Six Sigma is a data-driven methodology for process improvement originally developed by Motorola. It involves defining a project goal, measuring key aspects of the current process, analyzing data to determine root causes of defects, improving the process by addressing causes, and controlling future process variation. The document provides an overview of Six Sigma and its development, then gives an example project summary involving improving calcium levels in a product. The project uses Six Sigma tools like process mapping, measurement systems analysis, data analysis, design of experiments, and risk analysis to select and validate factors influencing calcium and develop improvements.

1. Six Sigma

2. What Is Six Sigma ?
• A metric — standard deviations in a normal curve
• A goal — 3.4 defects per million opportunities
• A rigorous, process focused methodology
– the DMAIC process
• A management
philosophy
Business
and/or Customer
Requirement
1
2
3
4
5
6σ
Defects Good

3. Where Does Six Sigma Come From?
• Necessity is the mother of invention
– Motorola was losing market share to foreign rivals who
had better quality and lower cost.
– A Japanese firm took over a Motorola television factory.
After implementing changes, the factory was producing
with 1/20th the defect rate. Same people, same
equipment, same designs…..different management and
different processes.
• “Our quality stinks.” — Art Sundry, Motorola
©2005 Tyg Lucas

4. • Late 1970s Mikel Harry (the man with two first names), a
senior staff engineer, is using statistical analysis for
problem solving. He was working in the Government
Electronics Group (GEG)
• Though certainly not the first to apply statistical thinking to
manufacturing analysis, he is the one who went on to
refine a methodology and then call it “Six Sigma.” He
wrote an internal paper called “The Strategic Vision for
Accelerating Six Sigma Within Motorola."
Who Developed Six Sigma ?

5. • Bill Smith and throughput yield:
– Motorola had quality issues even on products that had
highly capable processes. Why?
– Bill Smith (sometimes referred to as the father of Six
Sigma) examines the issue.
– Individual yields are combined into a “rolled throughput
yield.” (So a 100-component product with individual
yields of 99.9% still only gives a completed product with
90% reliability.)
– He also developed many of the tools and techniques
that became the Six Sigma methodology.
Why Six of Those Sigmas ?

6. Six Sigma Early Development
• By the mid-1980s, Bob Galvin, Motorola CEO, has the
company focused on improving quality.
• 1988, Motorola wins the first Malcolm Baldridge Quality
Award. Part of winning this national quality award is the
agreement to share the methods used to achieve the high
levels of quality.
• Other companies initiate “Six Sigma” programs, notably
Larry Bossidy at Allied Signal.
• Larry tells his friend Jack Welch about it. Jack applies it at
GE in a very big, very GE way.

7. Six Sigma’s Methodologies
Six Sigma
Process Management
Improve EXISTING
processes so that
their outputs meet
customer
requirements
Control and manage cross-function
processes to meet business goals
Design
NEW products
and processes
that meet
customer needs

8. • Project management
• Voice of the Customer
• Process mapping
• Data collection
• Data graphs
• Gage R&R
• Operational definitions
• Process Capability
Assessment
• Hypothesis testing
• Regression analysis
• Designed experiments
• Statistical process control
• FMEA
• Stakeholder analysis
• Implementation planning
• Tollgate reviews
Common Six Sigma Tools
nothing new here . . .

9. Analyze
Define
Measure
Improve
Control
The Powerful DMAIC Road Map

10. Define

11. Define
Project Charter
Problem Statement:
Goal:
Business Case:
Scope:
Cost Benefit Projection:
Milestones:
VOC Key Issue CTQ
Delighters
More Is Better
Must Be
Voice of the CustomerBusiness Case
Initial Process Mapping
OutputsProcessInputs
Yield: 60%
Yield: 90%
Yield: 45%
Yield: 98%
CUSTOMERS
SUPPLIERS

