Your SlideShare is downloading. ×
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Mpc mtcp six sigma [compatibility mode]
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Mpc mtcp six sigma [compatibility mode]

338

Published on

Six Sigma on Malaysian Technical Cooperation Program 2009.

Six Sigma on Malaysian Technical Cooperation Program 2009.

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
338
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. MTCP Program Six Sigma Introduction 15th June 2009 Topic: Six Sigma Introduction Malaysia Productivity Corporation (Statutory Body under MITI) P.O Box 64, Jalan Sultan, 46904 Petaling Jaya Website: www mpc gov my www.mpc.gov.my Consultant: Mohd Azlan Abas Tel: +6 012 308 7421 Email: Mohdazlan.abas@gmail.com 1 of 60Why Does One Need a Quality Initiative? Meet customer expectations for higher quality Provide a competitive differentiator in the service market Build greater pride and satisfaction in the team Drive other key goals: productivity and growth Tangible Costs Intangible Costs - Inspection - Expediting - Scrap - Lost Customers - Rework - Longer Cycles - Warranty - Lower Morale Enormous opportunity 2 of 60 1
  • 2. Six Sigma Provides focus on Critical to Quality (CTQ) Metrics in businesses  Vital Few CTQs that apply to all Customers: – Responsiveness – Marketplace Competitiveness p p – On-time, Accurate and Complete Deliverables – Product/Service Technical Performance  Key CTQs for a company: – Post Sales Issue resolution – Service Delivery Span – Contract Fulfillment – Parts Fulfillment – Pricing – Customer Escalation Cycle Time 3 of 60 Six Sigma Provides focus on Critical to Quality (CTQ) Metrics in businesses Practical Traditional Problem ApproachDefine MeasureD fi & M Statistical Problem Analyze Statistical Solution 6Quality Methodology Systematic Approach Focusing p Improve on Statistically Significant Root Practical Causes & Solutions Solution Driving Customer & Shareholder Benefits Control 4 of 60 2
  • 3. What is Six Sigma?Process to reduce defects per million opportunities B DPMO • From current levels to “Six Sigma” • “Sigma” is standard deviation from the ideal 2 308,537 • Can be applied to all business functions Can be applied to all business functions 3 66,807 66 807 » Manufacturing, Products, Transactions » Service, Sales Support 4 6,210Quantitative methodology 5 233 • Uses measurements and scientific process 6 3.4 3 to 6 ... 20 000 times improvement ... 20,000 ... A true quantum leap 5 of 60 Process = Hose Four Important Properties: (1) Centering (2) Spread (3) Shape (4) Stability Over Time 10.5 10 9.5 Y axis = Weight (lbs) 6 of 60 3
  • 4. Six Sigma ConceptEvery Human Activity Has Variability... Mean Lower Upper Customer C t Customer C t Specification Specification 1 p(defect) Target Reducing variability is the essence of six sigma 7 of 60 Mean SpecificationWhat is Sigma?  Limit Some Chance of Failure 1 The hi h th Th higher the  3 number (Z) in front of the sigma symbol the lower the chance of producing a Much Less Chance of defect Failure 1 6 Reducing variation is the key to reducing defects 8 of 60 4
  • 5. Reducing Variation is the Key What GE sees Order by Order Delivery Times Starting Point After Project 28 29 (Days) 18 6 6 10 13 17 23 12 5 4 8 10 What Customers feel 16 13 Mean 19 10 Big Change h 33 20 11 13 30% improvement VarianceAverage 17 13 No Significant Change 9 of 60 Reference: Six Sigma Performance 3.8 Sigma Six Sigma 99% Good 99.99966% Good• 20,000 lost articles of mail per • Seven articles lost per hour hour• Unsafe drinking water for • One unsafe minute every almost 15 minutes each day seven months• 5,000 incorrect surgical • 1.7 incorrect operations per operations per week week• Two short or long landings at • One short or long landing most major airports each day every five years• 200,000 wrong drug • 68 wrong prescriptions per prescriptions each year year 10 of 60 5
