2. 6 Sigma?
“Project oriented and statistical
fact based process improvement
methodology driven by Business
and customer expectations”
3. Quiz / Icebreaker
Choices we all make:
You have two routes to choose from when driving
to work…
I-635I-635 NorthWestNorthWest
HighwayHighway
35Min. Avg. 40Min. Avg.
WhichRouteDo YouTake?WhichRouteDo YouTake?
Or, Does it ReallyEvenMatter?Or, Does it ReallyEvenMatter?
4. Descriptive Statistics: I-635, NWH
Variable N Mean Median StDev
I-635 555 35.149 32.676 12.214
NWH 555 40.085 40.097 1.564
Sometimes it depends
onmorethanjust the
average!
So, which route would be easier to plan around?
Answer
Route 1 Route 2
20
30
40
50
60
70
80
90
C4
C5
I-635 NWH
If I take I-635, I could possibly make it to work in 25 mins, but
there is a high probability of a disabled vehicle, an accident or
two, weather or other variables that cause a lot of variation
around when/if I actually arrive.
On Northwest Highway there is typically a constant, slow stream
of traffic that trudges along each day and is not greatly
disrupted by changes in construction or accidents, since they
are worked around between lights anyway. (The exception is
heavy rain which invariably knocks out all the traffic lights!)
5. If you always leave
45 mins before you
need to be at work,
which route will
make you late more
often?
Answer
0 20 40 60 80 100
USLUSL
Process Capability Analysis for Route 1
USL
Target
LSL
Mean
Sample N
StDev (Within)
StDev (Overall)
Cp
CPU
CPL
Cpk
Cpm
Pp
PPU
PPL
Ppk
PPM < LSL
PPM > USL
PPM Total
PPM < LSL
PPM > USL
PPM Total
PPM < LSL
PPM > USL
PPM Total
45.0000
*
*
35.1491
555
11.8541
12.2197
*
0.28
*
0.28
*
*
0.27
*
0.27
*
196396.40
196396.40
*
202982.59
202982.59
*
210077.85
210077.85
Process Data
Potential (Within) Capability
Overall Capability Observed Performance Exp. "Within" Performance Exp. "Overall" Performance
Within
Overall
32 34 36 38 40 42 44 46
USLUSL
Process Capability Analysis for Route 2
USL
Target
LSL
Mean
Sample N
StDev (Within)
StDev (Overall)
Cp
CPU
CPL
Cpk
Cpm
45.0000
*
*
40.0846
555
1.61988
1.56463
*
1.01
*
1.01
*
Process Data
Potential (Within) Capability
Overall Capability Observed Performance Exp. "Within" Performance Exp. "Overall" Performa
Wi
Ove
Since NWH is more
predictable, it may
be the better choice
despite the higher
average time.
I-635
NWH
6. Practical ProblemPractical Problem
Overall Approach of 6-Sigma
Statistical ProblemStatistical Problem
…
Y
x x x
Statistical SolutionStatistical Solution
Y = f (X)
OUTPUT INPUTS
Y = f (X)
OUTPUT INPUTS
Practical SolutionPractical Solution
Approach for
problem solving
7. The DEFINE Phase
What is important?
Develop the Business
Case
Translate the Voice of the
Customer
High-level map the process
Create a Project
Charter
8. The MEASURE Phase
How are we doing?
Identify potential
causes
Develop a Data
Collection Plan
Verify the Measurement
System
Identify patterns in data Determine baseline
process capability
Count
PercentC4
Count
32.4 9.7 6.8 5.3 2.4
Cum % 43.5 75.8 85.5 92.3
90
97.6 100.0
67 20 14 11 5
Percent 43.5
OtherAEGBC
200
150
100
50
0
100
80
60
40
20
0
Cpk=0.36
Yield=86.2%
Process
Sigma=2.6
9. The ANALYZE Phase,
What is wrong? The finding the problem’s root causes.
Analyze the process
Visualize causes X
Discrete Continuous
Y
Discret
e Chi2 Logistic
Regression
Cont-
inuous
t-Test
F-Test
ANOVA
DOE
Regression
Verify and quantify causes
X
Y
1514131211109876
60
55
50
45
40
Scatterplot of Y vs X
12.612.011.410.810.29.69.08.4
A
B
Process
Analysis
Data
Analysis
10. The IMPROVE Phase,
What needs to be done?
