![Sigma level vs. Defect
DPMO = [ D ( N * O ) ] * 1 Million Sigma DPMO
2 308770
where: 2.25 226716
– D = total number of defects 2.5 158687
counted in the sample – 2.75 105660
3 66811
a defect is defined as failure to meet 3.25 40060
a Critical Customer 3.5 22750
Requirement or CCR 3.75 12225
– N = number of units of product or 4 6210
4.25 2980
service 4.5 1350
– O = number of opportunities per 4.75 577
unit of product or service for 5 233
5.25 88
a customer defect to occur 5.5 32
– M = million 5.75 11
6 3.4
M. S. Ramaiah School of Advanced Studies 18](https://image.slidesharecdn.com/cjb0412001-rtse-esd2525-presentation-130220052051-phpapp01/75/Six-Sigma-and-For-Software-Engineering-18-2048.jpg)
![Measure – Measure The Process Performance
DPMO – For attribute data
Total number Defects- UI D 283
SR’s caught internally in
R1.0 in different phases(
code review, Unit Test,
Design)
Number of Units N 11
processed- Modules
Number of defect O 164
opportunities- Test cases
DPMO = 1M *[D/(N*O)] = 1000000*[283/(11*164)] = 156874
Current Sigma level = 2.51
M. S. Ramaiah School of Advanced Studies 41](https://image.slidesharecdn.com/cjb0412001-rtse-esd2525-presentation-130220052051-phpapp01/75/Six-Sigma-and-For-Software-Engineering-41-2048.jpg)
![References
[1] Pyzdek,T., The Six Sigma Hand Book, Revised and
Expanded,New Delhi:McGraw-Hill
[2] CIC2000: Six Sigma Green Belt Training, Motorola
University
[3] Benbow,W.D.,Kubiak,T.M.,The Certified Six Sigma
Blackbelt Handbook,2005,USA,ASQ Quality Press
M. S. Ramaiah School of Advanced Studies 70](https://image.slidesharecdn.com/cjb0412001-rtse-esd2525-presentation-130220052051-phpapp01/75/Six-Sigma-and-For-Software-Engineering-70-2048.jpg)
Uploaded byAnshuman Biswal
Six Sigma and/For Software Engineering
The document presented information on Six Sigma and its application to software engineering. It began with an outline of the topics to be covered, including definitions of Six Sigma, its history and evolution, key statistical concepts, and the DMAIC methodologies. It then discussed Six Sigma as a metric, philosophy, and management system. Specific examples of how Six Sigma was implemented at Motorola in the 1980s-1990s were provided to help drive defect reduction and process improvement through statistical analysis.
More Related Content
![Chorus - Distributed Operating System [ case study ]](https://cdn.slidesharecdn.com/ss_thumbnails/chorus-171117082036-181113144303-thumbnail.jpg?width=640&height=640&fit=bounds)
Similar to Six Sigma and/For Software Engineering
Six Sigma and/For Software Engineering
- 1. CSN2501 RTSE Presentation SixSigma and/for Software Engineering Anshuman Biswal PT 2012 Batch, Reg. No.: CJB0412001 M. Sc. (Engg.) in Computer Science and Networking Module Leader: N. D. Gangadhar Module Name: Real-Time Software Engineering Module Code : ESD2525 M. S. Ramaiah School of Advanced Studies 1
- 2. Marking Head Maximum Score Technical Content 10 Grasp and Understanding 10 Delivery – Technical and 10 General Aspects Handling Questions 10 Total 40 M. S. Ramaiah School of Advanced Studies 2
- 3. Presentation Outline • What is Six Sigma ? • Evolution of Concepts behind Six Sigma • Six Sigma at Motorola • History of Six Sigma • Statistics behind Six Sigma • What is Six Sigma Performance? • Relationship between Sigma Level and defect • Six Sigma methodologies • Define • Measure • Analyze • Improve • Control • Conclusion • References M. S. Ramaiah School of Advanced Studies 3
- 4. What is SixSigma A Metric ? A A Philosophy What is Methodology ? Six Sigma ? A Management System M. S. Ramaiah School of Advanced Studies 4
- 5. Six Sigma- AMetric A metric that encourages measurements of process performance. Sigma is the Greek letter representing a statistical unit of measurement that defines the standard deviation of a population. It measures the variability or spread of the data. 6 sigma is also the measure of variability. It’s a name given to indicate how much of the data falls within the customers requirements. The higher the process sigma, the more of the process outputs, products and services, meets customers requirements – or fewer the defects The term sigma is often used as the scale for levels of “goodness” or quality. Using this scale, “ Six Sigma” equates to 3.4 defects per million opportunities (DPMO). M. S. Ramaiah School of Advanced Studies 5