  • 6. Key Terms: Process or Activity Process X’s or Factors Outputs X1 Y1 X2 X3 PROCESS Y2 Y3 X4 For any given product, procedure or transaction, there are inputs, a process, and outputs. You will need to measure the outputs to quantify how well you satisfy a CTQ requirement; the output measures are Ys To Ys. change the process performance however, you must find and change the critical Xs. Find and control the critical X’s 11 of 60 The People GM Global Quality MasterQuality LeadersQ Black Belts Greenbelts Apply Six Sigma tools and methodology in Develop tools andHelp choose projects, everyday work. teaching materials.interview Black Beltcandidates, tie projects Conduct training andto business needs. communicationRemove barriers and Six Sigma g sessions. Mentordrive Six Sigma into the Black Belts Black Belts and theirculture of their functions. projects. Project leaders, change agents, expert application of tools, mentor Green Belts and their projects. 12 of 60 6
  • 7. DMAIC Objectives Step A • DEFINE Phase Purpose – Step A: Customer & Project CTQs – Step B: Team Charter (4‐Blocker) – St C Th Hi h L l P Step C: The High Level Process Map  M • Change Acceleration Process (CAP) – E = Q x A – The Model – Key CAP tools:  • ARMI • GRPI • Threat/Opportunity Matrix • In/Out of the Frame 13 of 60 DMAIC Customer & Project CTQs Step ACTQ Definition and CTQ Elements Product/ Cycle Time to Deliver Process Drawings Characteristic From Notice To Proceed ToCustomer Measure Delivery Time of Drawings (Weeks) Need CTQ On Time Target 13 Weeks Delivery Specification/ Tolerance 15 Weeks Limit(s)A performance standard translates customer needs into quantified requirements for our product or process 14 of 60 7
  • 8. A Project Team Charter DMAIC Step B Business Case (Problem statement) Potential Issues/ Speed Inhibitors•Multilin has a small market share in Indonesia  • Data collection compared to the more established European relays manufacturer. Need to improve our service to  Project Teamcustomer and increase sales in Indonesia.  customer and increase sales in Indonesia Name Org. Role % Ded.Goal Statement Vince Tullo Manager         Champion Stuward Thompson Manager Champion•To increase sales figures in Indonesia in 2004 to   W.N. Yew          Sales                 Leader          70%$1.8M. Daniel Sutando Sales leader Member 10% W.Y. Tan Sr App Engr Member 10%• To ensure relevant projects are pursued and to  Steven Tao    BB                  Member         5%improve the market coverage for distributors, EPCs, and end users. Milestones esto es Owner Date Scope Included: Project Charter Approved         YWN /Tao        Apr 30 • Multilin relays, Indonesia Market Measure /Baseline                      YWN/Tan          July 30 Analyze                                         WN/Tao Aug 30 Deliverables  Improve                                       YWN/Tao       Sept 30 Control                                          YWN/Tao   Oct 30Order growth for 2004 Close                                              YWN/Tao   Nov 30 15 of 60 DMAIC WIRE MILL PROJECT Step C 2 3 4 5 6 7 8 9 10 11 12 correct wire on cold change wire correct wire on wire size pulley weld reel speed size pulley build stand # feet per on out of damaged spool out of damaged reel round pulley round pulley 1 measureme nt accuracy measureme nt accuracy Xs proper die Input Legend lubrication