Identify potential solutions Test and verify potential solutions
Select solution(s) Mitigate risk Implement solution
Process
Step/Part
Number
Potential Failure Mode Potential Failure Effects
S
E
V
Potential Causes
O
C
C
Current Controls
D
E
T
RP
N
1 xxxxxxxxxxxxxx xxxxxxxxxxxxxx
6
xxxxxxxxxxxxxx
3
xxxxxxxxxxxxxx
1 18
2 xxxxxxxxxxxxxxxxx xxxxxxxxx
4
xxxxxxxxxxxxxx
3
xxx
2 24
3 xxxxxxxxx xxxxx
5
xxxxxxxx
4
xxxxxxxxx
5 100
4 xxxxx xxxxxxxxxxxxxx
9
xxxxxxxxxx
1
xxxxx
9 81
5 xxxxxxxxxxxxxx xxxxxxxx
1
xxxxxxxxxxxxxx
1
xxxxxxxxxxxxxx
3 3
6 xxxxxxxx xxxxxxxxxx
2
xxxxxxxx
9
xxxxxxxx
4 72
7 xxxxxxxxxx xxxxxxxx xxxxxxxxxx xxxxxxxxxx
0
12.612.011.410.810.29.69.08.4
Before
Pilot
P-value 0.007
11. The CONTROL Phase,
How do we guarantee performance?
Close project and
celebrate
Clarify Process Management Plan
and responsibilities
Establish process
monitoring
Standardize
process
Document key learnings and
identify transfer opportunities
Evaluate
results
5045403530252015105
55
50
45
40
35
_
X=45.09
UCL=52.95
LCL=37.22
726456484032241681
30
25
20
15
10
_
X=16.40
UCL=20.91
LCL=11.90
Before After
1
29252117139
16
12
8
4
0
USL
Within
CCpk 0.238
Cp *
CPL *
CPU 0.238
Cpk 0.238
29252117139
24
18
12
6
0
USL
Within
CCpk 0.988
Cp *
CPL *
CPU 0.988
Cpk 0.988
Before
Aft er
12. NSN 6-Sigma Roles
Sponsor/Process Owner
6-Sigma Green Belt
6-Sigma Black Belt
6-Sigma
Master
Black Belt
Leads in project identification, prioritization & defining the project
scope. Reports the status to Executives.
Removes barriers for Belts and aligns resources.
Leads high level projects/programs
Leads deployment in own Business area
Coaches and mentors black belt and green belt.
Master 6-Sigma competence
Leading, executing, and completing large projects.
Mentor Green Belts.
Advanced 6-Sigma competence
Takes ownership projects
Maintaining the project’s gains.
Removes barriers for Belts.
Leading, executing and completing small/medium
Projects
Practical 6-Sigma competence
Deliverables
Clear problem statement including metrics
Business value of the project
Project plan (incl. milestones, team members)
Tools
Project Charter
High-Level Map of the Process (SIPOC map)
Voice of the Customer process
Business Case
Stakeholder Planning
Deliverables
Baseline capability
Executed data collection plan
Tools
Cause and Effect diagram (Fishbone)
Prioritization Matrix
FMEA
Gage R&R
Sample Size Calculation
Process Capability Calculation
Graphical tools (Frequency Plots, Time Series Plots, Pareto Charts)
Control Charts
The Analyze phase in a Six Sigma project is about finding the problem’s root causes.
The analysis includes process and data analysis tools.
In most projects Measure and Analyze are iterative not linear.
Deliverable
Verified/quantified root causes
Tools
Process Mapping
Cause and Effect Diagram
Relationship plots (Stratified Frequency Plots, Scatter Plots, Multi-vari Charts)
Hypothesis Tests
Regression Analysis
Design of Experiments
Deliverables
Evaluated project results
Established process monitoring system
Process handed over to Process Owner
Tools
Hypothesis Tests
Process Capability Calculation
Process Management Chart
Control Charts
Standardization