12. Measure

13. Measure
Col # 1 2 3 4 5 6
Inspector A B
Sample # 1st Trial 2nd Trial Diff 1st Trial 2nd Trial Diff
1 2.0 1.0 1.0 1.5 1.5 0.0
2 2.0 3.0 1.0 2.5 2.5 0.0
3 1.5 1.0 0.5 2.0 1.5 0.5
4 3.0 3.0 0.0 2.0 2.5 0.5
5 2.0 1.5 0.5 1.5 0.5 1.0
Totals 10.5 9.5 3.0 9.5 8.5 2.0
Averages 2.1 1.9 0.6 1.9 1.7 0.4
Sum 4.0 Sum 3.6
XA 2.0 XB 1.8R
A
R
B
Validate Measurement
Systems
Display Data
0
1000
-1000
10 20 30
UCL
X
LCL
D B F A C E Other
Identify the Metrics
Data Collection Plan
Operational Definition and Procedures
Data Collection Plan
What questions do you want to answer?
Data
What Measure type/
Data type
How
measured
Related
conditions
Sampling
notes
How/
where
How will you ensure
consistency and stability?
What is your plan for
starting data collection?
How will the data be displayed?
Prioritize the Metrics
I1
I2
I3
I4
O1 O2 O3 O4
FMEA
Identify Process Capability
LSL USL
Cp = 0.4
s = 2.7
Measure
the
process
I P O
Input
Measures
Process
Measures
Output
Measures

14. Analyze

15. Analyze
.
VA NVA
Process Door
Regression Analysis
Chi-Square
c²
Regression
t-test
ANOVA
X1
Y
Hypothesis-Testing Design of Experiments
.
Cause & Effect
Data Door
22
21
20
19
18
17
16
15
14
13
12
1 2 3 4 5 6 7 8 9 10
X
O
n
X
O
n
X
O
n
X
O
n
X
O
n
X
O
n
X
O
n
X
O
n
O
n
X
O
n
.

16. Improve

17. Perform
Cost-
Benefit
Analysis
Generate Solutions
A
B
C
D
4
1
3
2
Assess Risks
Run Pilot
Test
Full scale
Original
2 4 86 10
G
1 3 5 7 9
A
B
C
D
FE
JIH
G
Plan
Implementation
Select the Solution
Improve
FMEA

18. Control

19. Control
Evaluate Project
Results
.
UCL
LCL
Ownership &
Monitoring
Before After
Step 4 changes
implemented
} Improvement
Target
} Remaining Gap
Good
}Improvement
Before After
A1 A2 A3 A4 A2 A1 A3 A4
Process Change
Management
Learnings
Recommendations
Results
•
•
•
next
Key Learnings
QC Process Chart
Work
Instructions
Control/Check Points Response to Abnormality
NotesCode # Charac-
teristics
Control
LimitsMethodWho
Immediate
Fix
Permanent
Fix WhoFlowchart
2
12
Product Name
Process Name
Process Code #
Date of Issue: Issued by: Approved by:
Revision Date Reason Signature
1
Document &
Standardize
Training
Curriculum
Training
Manual
Fill to here
.
Closure
LSL USL
s = 3.7
Cp = 1.4
s = 2.7
Cp = 0.4
Process
Owner

20. A Philosophy ?
• Identify the process and customers right up front.
• Time spent identifying root causes and not just symptoms
is time well spent. D, M, A, then I !!
• “Show me the Data !” (valid data please)
• Great ideas with poor support will fail.
• Put good people in a bad process and the process will
win every time.
• Y = f (x)

21. An Illustrative Project Example
PROCESSES
TOOLS
SKILLS
TRAINING
LEAN SIGMA
(DMAIC +)
Integrated Improvement
Y1
y1
VOICE OF...
• Market
Customer•
Employee•
• Business
FEEDBACK
CORE & ENABLING PROCESSES
PROCESS
MAPS SYSTEMS
EXECUTION (PROCESS MANAGEMENT)
WORKOUT
SIX SIGMA
LEAN SIGMA
STRATEGY
If new product
or process
Big Y’s
Sub Y’s
PROCESS
DFSS(DMADV)
Fundamental Redesign
D
R
I
V
E
S
S
U
P
P
O
R
T
S
Flexible Problem Solving Models
Y1
y1
VOICE OF...
• Market
• Customer
• Employee
• Business
BUSINESS
OBJECTIVES
RESULTS:
Top-Level
Indicators
(Dashboards)
PROCESS
MAPS
SYSTEMS
PROCESS IMPROVEMENT
STRATEGY
If new product
or process
Projects
PROCESS
DFSS
D
R
I
V
E
S
S
U
P
P
O
R
T
S
PROCESS
CONTROL
ALIGNMENT
SIX SIGMA
(DMAIC)
Incremental Improvement
GE WORKOUT
Quick Wins
Accelerated Improvement
The power of the Lean Tools &
Principles fully integrated into
DMAIC & DFSS
The power of the Lean Tools &
Principles fully integrated into
DMAIC & DFSS