l b i ti direction of critical lubricant flow pH level controllabl e 13 14 15 16 17 18 19 20 21 22 23 noise correct accurat e fat level size tension setting pressure out of temperature round Legend 1) Start of shift 13) Wire size change 2) Change final capstan felt 14) Stop machine 3) Check for excessive 15) Replace dies (as required) slivers/fines 4) Check size 5) Check surface quality 16) 17) 18) ) Restring new setup Check wire size Attach wire to spool p CTQ’s A) Diameter B) Out-of-round ) Ys 6) Check stand supply 19) Set dancer air pressure C) Loops 7) Is reel spooled D) Tangles 8) Connect wire to new reel 20) Crack valve 21) Check lubricant flow E) Pinchouts 9) Check size 22) Start spooler F) Surface 10) Check surface quality quality 11) Check for acceptable spool23) Start capstan build 12) Store in enamel room 16 of 60 8
  • 9. Objectives DMAIC • Introduction to Measure - Measure Phase Deliverables - Using Statistics to Solve Problems • Measure Step 1: Select CTQs - Quality Function Deployment (QFD) Process Quality Function Deployment (QFD) Process - Process Mapping - Failure Modes & Effect Analysis (FMEA) - Pareto Chart - Cause & Effect Diagram • Measure Step 2: Define Performance Standards - Performance Standard / Defect / - Basic Nature of Data • Measure Step 3: Measurement System • Introduction to Measurement Systems Analysis (MSA) 17 of 60 DMAIC The Statistical Problem Goal: Find the Relationship Process - A Y = f(X1, …, Xn) Process - B     Shape of the Curves CHARACTERIZES the Process p Process B is Better than Process A** Assumes same scale 18 of 60 9
  • 10. DMAIC Select CTQ Characteristics Step 1 Quality Function Deployment (QFD) Process Mapping (Flow Chart) Functional Requirements (HOW’s) Requirements Customer (WHAT’s) Map Customer Needs to Potential Hows Identify Critical Few - Resource Planning Map “Information” Flow Identify All Touch Points 19 of 60 DMAIC Select CTQ Characteristics Step 1 Failure Modes & Effects Analysis (FMEA) Process Step Potential Potential SE Potential O Current D R Action Owner OR Failure Failure Effects V Causes C Controls E P Recommended Part Number Mode C T N Anticipate Potential Failures in Process / Products & Develop Proactive Mitigation Plans D e fe c ts b y O p e ra tio n 20% of causes account for 1 00 0 0 00 100 80% of the problem 80 PercentCount 60 50 0 0 00 40 20 0 0 W W PE LO P LO HA EF AT 1 INS 2 P H E3 RF D e fe c t CH .S ST AG CO HO LE X -R AY HO LE INS BE NC HO L LW A TE MA AL A FIN FIN C o unt 2 7 6 1 4 41 3 0 8 4 41 2 7 2 0 41 0 2 5 9 91 0 1 4 9 1 9 3 3 5 3 8 2 8 6 1 5 4 1 1 0 4 9 6 4 3 3 6 7 0 7 P erce nt 2 6 .2 1 2 .4 1 2 .1 9 .7 9 .6 8 .8 7 .9 5 .1 4 .7 3 .5 C um % 2 6 .2 3 8 .6 5 0 .6 6 0 .4 7 0 .0 7 8 .8 8 6 .7 9 1 .8 9 6 .5 1 0 0 .0 20 of 60 10
  • 11. Performance Standards Objectives DMAIC Step 2• Determine the Performance Standard• Define a Defect - What are the customer’s acceptance criteria for the part/product or process?• Established How to Measure the Quality of the Part / Product or Process - Where are the data coming from? - How do you measure the process? - What are the units of measure? - Is it a discrete or continuous measure?• Gained Consensus On the Performance Standard 21 of 60 DMAIC What’s Performance Standards Step 2• Are Requirement(s) or Specification(s) Imposed by the Customer on a Specific CTQ• Translate Customer Needs into Measurable Characteristic – Have Clear Operational Definition, i.e. Specifies What to Measure, How to Measure & Collect the Data – Specifies Target or Mean – Impose Specification Limits – Have Clear Defect Definition CTQs Quantified – Everybody on Same Page 22 of 60 11
