SVR 1 23/2/11 INTRODUCTION TO SIX SIGMA
Feb 08, 2016
SVR 123/2/11
INTRODUCTION TOSIX SIGMA
SVR 223/2/11
SIX SIGMA IS DEFINED AS
• A STRUCTURED , SYSTEMATIC , TEAM DRIVEN, DATA DRIVEN APPROACH TO ACHIEVING BUSINESS EXCELLENCE.
• SIX SIGMA IS A PHILOSOPHY
• WE CAN ALSO LOOK AT IT AS A SET OF TOOLS AND TECHNIQUES FOR ACHIEVING BUSINESS RESULTS.
WHAT IS SIX SIGMA?
SVR 323/2/11
Six Sigma – a way of life for people who may not have even heard of management tools
No of defects : 1 in 8 million transactions, Sigma level : 7 sigma
Dabbas are collected from customers' homes
At the nearest railway station, they are sorted according to the destination station and are loaded in crates onto trains.
At the destination station they are unloaded by other dabbawalas and re-
sorted. The 100- kg crates of carriers are delivered, picked up at 1:30 p.m., and
returned whence they came
7-8 am 10:15 – 10:45 am 12:00 to 12:30 pm
SVR 423/2/11
They do it with codes and colours
Collection Point
Andheri
LokhandwalaComplex
Destination Details
Nirmal Bldg
11th Floor
Nariman Point
Dabbawala identification at destination
SVR 523/2/11
What does this tell us Attaining six-sigma is not about usage of statistics but about
• discipline, • customer focus,• simplicity, • robust process, • robust system for managing the process • and belief in the process
But even if a company operated in Six Sigma fashion if its overall strategy is weak or the market environment is changing, Six Sigma might be less successful if not linked to the business needs.
It is about getting our Corporate Strategy & Operational Excellence right.
SVR 623/2/11
Six Sigma can also be looked at as1. Six Sigma as a culture of continuous improvement
• Improving business performance through greater customer satisfaction, profitability & competitiveness by process improvements
• it is about learning to build processes that deliver flawless quality of output continuously
2. Six Sigma as a stretch goal• Improvement in performance 10 X every 2 years• Achieving 3.4 defects per million
SVR 723/2/11
TO REDUCE VARIATION TO REDUCE LEAD TIME TO REDUCE DEFECTS/ ERRORs TO IMPROVE YIELD TO ENHANCE CUSTOMER SATISFACTION TO IMPROVE THE BOTTOMLINE
THE GOALS OF SIX SIGMATHE GOALS OF SIX SIGMA
SVR 823/2/11
TWO MEASURES OF A PROCESS
•Central Tendency
•Dispersion
SVR 923/2/11
Customer may never observe mean performance
But
Always feel variation around the mean
SVR 1023/2/11
We normally study and report Averages
The problem is - Average never happens
What Customer Feels is – Variation
And The Organization's performance depends on Variation
Don’t Look at Averages
SVR 1123/2/11
What we See and What We Get
Our Performance – Average - 20 Days
Delivery Commitment to Customer = 25 Days
SVR 1223/2/11
TargetValue
USL
CustomerDissatisfaction
What we See and What We Get
20 Days
25 Days
SVR 1323/2/11
What we See and What We Get
Our Performance – Average = 25.0 kgs
Bag Weight = 25 Kgs
SVR 1423/2/11
TargetValue
USL
Loss
What we See and What We Get
25 Kgs
24.8 Kgs 25.2 Kgs
CustomerDissatisfaction
SVR 1523/2/11
CenterProcess
ReduceSpread
Off-Target Too Much Variation
Centered On-Target
The Statistical Objective of Six Sigma
SVR 1623/2/11
The Normal Curve
2 345 6
68.26 %
95.46 %
99.73 %99.9937 %
99.999943 %99.9999998 %
2 345
-6 -5 -4 -3 -2 +1 +2-1 +3 +4 +5 +6
SVR 1723/2/11
Six Sigma Statistically
Statistically Six Sigma is the area covered by the specification limits is 6 standard deviations on either side from mean for a normal distribution.
SVR 1823/2/11
3 Sigma Process
2 345 6
68.26 %
95.46 %
99.73 %99.9937 %
99.999943 %99.9999998 %
2 345
-6 -5 -4 -3 -2 +1 +2-1 +3 +4 +5 +6
LSL USL
Rej. Rej.
SVR 1923/2/11
3
6
99.73 %
99.9999998 %
3
-6 -3-2
+3 +6
LSL USL
Rej. Rej.
Six Sigma Process
SVR 2023/2/11
SIGMA RATING & PPM ARE CORRELATEDSIGMA RATING & PPM ARE CORRELATED
Sigma Rating PPM2 308,5373 66,8074 6,2105 2336 3.4
Process Defects per Capability Million Opportunities
SIGMA RATING A PERFORMANCE MEASURE
SVR 2123/2/11
SHOULD ALL PROCESSES BE AT SIX SIGMA?
The airline fatality rate is about 6.5 sigma, but airline baggage handling is about 4 sigma
SVR 2223/2/11
SIX SIGMA PROJECT - STEPS
There are 5 fundamental steps involved in applying the breakthrough strategy for achieving Six Sigma.