- 6. Six Sigma –A business methodology A methodology that focuses on the following Utilizing Driving rapid Understandin Aligning key rigorous data and g and business analysis to sustainable managing processes to minimize improvement customer achieve those variation in to business requirements requirements those processes processes M. S. Ramaiah School of Advanced Studies 6
- 7. Six Sigma –A management system Six Sigma is management system for executing business strategy. Six Sigma is a solution to help organizations to: Align their Govern efforts to business strategy Mobilize teams Accelerate ensure to critical to attack high improved business improvements are improvement impact projects results sustained efforts M. S. Ramaiah School of Advanced Studies 7
- 8. Six Sigma –A Philosophy Business Excellence Inferential Statistics Customer Focus Transition from Basic statistical intuition towards analysis inferential statistics in decision making Basic Charts and First time right at Graphs source Zero defect Brainstorming Tools Extra ordinary processes that deliver Intuition or Gut predictable results feeling M. S. Ramaiah School of Advanced Studies 8
- 9. Evolution of conceptsbehind Six Sigma FMEA was formally introduced in 1940’s for military usage by US armed forces. Later it was used for aerospace/rocket Ronal Fisher development. Example of introduced Design this is Apollo space In 1920’s , Walter A of Experiment program. The primary Carl Friedrich Shewart showed that push came during 1960’s 3 sigma, from the through a book in Gauss (1777-1855) ,while developing means mean is the point 1935. This was a introduced the to put a man on the moon where a process result of series of and return him back concept of Normal requires correction. studies that started safely. In 1970 Ford Curve This finally led to with study of introduced FMEA in Control Charts variation in crop automotive industry yield Automotive Industry Action Group ( AIAG) published the most accepted document on Measurement Systems Analysis (MSA). MSA is an essential step in 6 sigma methodologies and is used to ensure reliability of data. M. S. Ramaiah School of Advanced Studies 9
- 10. Six Sigma atMotorola In the late 1970’s Dr. Mikel Harry, a senior staff engineer at Motorola’s Government Electronics Group ( GEG),experimented problem solving through statistical analysis. Using this approach, GEG’s product were being designed and produced at a faster rate and at lower cost. Subsequently Dr Harry began too formulate a method for applying six sigma through out Motorola. M. S. Ramaiah School of Advanced Studies 10
- 11. Six Sigma atMotorola In 1987 when Bob Galvin was the chairman, Six Sigma was started as a methodology in Motorola. Bill Smith, an engineer and Dr. Mikel Harry together devised a 6 step methodology with the focus on defect reduction and improvement in yield through statistics. M. S. Ramaiah School of Advanced Studies 11
- 12. Six Sigma atMotorola The term Six Sigma was coined by Bill Smith, who is now called the Father of Six Sigma. Terms such as Black belt and Green belt were coined by Mikel Harry in relation to martial arts Using this methodology the company saved $16 billion in 10 years M. S. Ramaiah School of Advanced Studies 12
- 13. History of SixSigma 1987 Motorola Strength in Manufacturing 1994 Allied Signal Linked to Financial Returns General Electric Linked to Design and Service 1996 3M / Phillips Linked to Sales and Product Commercialization Sony, Seagate, Raytheon, Toshiba and many other 2000 companies 2006 Motorola Begins Lean Transformation Uses Lean concepts in 6 Sigma 2008 Motorola methodologies and termed it as Lean 6 sigma M. S. Ramaiah School of Advanced Studies 13