  • 12. DMAICMore Performance Standards . . . Step 2 Target Good Product Defective Product Tolerance Defective Product = USL - LSL LSL USL Lower Spec Limit Upper Spec LimitSpecification limits are set in order to divide customer satisfaction from customerdisappointment. While the exact limits may not be explicitly stated by the customer(and captured in Step 1), their specific values come from what the customer defines asa defect. 23 of 60 DMAIC The Basic Nature of Data Step 2 • Continuous Data - Characterizes a product or process feature in terms of its size, weight, volts, time, or currency - The measurement scale can be meaningfully divided into finer and finer increments of precision - Distributions: To apply the normal distribution, one must necessarily use continuous data • Discrete Data - Counts the frequency of occurrence: e.g., the number of times something happens or fails to happen - Is not capable of being meaningfully subdivided into more precise increments - The validity of inferences made from discrete data are highly dependent upon the number of observations. The sample size required to characterize a discrete product or process feature is much larger than that required when continuous data is used. - Distributions: The Poisson and binomial models are used in connection with this type of data 24 of 60 12
  • 13. DMAIC Making Data Driven Decisions Step 3• Six Sigma is all about reducing defects for the customer.  Fewer defect result in  a more satisfied customer.• The project team must decide what needs to be done to improve quality for  the customer based on actual data – measurements of the product or process.• The measurement system (gauge) used to collect the project data has to be The measurement system (gauge) used to collect the project data has to be  sufficiently good to allow the project team to make the correct decisions.• A measurement system must deliver data that accurately represents the  project or product. It is defective if it does not.• The Six Sigma project team becomes the customer for the measurement  process. There are many CTQs a project team must consider when evaluating  the quality of a gauge…If the gauge isn’t good enough for the needs of the project, th i ’t d h f th d f th j t you have to fix it (using DMAIC) before you can move on! 25 of 60 The Measurement Process DMAIC Step 3 In order to improve a product or process, we must measure it. We have submit the output from that process to a second measurement process. Parts (Example) As a result, all of our observations of the original Inputs Process ocess Outputs p process are distorted byy errors in our measurement Inputs system. We need to make sure these error don’t dominate our view of the process! • Observations Measurement Outputs • Measurements Process • Data 26 of 60 13
  • 14. DMAIC Accuracy (Bias) ‐ Shift in the Average Step 3 Inputs Process Outputs Inputs Measurement Outputs ProcessActual(Part) + Meas. System = Observed(Total) True Avg Bias Obs. Avg Measurement System Bias – Determined through “Calibration Study” 27 of 60 DMAIC Measurement System Precision Step 3 Inputs Process Outputs Inputs Measurement Outputs Process Product Variability Measurement Total Variability (part) Variability (Observed total) actual(part) + meas. system = observed(total) Measurement System Variability – Investigated through “R&R Study” 28 of 60 14