» Define Define» Measure Measure» Analyze Analyse» Improve Design» Control ValidateDMAIC DMAD
V
SVR 2323/2/11
The Six Sigma Process
Measure Analyze
Practical Problem Statistical Problem
Practical Solution Statistical Solution
Control Improve
SVR 2423/2/11
Objective of Six Sigma
Y = f (X)
SVR 2523/2/11
MAIC Funnel
Process Map, C & E Matrix, FMEA, MSA, Capability Study
Multi-Vari, Correlation/
Regression Hypothesis
Tests, ANOVA
Multi-Factor ANOVA,
2k Factorial Expts.,
Fractional Factorials
Control Plans, SPC,
Mistake Proofing
3 – 6Critical x’s
4 – 8Critical x’s
8 – 15 x’s
30 – 50 Inputs
Analyse
Measure
Improve
Control
SVR 2623/2/11
METHODOLOGY
Identify Six Sigma Projects
Carryout Six Sigma Projects
Integration with Business Planning Process
Institutionalization
Management Team Buy-in
Creating Organization for Six Sigma
Training People on Six Sigma
SVR 2723/2/11
Basic Tenets of Six Sigma
•Customer Orientation
•Process Orientation
•Speak with facts
SVR 2823/2/11CWCM / SS / ABCTL Chloro Alkali / 2 Oct
Speak with Facts
•If we can’t express something in numbers, we don’t know much about it.
•If we don’t know much about it, we can’t control it
•If we can’t control it, we are at the mercy of chance.
Let us talk with facts because;
SVR 2923/2/11
Where Six Sigma does not work
Poorly defined problems
No clear measurement metrics
Non availability of data/development actions
One off assignments/ no process exists
No team assigned or No Review mechanism planned
Not linked to Reward and Recognition system
Some softer reasons: Low on priority, focus, support, guidance etc.
SVR 3023/2/11
Roadmap to Six Sigma
SVR 3123/2/11
Identify All Possible CausesDefine Performance Parameter ‘Y’
Validate Measurement System for ‘Y’
Establish Baseline Process CapabilityDefine Performance ObjectivesIdentify Variation Sources
Establish Cause & Effect RelationshipDevelop Feasible SolutionsValidate & Implement Solutions
Step 4 Step 5Step 6 Step 7Step 8Step 9
Step 10Step 11Step 12
Develop & Implement Control Plans Monitor Process for SustenanceInstitutionalize Improvements
Step 13Step 14 Step 15
Step 1 Step 2Step 3
Identify Improvement OpportunityDevelop Team CharterDefine Process Map
The DMAIC Road Map
Define
Measure
Analyze
Improve
Control
SVR 3223/2/11
Step 1 - Identify Improvement Opportunities
• Identify Project Ys & Link it to Big Y
Step 2 - Develop Team Charter
• Develop The Business Case• Develop The Problem and Goal Statement• Determine Project Scope• Select Team & Define Roles• Set Project Milestones
Step 3 - Define Process Map
• Process Map Connecting Customer to Process
Define Phase Steps
SVR 3323/2/11
• Project Charter• CTQ Matrix• SIPOC• QFD
Tools Commonly Used during define Phase
SVR 3423/2/11
PROCESS CTQKPOVKPIV
KPIV
DEFINE Clearly state the Problem in terms of “Defects” or “Variation” which Upset the Customer
Define Phase - Goal
KPIV
SVR 3523/2/11
Measure Phase - Steps
Step 4 – Identify All Possible Causes
• Identify Measurable Y and All Possible Causes of Variation (Xs)
Step 5 - Define Performance Parameter
• Design the Measurement System• Determine & Confirm Specifications Limits For “Y”
Step 6 – Validate Measurement System for ‘Y”
• Confirm Adequacy for Measurement System for “Y”
SVR 3623/2/11
1. Data collection Plan/Check Sheets2. Stratification (Multi Vari Analysis)3. Control Charts4. Frequency Plot/Normal Probability Plot (Six Pack)5. Gage R & R6. Pareto Chart7. Prioritization matrix8. FMEA/Risk Analysis9. Cp, Cpk, Pp, Ppk (Process Capability - Normal, Binomial or Poisson)
Tools Commonly Used during Measure Phase
SVR 3723/2/11
Control Charts
1,1
1,2
1,3
1,4
1,5
1,6
1,7
1,8
1 3 5 7 9 11 13 15
Number
Wal
l thi
ckne
ss (m
m)
Actual Reading
UCL 1,69
LCL 1,19
Histograms
34 4 4
5 5
8
4
100
1
2
3
4
5
6
7
8
9
Number of sick days taken per emplyee
0
5
10
15
20
25
30
0 50 100 150
Cusomer size (KUSD)
Def
ect p
rodu
cts
(pcs
)
Scatter Diagram
Check SheetsCustomer October
Complaints Week 1 Week 2 Week 3 Week 4 TotalError onstatement
IIII IIIIIIII IIII I
IIII IIII IIIIIIII IIII
IIII IIII IIIIIIII IIII I
IIII IIIIIIII IIII I
92
Latestatement
IIII IIII IIII I IIII IIII II 28
Products notoperated
IIII IIII IIIIIIII IIII IIIIIIII
IIII IIII IIII IIII IIII IIIIIIII IIII IIIIIIII II
IIII IIII IIIIIIII IIII IIII
117
Total 66 45 68 58 248
Charts and Graphs
0
50
100
1987 1988 1989 1990 1991
7 Quality Tools7 Quality Tools
Charts and GraphsCheck SheetsPareto ChartsCause and Effect (Ishikawa/Fishbone) DiagramsScatter DiagramsHistogramsControl Charts Pareto Charts
050
100150200250300350400450
Ordershipped
late
Wrongparts
shipped
Brokenparts
Wrongorder
Rudeservice
Clericalerrors
Other Total
Num
ber o
f cus
tom
er c
ompl
aint
(J
une
- Dec
embe
r)
0102030405060708090100
%
Cause and Effect Diagram(Ishikawa/Fishbone)
Problem
SVR 3823/2/11
MEASURMENTS
Run 1
Run 2
Run 3
Run 4
Run 5
Run 6
Run 7
IF WE CAN’T EXPRESS WHAT WE KNOW IN THE FORM OF NUMBERS,WE REALLY DON’T KNOW MUCH ABOUT IT
Dr Mikel Harry
LET DATA BE THE BASE TO OUR ACTIONS!LET DATA BE THE BASE TO OUR ACTIONS!