- 14. Statistics behind SixSigma pizza santas pizza hut 21 17 If two pizza delivery outlets have 9 18 same average delivery time of 20 11 21 minutes, against the promised 12 22 29 23 delivery time of 30 minutes then 14 16 would you say that they are equally 28 24 good? 26 23 20 19 24 18 22 21 23 16 23 21 7 19 29 22 average 19.8666667 20 M. S. Ramaiah School of Advanced Studies 14
- 15. Statistics behind SixSigma santas pizza pizza hut 21 17 9 18 To comment on which one is 11 21 better would you like to consider 12 22 variation in delivery time? If yes 29 23 how would you like to measure 14 16 28 24 variation? 26 23 20 19 24 18 22 21 23 16 23 21 7 19 29 22 2.618614 standard deviation 7.42454103 7 M. S. Ramaiah School of Advanced Studies 15
- 16. Statistics behind SixSigma santas pizza pizza hut 21 17 9 18 11 21 Can we find a single measure for 12 22 process performance that 29 23 14 16 incorporates the central tendency, 28 24 variation and specification limits? 26 23 20 19 24 18 22 21 23 16 23 21 7 19 29 22 average 19.8666667 20 standard deviation 7.42454103 2.6186147 Upper Specification Limit 30 30 sigma level 1.36484306 3.8188131 Probability 91.384881% 99.99330% M. S. Ramaiah School of Advanced Studies 16
- 17. What is SixSigma performance ? Your process is performing at sigma level of six if the difference between the mean and the specification limit is six times the standard deviation. To get a six sigma performance the variation should be as small as possible so that we can fit six standard deviations between the mean and the USL. M. S. Ramaiah School of Advanced Studies 17
- 18. Sigma level vs.Defect DPMO = [ D ( N * O ) ] * 1 Million Sigma DPMO 2 308770 where: 2.25 226716 – D = total number of defects 2.5 158687 counted in the sample – 2.75 105660 3 66811 a defect is defined as failure to meet 3.25 40060 a Critical Customer 3.5 22750 Requirement or CCR 3.75 12225 – N = number of units of product or 4 6210 4.25 2980 service 4.5 1350 – O = number of opportunities per 4.75 577 unit of product or service for 5 233 5.25 88 a customer defect to occur 5.5 32 – M = million 5.75 11 6 3.4 M. S. Ramaiah School of Advanced Studies 18
- 19. Six Sigma methodologies DMAIC Define – Measure – Analyze – Improve – Control Structured and repeated process improvement methodology Focuses on Defect reduction Improves existing products and processes DMADV Define – Measure – Analyze – Define – Verify Strict approach to design so as to exceed customer expectations Focuses on preventing errors and defects Develop new product/process or redesign existing product/process M. S. Ramaiah School of Advanced Studies 19
- 20. Six Sigma methodologies DMAIC Define – Measure – Analyze – Improve – Control Structured and repeated process improvement methodology Focuses on Defect reduction Improves existing products and processes DMADV Define – Measure – Analyze – Define – Verify Strict approach to design so as to exceed customer expectations Focuses on preventing errors and defects Develop new product/process or redesign existing product/process M. S. Ramaiah School of Advanced Studies 20
- 21. DMAIC and ProblemSolving Flow Define Measure Improve Control Analyze Opportunities Performance Performance Performance Opportunity (Establish (Characterize (Modify) (Ensure (Decompose) Requirements) Performance) Consistency) Practical Quantitative Quantifiable Practical Problem Problem Analysis Conclusion Conclusion statement Statement Key Questions to Ask How do we What is How are we What is What needs to guarantee important? doing? wrong? be done? performance? M. S. Ramaiah School of Advanced Studies 21
- 22. Define – Theimprovement Opportunity Select the Define and Develop the Map the improvement scope the team process opportunity project charter •Business leaders identify Six Sigma projects with Master Black Belts supporting the project portfolio process. • Black Belts, Green Belts and Six Sigma Teams with Sponsors & Champions are other sources of projects in the portfolio. • A continuous review of Scorecard Goals, dashboard metric results, and audit results suggest possible projects/opportunities for Improvement. M. S. Ramaiah School of Advanced Studies 22