  • 15. DMAIC Accuracy vs. Precision Step 3 x xx x xx x xx Accurate? Accurate? x Yes NoYes No x xx xx x xx Precise? Precise? x Yes NoYes NoMeasured Value The True Value x x x x Accurate? Accurate? x x x YesYes No x x x No x x x x x Precise? Precise? x x Yes NoYes No x x x x x Accuracy: the difference between the observed average and the truth. Precision: the amount of inconsistency between measurements. 29 of 60 Characteristics of a Measurement System DMAIC Step 3 • Accuracy : Differences between observed average measurement  and a standard. • Resolution : The smallest scale of the measurement. • Stability : Do measurements change with time? • Linearityy : Is the measurement proportional to the magnitude (size,  p p g ( , weight, etc) of the sample. • Precision : Noise of the measurements. – Repeatability: variation when one person repeatedly measures the  same unit with the same measuring equipment.  Also called Equipment  Variation (EV). – Reproducibility: variation when two or more people measure the same  unit with the same measuring equipment.  Also called Appraiser  Variation (AV). Variation (AV) 30 of 60 15
  • 16. WHAT IS MINITAB?• Statistical software package• Has many of the tools needed to successfully analyze data with rigor - Graphs - Statistical tools for data analysis - Six Sigma Reports• Descriptive Statistics• Gage R & R• Capability Analysis (ZST/ZLT)• Graphing - Many Types! (Try them all!)• Pareto, Fish bone• Hypothesis Testing• Product & Process Six Sigma Reports• Generation of Test Plans for Designed Experiments• Analysis of DOE results• Statistical Process Control A Great Toolbox for Six Sigma Projects! 31 of 60 What does Minitab Look Like? Menu Bars: For quick access to common commands. Session Window: Shows Minitab text output (One only is present) Worksheet Window: Kind of like an Excel worksheet (at least one is always present). A tool is available to manage multiple worksheets. 32 of 60 16
  • 17. Objectives DMAIC • Calculate baseline capability of the process using either  continuous or discrete data • Statistically define the improvement goals y p g • Generated a list of Statistically Significant Xs based on  analysis of historical data • Identified which Xs to further investigate in the Improve  phase • Gained consensus with the project team on the list of Xs for Gained consensus with the project team on the list of Xs for  investigation 33 of 60 DMAIC Step 4: Establish Process Capability Step 4 USL Y = f (X 1 . . . X n)The variation inherent to any dependent variable (Y) is determined bythe variations inherent to each of the independent variables. (X) Poor Excellent Process Capability Process Capability Very High Very High Very Low Very LowProbability of Probability of Probability of Probability of Defects Defects Defects Defects LSL USL LSL USL Low Z High Z Z is a Measure of Process Capability 34 of 60 17
  • 18. Step 5: Define Performance Objective p(x) Benchmark Entitlement Baseline DefectsBenchmark: World-Class performance Entitlement: The level of performance a business should be able to achieve given the investments already made Baseline: The current level of performance Benchmarking Sets the Ultimate Goal, while Baselining Takes Current Measurements to Monitor a Process 35 of 60 DMAICStep 6: Identify Sources of Variation Step 6 Y= f (X)To get results, should we focus our behavior on the Y or X ?n Y n X1 . . . Xnn Dependent n Independentn Output n Input-Processn Effect n Causen Symptom n Problemn Monitor n Control Historically the Y, … with Six Sigma the Xs 36 of 60 18
