measure defect rates and document
the process in its current incarnation. Analyze
process data and determine the capability of
the process. Improve the process and remove
defect causes. Control process performance and
ensure that defects do not recur.
Define
Control
Measure
Improve
Analyze
Six Sigma Innovation
38
Control The Goal of the Improve Phase is to
Test Sources of Variation to Determine which of
These Actually Cause Process Variation. 10.
Validate the measurement system of the control
variables. 11. Determine process capability. 12.
Implement process control system bring the
process to a close.
39
Control 10. Validate
the measurement system of the control variables.
GOAL Make sure the implemented solution
remains effective and in control HERE WE
ESTABLISH THE Process Control System Even though
our solution may be an excellent one, the nature
of most systems is toward entropy or
degradation, thus we will (a) Create an
implementation plan with controls for each x
(b) Prepare documentation and provide (for)
training, and (c) Collect data to re-evaluate
process capability. RECALL our Measurement
System Analysis, as used in Step 3 of
MEASURE a. what is the measurement process
used? b. describe that procedure c. what is
the precision of the system? d. how
was precision determined e. what does the gage
supplier state about f. Do we have results
of either a Accuracy Precision
Resolution Test-Retest Study? or a
Gage RR Study? In MEASURE the MSA was applied
to y. In CONTROL it is applied to x.
40
Control 11. Determine process capability. The
GOAL at this stage is to statistically confirm
that the implemented changes have produced
improved performance. Process capability is
reassessed. A Hypothesis Test may be appropriate
to evaluate the difference in performance
prior to and after the implemented changes
41
Control 12. Implement process control system
and bring the project to a close Three primary
approaches may be used at this stage Risk
Management This is similar to FMEA but now
focus is trained on x, rather than y. Risk
Management Score RMS (Impact)(Probability)
RM identifies and quantifies risks, establishes
a risk abatement plan, and monitors the progress
of the plan. Mistake Proofing This is a
technique for eliminating errors by making it
impossible to make them in the process. To quote
It is good to do it right the first time. It is
even better to make it impossible to do it
wrong. Statistical Process Control (Charts)
This is a feedback system with sequential data
and ongoing process data collection.
42
Black Belt Perspective
1 Select CTQ Characteristic 2 Define Performance
Standards 3 Validate the Measurement
System 4 Establish Product Capability 5 Define
Performance Objectives 6 Identify Variation
Sources 7 Screen Potential Causes 8 Discover
Variable Relationships 9 Establish Operating
Tolerances 10 Validate Measurement
System 11 Determine Process Capability 12 Implemen
t Process Controls
Measure Analyze Improve Control
43
Business Level View R Recognize the true
states of your business. D Define what plans must
be in place to realize improvement of each
state. M Measure the business systems that
support the plans. A Analyze the gaps in system
performance benchmarks. I Improve system
elements to achieve performance goals. C Control
system-level characteristics that are critical to
value. S Standardize the systems that prove to
be best-in-class. I Integrate best-in-class
systems into the strategic planning framework.
44
Operations Perspective R Recognize operational
issues that link to key business
systems. D Define Six Sigma projects to resolve
operational issues. M Measure performance of the
Six Sigma projects. A Analyze project performance
in relation to operational goals. I Improve Six
Sigma project management system. C Control inputs
to project management system. S Standardize
best-in-class management system
practices. I Integrate standardized Six Sigma
practices into policies and procedures.
45
A Process View R Recognize functional problems
that link to operational issues. D Define the
processes that contribute to the functional
problems. M Measure the capability of each
process that offers operational leverage. A Analy
ze the data to assess prevalent patterns
trends. I Improve the key product / service
characteristics created by the key processes.
C Control the process variables that exert undue
influence. S Standardize the methods
processes that produce best-in-class
performance. I Integrate standard methods
processes into the design cycle.
46
S
S
IX
IGMA
I
NNOVATION
End of Session
DEPARTMENT OF STATISTICS
REDGEMAN_at_UIDAHO.EDU
OFFICE 1-208-885-4410
DR. RICK EDGEMAN, PROFESSOR CHAIR SIX SIGMA
BLACK BELT