- 23. Define – Theimprovement Opportunity Select the Define and Develop the Map the improvement scope the team process opportunity project charter Translating Voice of the Customer (VOCs) to Critical Customer Requirements (CCRs) Often the voices of customers (VOC) are not clear and not in technical language. Therefore VOC needs to be translated to the Critical Customer Requirements (CCR) which can be translated into the technical requirements known as Critical to Quality (CTQ), and Critical to Process (CTP) Tools: QFD/ House of Quality Input Process Output Measures M. S. Ramaiah School of Advanced Studies 23
- 24. Define – Theimprovement Opportunity Example of Input Process Output measures CTQ CCR CUSTOMER ISSUES VOC Defect 1> Defect back log Defect back log Higher Reduction needs to be cleared by increased so more staff number of June 2012. count were needed defects 2> Not more than 100 and also some defects found in XXX VOC: Voice Of Customer defects in XXXX r3.0 were pushed to next R1.0 CCR: Critical Customer Requirement 3> Try to attain virtual release by CTQ: Critical To Quality zero goal at the end of compromising with the release of the certain features that product. got pushed to next release BUSINESS VOB ISSUES CBR CTP Quality of To clear the Staff count can not Minimize product back log increase which can the rework needs to be more staff increase cost cost by improved i.e. counts reducing Defects needed and Identify defects in early the defects Backlog more effort phase before build going by 80% needs to be needs to be to BT cleared put. without Reduce number of defects increasing in Some ,so it reduces the rework staff count features had cost to be pushed to next release. VOB: Voice Of Business CBR: Critical Business Requirement CTP: Critical To Process M. S. Ramaiah School of Advanced Studies 24
- 25. Define – Theimprovement Opportunity Select the Define and Develop the Map the improvement scope the team process opportunity project charter Team Charter is a business document that communicates leadership expectations, expected business outcomes and personal relevancy of the project in order to focus and motivate team. Purpose of the Team Charter • Communicate leadership expectations, expected business outcomes and personal relevancy of the project in order to motivate and focus team. • Clarify the scope, resources and deliverables of improvement project. M. S. Ramaiah School of Advanced Studies 25
- 26. Define – Theimprovement Opportunity Project Title: XXX R3.0 Defect Reduction Business Case Opportunity Statement The number of defects ,raised by the Box Test team in the XXX R1.0 is In the past seven months ( April 2011 to November on the higher side resulting in massive rework and need of more 2011), which was the XXX R1.0 phase had a total of 532 resource working to clear the backlog. This accounts to USD 618K of BT SR out of which UI SR were 169, which was costs and out of which UI defects accounts to USD 196K. We need to approximately 80% greater than expected. reduce the number of defects by at least 80% in XXX R3.0 which will Improvement Opportunity is targeted at the front end help us to achieve virtual zero goal for future releases phase of the UI Dev ( Req., Design, Code and Unit test phases ) to achieve defect reduction from 169 to 32. Estimated Savings = USD 37K( Cost per defect taken as USD 1162). So a total of USD 138K can be saved if the UI defects is controlled to 32 numbers or less in future releases. Goal statement Project scope Reduce the BT defects against UI development in XXX R3.0 at least This project will involve review of main reasons for UI by 80% by November 2012. development bugs raised during Box test. It will not examine phase after the box test like system tests and also it will not examine the bugs raised in the stack and middle ware development. Project plan Team Selection Phase Start End Remarks Sponsor: Mr. X Define 4/30/2012 5/18/2012 Champion: Mr. Y Measure 5/21/2012 7/27/2012 MBB: Mr Z Analyze 7/30/ 2012 9/28/2012 MBB : Mr. A Black Belt Mentor : Mr. B Improve 10/1/2012 10/31/2012 Project Manager: Mr. C Project Lead: Mr. D Control 11/1/2012 11/30/2012 GB Candidate : Mr. E Team Members: Mr. F M. S. Ramaiah School of Advanced Studies 26