  • 19. Process Capability DMAIC Step 4Determining Process Capability (Steps 1 – 4)1. Identify Requirements on the Customer or Internal Y These are often expressed quantitatively as target values plus specification limits. - Anything “outside of the specification limits” is a defect. Target Value g (lower specification limit) LSL USL (upper specification limit) 90 100 110 Defects Defects Tolerance2. Determine Process Distribution We expect variation in our process. Mean = 102 The Y’s measured on a large number Ys  = 5.0 of parts will form a distribution. For simplicity, we’ll assume that they form a normal distribution with a known mean and standard deviation. 85 90 95 100 105 110 115 37 of 60 Process Capability DMAIC Step 43. Superimpose your process distribution on to the target and specification limits. Target Mean = 102  = 5.0 LSL USLProbability of Probability of a Defect a Defect 85 90 95 100 105 110 115 4. Apply some Basic Statistics. 1. We can relate our distribution to the standardized normal distribution. • The total area under the standard curve = 1 (or 100%) 2. If we can determine the Z l corresponding to a specification li i 2 d i h Z-value di ifi i limit, then we can calculate the area beyond that limit (out of specification) • The area beyond a given specification limit is the fraction of the population that is defective wrt that limit. 38 of 60 19
  • 20. Process Capability DMAIC Step 4 Putting it all together (Steps 1 – 4) Total Area Outside of Specification Limits = (0.055 + 0.008) = 0.063 6.3% of what we produce is defective 90  102 Notice also that our 110  102Z(LSL)   2.4 Z(USL)   1.6 5 process is not centered 5Probability  0.008 on the target Probability  0.055 Target Mean = 102  = 5.0 LSL USL 85 90 95 100 105 110 115 39 of 60 Single- Single-Tail Z Table (values from 0.00 to 3.99) Z 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0 0.5000 0.4960 0.4920 0.4880 0.4840 0.4801 0.4761 0.4721 0.4681 0.4641 0.1 0.4602 0.4562 0.4522 0.4483 0.4443 0.4404 0.4364 0.4325 0.4286 0.4247 0.2 0.4207 0.4168 0.4129 0.4090 0.4052 0.4013 0.3974 0.3936 0.3897 0.3859 0.3 0.3821 0.3783 0.3745 0.3707 0.3669 0.3632 0.3594 0.3557 0.3520 0.3483 0.4 0.3446 0.3409 0.3372 0.3336 0.3300 0.3264 0.3228 0.3192 0.3156 0.3121 0.5 0.3085 0.3050 0.3015 0.2981 0.2946 0.2912 0.2877 0.2843 0.2810 0.2776 0.6 0.2743 0.2709 0.2676 0.2643 0.2611 0.2578 0.2546 0.2514 0.2483 0.2451 z 0.7 0.2420 0.2389 0.2358 0.2327 0.2296 0.2266 0.2236 0.2206 0.2177 0.2148 0.8 0.2119 0.2090 0.2061 0.2033 0.2005 0.1977 0.1949 0.1922 0.1894 0.1867 0.9 0.1841 0.1814 0.1788 0.1762 0.1736 0.1711 0.1685 0.1660 0.1635 0.1611 1 0.1587 0.1562 0.1539 0.1515 0.1492 0.1469 0.1446 0.1423 0.1401 0.1379 1.1 0.1357 0.1335 0.1314 0.1292 0.1271 0.1251 0.1230 0.1210 0.1190 0.1170 1.2 0.1151 0.1131 0.1112 0.1093 0.1075 0.1056 0.1038 0.1020 0.1003 0.0985 1.3 0.0968 0.0951 0.0934 0.0918 0.0901 0.0885 0.0869 0.0853 0.0838 0.0823Calculates the 1.4 0.0808 0.0793 0.0778 0.0764 0.0749 0.0735 0.0721 0.0708 0.0694 0.0681 1.5 0.0668 0.0655 0.0643 0.0630 0.0618 0.0606 0.0594 0.0582 0.0571 0.0559probability of a USL 1.6 0.0548 0.0537 0.0526 0.0516 0.0505 0.0495 0.0485 0.0475 0.0465 0.0455 1.7 0.0446 0.0436 0.0427 0.0418 0.0409 0.0401 0.0392 0.0384 0.0375 0.0367defect above the 1.8 0.0359 0.0351 0.0344 0.0336 0.0329 0.0322 0.0314 0.0307 0.0301 0.0294 1.9 0.0287 0.0281 0.0274 0.0268 0.0262 0.0256 0.0250 0.0244 0.0239 0.0233given value of z 2 0.0228 0.0222 0.0217 0.0212 0.0207 0.0202 0.0197 0.0192 0.0188 0.0183 2.1 0.0179 0.0174 0.0170 0.0166 0.0162 0.0158 0.0154 0.0150 0.0146 0.0143 2.2 0.0139 0.0136 0.0132 0.0129 0.0125 0.0122 0.0119 0.0116 0.0113 0.0110 2.3 0.0107 0.0104 0.0102 0.0099 0.0096 0.0094 0.0091 0.0089 0.0087 0.0084 2.4 0.0082 0.0080 0.0078 0.0075 0.0073 0.0071 0.0069 0.0068 0.0066 0.0064 2.5 0.0062 0.0060 0.0059 0.0057 0.0055 0.0054 0.0052 0.0051 0.0049 0.0048 2.6 0.0047 0.0045 0.0044 0.0043 0.0041 0.0040 0.0039 0.0038 0.0037 0.0036 2.7 0.0035 0.0034 0.0033 0.0032 0.0031 0.0030 0.0029 0.0028 0.0027 0.0026 2.8 0.0026 0.0025 0.0024 0.0023 