Importance of Measurement
SVR 3923/2/11
PROCESS CTQ
KPIV
KPOVKPIV
KPIV
KPIV
MEASUREDescribe the Performance of the process Quantitatively and Identify Key Inputs
Measure Phase - Goal
SVR 4023/2/11
Analyze Phase - Steps
Step 7 - Establish Baseline Process Capability
• Establish Current Performance (9-12 Months Data)
Step 8 - Define Performance Objective
• Statistically Define the Goal of the Project
Step 9 - Identify Variation Sources
• Identify Key Xs which are expected to influence Y
SVR 4123/2/11
1. Brainstorming (Why - Why Analysis)2. Cause and Effect Diagram3. Control Charts4. Multi vari Analysis5. Test of hypothesis6. Design and Analysis of Experiments7. Scatter Plots/Regression Analysis
Tools Commonly Used During Analyze Phase
SVR 4223/2/11
PROCESS CTQ
KPIV
KPOVKPIV
KPIV
KPIVAnalyzeStatistically Describe the Relationship BetweenINPUTS and the OUTPUTSFind the Inputs with the BIGGEST IMPACT on the OUTPUT
Analyze Phase - Goal
SVR 4323/2/11
Improve Phase Steps
Step 10 – Establish Cause & Effect Relationship
• Determine Vital few Xs which Cause Change in Y• Establish Relationship Y v/s Vital Xs
Step 11 – Develop Feasible Solutions
• Determine Optimal Conditions• Develop Counter Measures and Solutions to achieve desired
state• Carry out Risk Analysis
Step 12 – Validate & Implement Solutions
• Pilot the Solution & Implement
SVR 4423/2/11
1. Consensus2. Creativity3. Data Collection/Check Sheets4. DOE5. FMEA6. Hypothesis testing7. Multi-Vari analysis8. Process capability Evaluation
Tools commonly Used during Improve Phase
SVR 4523/2/11
PROCESS CTQ
KPIV
KPOVKPIV
KPIV
KPIV
ImproveSystematically EXPERIMENT with the INPUTS to Find the Combination Which Delivers the OPTIMAL OUTPUT
High
High
Low
Medium
Improve Phase - Goal
SVR 4623/2/11
Control Phase Steps
Step 13 – Development & Implement Control Plans
• Implement Control Plans
Step 14 – Monitor Process for Sustenance
• Monitor Process for 3 Months to ensure sustenance• Determine Capability & Performance after Improvement• Ensure there is no adverse effect
Step 15 – Institutionalize Improvements
• Communicate Result • Link changes with the Quality & Operational Systems
SVR 4723/2/11
1. Control Charts2. Data Collection/Check Sheet3. Charts to Compare Before & After4. Process Capability Evaluation5. Control Plans6. SOP’s/WI’s/Operator Training7. Poka-Yoke/Fool Proofing
The Tools Commonly Used in control Phase
SVR 4823/2/11
PROCESS CTQ
KPIV
KPOVKPIVKPIV
KPIV
Lock In the INPUTS to Routinely generatethe OPTIMAL OUTPUT
Control Phase - Goal
SVR 4923/2/11
PROCESS CTQ
KPIV
KPOVKPIV
KPIV
KPIV
1. Statistically proven Relationship Between Inputs & Outputs
2. Systematic (Improvement & Control)
What makes Six Sigma Different
SVR 5023/2/11
Target
USLLSL
CenteredProcess
ReducedSpread
Defects
The Statistical Objective of Six Sigma
Reduce Variation & Center Process—Customers feel the variation more than the mean
Process Off Target Excessive VariationTarget
USLLSL
Target
USLLSL
SVR 5123/2/11
End Of Introduction