- 27. Define – Theimprovement Opportunity Select the Define and Develop the Map the improvement scope the team process opportunity project charter Process Mapping Purpose – Understand the relationships among inputs and outputs of a process – Identify non-value added activities so they can be eliminated Types of Process Maps – SIPOC – Top-down chart – Functional Deployment Chart – Value Stream Mapping – Spaghetti Diagram M. S. Ramaiah School of Advanced Studies 27
- 28. Define – Theimprovement Opportunity Example of SIPOC- Development Process SUPPLIER 1. INPUT OUTPUT CUSTOMERS 1. PDM Raw Requirement s Development Process CAB file Marketing Team Release 1. Notes Box Test Team 1. L3 Requirement Development Design Team Documents SUB PROCESSES Requirement Design Coding Testing (BT) Release Analysis Analysis Start Boundary: End Boundary: Requirement Gathering by PDM CAB file delivered to BT team M. S. Ramaiah School of Advanced Studies 28
- 29. Define – Theimprovement Opportunity TOP DOWN CHART Start Boundary: End Boundary: Requirement Gathering by PDM CAB file delivered to BT team Process: Software Development Requirement Design Coding Testing (BT) Release Analysis Analysis Design Write source code Collect architecture Develop test cases requirements from Write comments for Perform impact each source code Release test results field, customer, competitor analysis as per standards Develop test scripts product, external Prepare design Achieve in clear and automation demos doc case Perform code Prepare L3 Collect internal reviews document Review test cases and customer defects Review and Perform Unit Testing Raise defects if any Update L3 Req to fix in next build Perform Tests Review and Perform integration Update Testing Release L3 req. requirement spec, and Design doc Review test results UB Spec Prepare build M. S. Ramaiah School of Advanced Studies 29
- 30. Define – Theimprovement Opportunity Key Take Away INPUTS PROCESS OUTPUTS • Select the • Strategic priorities • Team charter improvement • Scorecard development • Project Plan opportunity • Core process selections • Prepared team • Define and scope the • Improvement • Critical to customer project expectations requirements • Develop the team • Improvement project • Process Map (As-Is) charter team sponsor, • Risk and mitigation plan • Map the process champion, team leader • Quick win opportunities • Identify quick wins M. S. Ramaiah School of Advanced Studies 30
- 31. Measure – MeasureThe Process Performance Create Collect Establish Select Define MSA- MSA- Measu and Baseline Measur Measu Variabl Attribu rement validat performan es res e Data te data plan e Data ce Example of Input Process Output Indicators Process Output Input Indicators Indicators Indicators Number of cust omer Number of FTR comment s on requirement s Document reviews Reduce t he number of SR Number of change requirement Availabilit y of resources Cust omer sat isfact ion Lines of Code Types of int ernal defect s Reduce t he Rework effort s Number of BT Sr's raised in R1.0 Project Schedule R1.0 in each module Planned Effort Act ual Effort s Number of design review comment s Number of code review comment s Act ual size of each module Number of int ernal Defect s M. S. Ramaiah School of Advanced Studies 31
- 32. Measure – MeasureThe Process Performance Cause and Effect Diagram M. S. Ramaiah School of Advanced Studies 32
- 33. Measure – MeasureThe Process Performance Cause and Effect Matrix High priority indicators for analysis a. Number of internal defects b. Types of internal defects c. Number of change requirements d. Number of BT Sr’s M. S. Ramaiah School of Advanced Studies 33
- 34. Measure – MeasureThe Process Performance Create Collect Establish Select Define MSA- MSA- Measu and Baseline Measur Measu Variabl Attribu rement validat performan es res e Data te data plan e Data ce Once the team knows what to measure, they need to further define the measurement. • This definition is called an operational definition. ( A operational definition is a precise description of What ? The specific criteria used for measurement How? The methodology to collect the data How much? The amount of data to be collected Who? Who has responsibility to collect data ) • The operation definition helps assure the data is: – Collected and measured in a consistent way – Representative of the process M. S. Ramaiah School of Advanced Studies 34