0.0023 0.0022 0.0021 0.0021 0.0020 0.0019 2.9 29 0.0019 0 0019 0.0018 0 0018 0.0018 0 0018 0.0017 0 0017 0.0016 0 0016 0.0016 0 0016 0.0015 0 0015 0.0015 0 0015 0.0014 0 0014 0.0014 0 0014 3 0.0013 0.0013 0.0013 0.0012 0.0012 0.0011 0.0011 0.0011 0.0010 0.0010 3.1 9.68E-04 9.36E-04 9.04E-04 8.74E-04 8.45E-04 8.16E-04 7.89E-04 7.62E-04 7.36E-04 7.11E-04 3.2 6.87E-04 6.64E-04 6.41E-04 6.19E-04 5.98E-04 5.77E-04 5.57E-04 5.38E-04 5.19E-04 5.01E-04 3.3 4.83E-04 4.67E-04 4.50E-04 4.34E-04 4.19E-04 4.04E-04 3.90E-04 3.76E-04 3.62E-04 3.50E-04 3.4 3.37E-04 3.25E-04 3.13E-04 3.02E-04 2.91E-04 2.80E-04 2.70E-04 2.60E-04 2.51E-04 2.42E-04 3.5 2.33E-04 2.24E-04 2.16E-04 2.08E-04 2.00E-04 1.93E-04 1.85E-04 1.79E-04 1.72E-04 1.65E-04 3.6 1.59E-04 1.53E-04 1.47E-04 1.42E-04 1.36E-04 1.31E-04 1.26E-04 1.21E-04 1.17E-04 1.12E-04 3.7 1.08E-04 1.04E-04 9.96E-05 9.58E-05 9.20E-05 8.84E-05 8.50E-05 8.16E-05 7.84E-05 7.53E-05 3.8 7.24E-05 6.95E-05 6.67E-05 6.41E-05 6.15E-05 5.91E-05 5.67E-05 5.44E-05 5.22E-05 5.01E-05 3.9 4.81E-05 4.62E-05 4.43E-05 4.25E-05 4.08E-05 3.91E-05 3.75E-05 3.60E-05 3.45E-05 3.31E-05 40 of 60 20
  • 21. Identify Variation Sources DMAIC Step 6 Goal: Process on Target with Minimum Spread Desired Problem with Spread Accurate but Current not Precise Situation LSL T USL Problem with Centering – Not on Target Current Desired SituationPrecise but not Accurate LSL T USL  Is the Problem Centering, Spread, or Both? 41 of 60 DMAIC Continuous Data Analysis Road Map Step 6 One Sample Two Samples Multiple Samples Study Stability (Run Chart) Study Shape (Histogram, Dot Plot, Normality)Study Spread Data paired?(Chi Square-Test) Study Spread (Homogeneity of Variance) Study Spread Study Centering (F test) (F-test) (Paired t test) t-test)Study Centering Study Centering (1 sample t-test) (ANOVA) Study Centering (2 sample t-test) 42 of 60 21
  • 22. Descriptive Statistics DMAIC Step 6Stat > Basic Statistics >Display DescriptiveStatistics Provides summary report of basic statistical information p-value > .05 Cannot Histogram Reject H0 Minitab Output No evidence that data is non-normal Dot Plot Basic Statistical Information Confidence Intervals 43 of 60 DMAIC X and Y Data Correlation Step 6 25 Strong Positive Correlation 20 15 Y 10 5 0 0 5 10 15 20 25 X 25 No Correlation 20 15 Y 10 5 0 0 5 10 15 20 25 X 44 of 60 22
  • 23. Improve Phase Objectives• The benefits of Design of Experiments (DOE’s)• Key concepts and terms associated with DOE’s• Performing a simple full factorial and fractional DOE’s and  interpreting the results• Awareness of screening designs and higher level response  surface designs 45 of 60 DMAIC What’s Improve Phase About. . .• Develop an Improvement Strategy • Determine which candidate x’s identified in the Analyze Phase are truly “critical X’s”. • If possible, determine a quantitative transfer function that relates your Y to these critical X’s• Identify Improvement Actions • Determine optimal settings for the X’s • Show the impact of the changes on meeting project or business objectives. Y = f(x)• Validate the Improvement • Demonstrate the validity of your identified improvement actions via additional experiments or a pilot study• Develop a Plan to Implement the Change It’s More than Just Designed Experiments 46 of 60 23