22. Problem Statement:
In the past 3 years, Calcium in product has averaged at
11.3 mg/50 cal, which is below label claim of 11.5 mg/50
cal. Disposal of product due to low Calcium was $47,000
in 2003 and $15,000 in January 2004.
Goal Statement:
Calcium in product shall be at label claim upon product
release. Disposal of product due to low Calcium shall be
zero after improvement to the process is made (by July
2004).
Calcium Recovery Improvement

23. %Contribution
%Study Var
%Tolerance
Gage R&R Part-to-Part
0
50
100
Components of Variation
Percent
Gage R&R (ANOVA) f or Ca (mg/100 c
Measurement System Analysis
Acceptable Gage on Project Metric = 27.48 %
Number of distinct categories = 5
Method to Validate Measurement System:
• 1 operator
• 15 samples
• 3x repeated
• Method used: Calcium determination by Inductively Coupled Argon
Plasma emission spectrophotometer
Results: Gage R&R (ANOVA) for CA (mg/100 c)

24. 24 ©2005 Tyg Lucas
Y = Calcium in mg/50 cal on product
Measure: Baseline Performance
453525155
12.0
11.0
10.0
Observation
Calcium
Run Chart for Calcium
Mean is 98.5% of Target

25. ©2005 Tyg Lucas
10.0 12.0 14.0 16.0 18.0 20.0 22.0 24.0
LSL USL
Process Capability Analysis for Calcium
Within
Overall
Average = 11.3 mg/50 cal Cpk = 0.49
Target = 11.5 mg/50 cal S.D. = 6.49
Process sigma = 2.8
DPMO = 96,801 ( approximately 10% probability of failure)
Measure: Baseline Performance

26. ©2005 Tyg Lucas
Measure
The team had a several page detailed process flow map

27. Analyze: Prioritization Matrix
Initial: 20 potential X’s
Prioritize to 7 potential X’s
Output Variable Calcium
content
agitator speed of f.p. tank 7
hold time of f.p. tank 9
liquid level of f.p. tank 5
finished product viscosity 7
product pH 3
calcium in f.p. tank 9
ProcessVariables
raw materials lot to lot
variation
5
Input
Variable

28. 28 ©2005 Tyg Lucas
Analyze: Potential X’s
Y = mg/50 cal Calcium in product
X1 = Product pH
X2 = Product viscosity
X3 = Raw material lot to lot variation
X4 = Calcium in finished product tank (=bulk liquid)
X5 = Hold time in finished product tank
X6 = Agitator speed in finished product tank
X7 = Liquid level in finished product tank

29. ©2005 Tyg Lucas
Analyze: Data Analysis
Calcium
(mg/50 cal)
pH of finished product
X1 = product pH
No correlation between Calcium and product pH
3.0 +
- x
- x
- x 2
- x
12.0+ x 3
- x x x 2
- 3 2
- x x x 3 x
- 4 x
11.0+ 2 5
- x x
- 2
- x x
- 2
10.0+ x
----+---------+---------+---------+---------+----
5.55 5.70 5.85 6.00 6.15

30. ©2005 Tyg Lucas
Analyze: Data Analysis
Calcium
(mg/50 cal)
Product viscosity (cPs)
X2 = product viscosity
-
13.5+
-
-
- x
- x x x
12.0+ x x x x x
- x x x 3 x 2
- x x 2 2 x x x
- 2 2 3 2
- x x x 2
10.5+ x x
- x x 2
- x
+---------+---------+---------+---------+---------+
0.0 4.0 8.0 12.0 16.0 20.0
No correlation between Calcium and product viscosity