- 35. Measure – MeasureThe Process Performance Create Collect Establish Select Define MSA- MSA- Measu and Baseline Measur Measu Variabl Attribu rement validat performan es res e Data te data plan e Data ce •Determining current process performance usually requires the collection of data. When developing a measurement plan ensure that: – The data collected is meaningful – The data collected is valid – All relevant data is collected concurrently • Before data collection starts, classify the data into different types: continuous or discrete. This is important because it will: – Provide a choice of data display and analysis tools – Dictate sample size calculation – Provide performance or cause information – Determine the appropriate control chart to use – Determine the appropriate method for calculation of Sigma Level M. S. Ramaiah School of Advanced Studies 35
- 36. Measure – MeasureThe Process Performance Measurement Plan M. S. Ramaiah School of Advanced Studies 36
- 37. Measure – MeasureThe Process Performance Create Collect Establish Select Define MSA- MSA- Measu and Baseline Measur Measu Variabl Attribu rement validat performan es res e Data te data plan e Data ce What is MSA? The study of the extent to which systematic and random factors are affecting our ability to correctly measure some phenomenon. When MSA is done? Before data collection. For variable data use Gage R and R test, Bias and Linearity Test and Repeatability and reproducibility test For attribute data use Kappa analysis M. S. Ramaiah School of Advanced Studies 37
- 38. Measure – MeasureThe Process Performance Create Collect Establish Select Define MSA- MSA- Measu and Baseline Measur Measu Variabl Attribu rement validat performan es res e Data te data plan e Data ce P value is 0.095 which is greater than 0.05 hence we conclude that data is normal M. S. Ramaiah School of Advanced Studies 38
- 39. Measure – MeasureThe Process Performance Create Collect Establish Select Define MSA- MSA- Measu and Baseline Measur Measu Variabl Attribu rement validat performan es res e Data te data plan e Data ce What is Process Performance? • Process performance is defined as the ability of a process to produce outputs that meet engineering and/or customer specifications. • For continuous data we measure the process capability by calculating Cp and Cpk •For attributes data we calculate the process capability by DPMO method M. S. Ramaiah School of Advanced Studies 39
- 40. Measure – MeasureThe Process Performance Process capability analysis M. S. Ramaiah School of Advanced Studies 40
- 41. Measure – MeasureThe Process Performance DPMO – For attribute data Total number Defects- UI D 283 SR’s caught internally in R1.0 in different phases( code review, Unit Test, Design) Number of Units N 11 processed- Modules Number of defect O 164 opportunities- Test cases DPMO = 1M *[D/(N*O)] = 1000000*[283/(11*164)] = 156874 Current Sigma level = 2.51 M. S. Ramaiah School of Advanced Studies 41
- 42. Measure – MeasureThe Process Performance Key Take Away INPUTS PROCESS OUTPUTS • Team charter •Select Measures • Input, Output, and • Project Plan •Define Measures Process Measures • Prepared team •Create Measurement Plan • Operational Definitions • Critical to customer •Conduct MSA Collect and Validate Data • Data collection plan requirements •Establish Baseline • Capable measurement • Process Map (As-Is) Performance system • Risk and mitigation plan • Baseline Performance • Quick win opportunities M. S. Ramaiah School of Advanced Studies 42
- 43. Analyze – Analyzethe Opportunity Validate Root Map the process Focus on Vital Few Cause •During the measurement phase, the team learned about the critical x’s and collected measurements on these x’s. • During the analysis phase, we need to understand the root cause(s) of the variation affecting the critical x’s. – Requires a more detailed level of investigation. • A detailed process map helps the team identify all the x’s and use this as a guide to get to root cause(s). M. S. Ramaiah School of Advanced Studies 43
- 44. Analyze – Analyzethe Opportunity A detailed process map example M. S. Ramaiah School of Advanced Studies 44
- 45. Analyze – Analyzethe Opportunity A detailed process map example M. S. Ramaiah School of Advanced Studies 45
- 46. Analyze – Analyzethe Opportunity A detailed process map example M. S. Ramaiah School of Advanced Studies 46
- 47. Analyze – Analyzethe Opportunity A detailed process map example M. S. Ramaiah School of Advanced Studies 47
- 48. Analyze – Analyzethe Opportunity Validate Root Map the process Focus on Vital Few Cause Determine critical inputs – Cause and Effect Matrix – 5 Why’s – FMEA • Stratify data and narrow the focus – Pareto – Sources of Variation • Validate root cause statistically – Comparative Methods – Regression – DOE M. S. Ramaiah School of Advanced Studies 48
- 49. Analyze – Analyzethe Opportunity Pareto Chart M. S. Ramaiah School of Advanced Studies 49
- 50. Analyze – Analyzethe Opportunity 5 why M. S. Ramaiah School of Advanced Studies 50
- 51. Analyze – Analyzethe Opportunity Validate Root Map the process Focus on Vital Few Cause Comparative methods M. S. Ramaiah School of Advanced Studies 51
- 52. Analyze – Analyzethe Opportunity Regression Analysis : Scatter Plot interpretation M. S. Ramaiah School of Advanced Studies 52
- 53. Analyze – Analyzethe Opportunity Regression Analysis : Residual Plot M. S. Ramaiah School of Advanced Studies 53
- 54. Analyze – Analyzethe Opportunity Regression Analysis : Residual Plot Normality plot for the residuals. Points (residuals) falling along a reasonably straight line indicates that residuals are normally distributed. • Residuals versus fitted values showing no pattern, funnelling out, curvature, etc. indicating constant variance assumption is not violated. A curvature in this plot indicates that the model is not good and higher orders in the model may be needed. A funnelling out indicates variation is not constant and needs a transformation (square root, log, etc.) • Histogram is skewed to the left. However, this is not uncommon with a small number of observations. As long as the other three charts look okay, the skewing is probably acceptable. • Residuals versus order of the observation showing no pattern, funnelling out, curvature, etc. No obvious violation of the independence assumption. M. S. Ramaiah School of Advanced Studies 54
- 55. Analyze – Analyzethe Opportunity Regression Analysis M. S. Ramaiah School of Advanced Studies 55
- 56. Analyze – Analyzethe Opportunity Key Take Away INPUTS PROCESS OUTPUTS • Input, Output, and Process Measures • Develop Detailed Process • Data analysis • Operational Definitions Map • Process FMEA • Determine critical inputs • Data collection plan • Stratify the data • Sources of Variation • Capable measurement • Identify and Validate • Standardized Work system Root Causes • Validated root causes • Baseline Performance M. S. Ramaiah School of Advanced Studies 56
- 57. Improve – ImprovePerformance Identify and Full scale Select Best implementation Solution Brain Storming AFFINITY Diagram Solution Mapping M. S. Ramaiah School of Advanced Studies 57
- 58. Improve – ImprovePerformance Identify and Full scale Select Best implementation Solution Brainstorming : Problem1 : Long Approval Time Suggestions from Brainstorming Session: Use informal meeting procedure than formal for review meeting Not all test cases should be given for review to Test lead Peer review should be encouraged Define review frequency and time for final approval Use review tool to monitor and track review comments status Automatic assignment of test cases on completion using review tool Independent reviews by test lead Communicate review time to customer and hence include in project plan Provide training on using review tool and to perform reviews Improve review checklists and guidelines Approve test cases in batch and not all at once Communicate product critical review results to customer Review tool should email review results on review completion Improve Review tool interface for user friendly M. S. Ramaiah School of Advanced Studies 58