31. ©2005 Tyg Lucas
X4 = Calcium in finished product tank (bulk liquid)
-
13.5+
-
-
-
- x
12.0+ x x x
- x x x x
- x2 x
- x x
- x
10.5+ x
- x x
- x
------+---------+---------+---------+---------+------
10.4 11.2 12.0 12.8 13.6
Calcium in
product
(mg/50 cal)
Calcium in bulk liquid (mg/50 cal)
Some correlation between Calcium in product
and Calcium in bulk liquid
Analyze: Data Analysis

32. ©2005 Tyg Lucas
Calcium
in product
(mg/50 cal)
hold time (hrs)
0 10 20 30 40
10.0
10.5
11.0
11.5
12.0
12.5
O = B01
+ = B53
x = B48
X5 = Hold time in finished product tank
Some correlation between Calcium in product
and hold time in finished product tank
Analyze: Data Analysis

33. ©2005 Tyg Lucas
Analyze: Critical X’s
Y =f (X4, X5, X6, X7, X8)
X4 = Calcium in finished product tank
X5 = Hold time in finished product tank
X6 = Agitator speed in finished product tank
X7 = Liquid level in finished product tank
X8 = Agitator on-off

34. ©2005 Tyg Lucas
0 10 20 30 40
10.0
10.5
11.0
11.5
12.0
12.5
ti i h f C 1
Calcium
in product
(mg/50 cal)
hold time (hrs)
O = B01-15 h
+ = B53-17 h
x = B48-29.5 h
X5 = Hold time in finished product tank
Analyze: Validate X’s
filler starts here
filler starts here
filler starts here
Hold time :

35. ©2005 Tyg Lucas
Improve: Pilot
0.0
0.5
6.0
12.0
16.5
17.5
18.5
19.5
22.0
0 10 20
0
.50
10.0
hold time-hr
Calcium(mg/50cal)
124
122
120
123 123 124 123 122
119
processed March 10, 04; packaged March 11, 04
Batch size: 4561 gals
Filling time: 5.5 hrs
(stopped for 45 mins)
Conclusion: continuous agitation helps keep Calcium
in suspension in the finished product tank
X8 = Agitator on-off
filler starts here
Results:

36. ©2005 Tyg Lucas
Improve: Risk Analysis
Potential Risks:
• Emulsion stability of the product might be negatively
impacted
• Greater vitamin C degradation
Risk Abatement Plan:
• Submitted Change Management Plan proposal and
discussed change on April 21,2004
• Analyzed stability samples for emulsion stability and
vitamin C degradation at zero and 1 month

37. ©2005 Tyg Lucas
Control & Monitoring Plans
Control Plan:
• Manufacturing Batch Records will be modified to
include change in agitation schedule from 30 minutes
on-off to continuous
Monitoring Plan:
• Routine QC practice will be utilized (each batch
produced is tested for Calcium)
• Pull out data 3 months from now and review results

38. ©2005 Tyg Lucas
Project Summary
Lessons Learned:
• Validating measurement system using Gage R&R is critical to
understanding the source of variation in the data.
• Drilling down the potential X’s and analyzing the data are keys to
finding the right solution
• Team members participation and involvement are key success
factors in any given projects
• Continuous agitation in tanks 20–23 helps keep Calcium in
suspension. Agitation schedule for other products stored in these
tanks needs to be re-evaluated
• Future/on-going projects that require products to be run at small
batch size (for example:Batjuice) should avoid using tanks with poor
agitation design (tanks 20–23)
Additional Project Opportunities:
• Product B+

39. ©2005 Tyg Lucas
But Does it Work in IT/Systems ?
• Do you have processes ?
The IT organization at Raytheon Aircraft saved $500,000 from a
single project in 2002.
The nine CIOs at Textron saved a total of $5 million in six months.
One team of engineers at Fidelity Wide Processing expects to
deliver $6 million to $8 million in cost reductions this year.
— Tracy Mayor
The CIO Service Center

40. ©2005 Tyg Lucas
What’s Next ?
• More Six Sigma
• Design for Six Sigma
• Lean Six Sigma