- 59. Improve – ImprovePerformance Affinity Diagram: Problem1 : Long Approval Time M. S. Ramaiah School of Advanced Studies 59
- 60. Improve – ImprovePerformance Affinity Diagram: Problem1 : Long Approval Time M. S. Ramaiah School of Advanced Studies 60
- 61. Improve – ImprovePerformance Solution Mapping M. S. Ramaiah School of Advanced Studies 61
- 62. Improve – ImprovePerformance Solution Selection Matrix M. S. Ramaiah School of Advanced Studies 62
- 63. Improve – ImprovePerformance Identify and Full scale Select Best implementation Solution using the best solutions change the process and draw the new process map Do a pilot solution plan Run the project and collect data again. Calculate the capability index and sigma level (DPMO) of the current process. Nevertheless the current process should be better than the previous process. Else do Analyze phase again and find for another set of root causes and repeat the improve phase. Keep on repeating this step until the new process is not improved . M. S. Ramaiah School of Advanced Studies 63
- 64. Improve – ImprovePerformance Key Take Away INPUTS PROCESS OUTPUTS • Data analysis • Solutions • Process FMEA • Identify and Select the • Process maps and • Sources of Variation Best Solution documentation • Standardized Work • Plan for Full-Scale • Cost/benefit analysis • Validated root causes Implementation • Improvement impacts and benefits M. S. Ramaiah School of Advanced Studies 64
- 65. Control – ControlPerformance Develop Evaluate Transfer integrated project Ownership control plan results A Control Plan documents the data needed to accurately measure the process on an ongoing basis and react to out of control processes. • Based on the work performed during the improve phase, your organization implemented your team’s recommendations and verified the impact they actually made on the process you targeted. • At this point, the objective is to sustain the gains (i.e., to make sure that the changes you helped to make become an integral part of the way your organization operates). • This is done by making sure we have a thorough control plan, which should address four different areas: error proof, measure and monitor, communicate, and document M. S. Ramaiah School of Advanced Studies 65
- 66. Control – ControlPerformance Develop Evaluate Transfer integrated project Ownership control plan results M. S. Ramaiah School of Advanced Studies 66
- 67. Control – ControlPerformance Develop Evaluate Transfer integrated project Ownership control plan results Checklist for Transition to Original Owner(s) • Has the original owner been made aware of the changes and the needed controls to maintain the process? • How will the owner monitor these improvements over time? • Is there a recommended audit plan for follow-up by the team / experts to assure that the process is still being maintained? • Have employees effected by the change been trained? • Did we have a good change plan? • Has the change been embraced by the owner and employees effected by the process? M. S. Ramaiah School of Advanced Studies 67
- 68. Control – ControlPerformance Key Take Away INPUTS PROCESS OUTPUTS •Develop a Detailed •Solution process maps and •Process Control Plan Control Plan documentation •Updated Standardized •Document final •Cost/benefit analysis Work implementation results •Improvement impacts and •Project Final Report / •Transfer ownership benefits Closure back to M. S. Ramaiah School of Advanced Studies 68
- 69. Conclusion • What is Six Sigma? • History of Six Sigma? • Methodologies of Six Sigma? • DMAIC process with examples from the real case studies M. S. Ramaiah School of Advanced Studies 69
- 70. References [1] Pyzdek,T., TheSix Sigma Hand Book, Revised and Expanded,New Delhi:McGraw-Hill [2] CIC2000: Six Sigma Green Belt Training, Motorola University [3] Benbow,W.D.,Kubiak,T.M.,The Certified Six Sigma Blackbelt Handbook,2005,USA,ASQ Quality Press M. S. Ramaiah School of Advanced Studies 70