  • 24. DMAIC Common “Improve” Tools Basic Intermediate Advanced Process Map  DOE  DOE Fishbone  Full Factorial  Response Surface Box Plot  Fractional  Taguchi (Inner / Factorial Outer Array) Time Order Plots  Intro to  Simulation Models Hypothesis Tests Response Linear Regression Surface  Already Covered  Covered in Improve Mistake Proofing  Multivariate  Covered in DFSS Regression  Adv. Level III e.g. ProModel LOW Problem Sophistication• Complexity • Risk HIGH Match the Tool to the• Business Impact Problem • Data Availability 47 of 60 DMAIC DOE – Design of Experiments Steps 7-8 Y = f (x1, x2, x3,……xn)Response (Y) Factors (x’s) (x s)• The measured outcome of an • The critical X’s which determine the experiment response,Y• The value observed for the CTQ • They can be categorical or being explored numericalRanges Levels• The extreme values for each factor • In DOE’s we investigate the effect determines the range for that factor of each factor at more than one - the region of interest/investigation setting or value 48 of 60 24
  • 25. DMAIC Benefits of DOEs Steps 7-8Classical ApproachOFAT - One Factor at a Time• Change one variable, X2, 100 90 while holding all others constant. 80• Find a maximum 70• Hold X2 at the 60 “maximum effect” level and Factor X1 repeat the process for the other variables.OFAT• Requires more experiments than a DOE• Becomes unmanageable as the number of factors increases• Can be very expensive and time consuming – and may not work very well 49 of 60 Benefits of Design of Experiments DMAIC Steps 7-8 DOE Approach • Select factors and levels • Select mathematical model designed to obtain maximum information for the number 100 00 90 of factors/levels selected. • In your experiments change 80 the factor levels in a systematic 70 manner so that all coefficients in 60 the model can be uniquely computed. (Orthogonality) Factor X1 • Solve the resulting set of simultaneous equations to obtain the coefficients. coefficients • Use statistical tests to determine if the coefficients are statistically significant, and if the resulting model (transfer function) is adequate. • Use the results of your DOE to plan the next DOE (if needed). 50 of 60 25
  • 26. DMAIC Screening Designs Steps 7-8 The Team’s understanding at the beginning of the project What Mother Nature KnowsThe following 7 Factors may be critical Xs Yield = 64.25 +A. Temperature (160C - 180C) 11.50*AB. Monomer Concentration (20% - 40%) -C. Catalyst Vendor (Sally - Ed) 2.50*BD. Stirring Speed ( 50 RPM - 100 RPM) +E. Monomer Purity ( 90% - 98%) 0.75*CF. Pressure ( 100 PSI - 500 PSI) +G. Acetone/Methanol Ratio - ( 0.25 - 0.50) 5.00*A*C We need an efficient method for screening these "candidate critical Xs" so that we can identify the Vital Few" 51 of 60 DMAIC Objectives • In the physical world, the law of Entropy explains the gradual loss of  order in a system.  The same law applies to business processes.  • Unless we add “energy” (in the form of documentation and ongoing  process controls), processes will tend to degrade over time, losing the  gains achieved by design and improvement activities. • The quality plan is the structure through which we add this “energy”  to business processes.   This is Control and the main objectives: – To make sure that our process stays in control after the solution has been  implemented.  – To quickly detect the out of control state and determine the associated  special causes so that actions can be taken to correct the problem before  special causes so that actions can be taken to correct the problem before nonconformance are produced. 52 of 60 26
  • 27. DMAIC Control – Control – Keep It On Target • Focus on the Right CTQ • Quantify the ProblemDefine • Determine the DriversMeasure Y = f(X)Analyze • Identify Needed ChangeImprove • Implement the Change • Validate Measurement System (Xs) • Determine Process Capability • Develop/Modify Quality Plan Control > Process Documentation > Process Controls • Implement Process Controls • Audit Plan Established • Transition to Operating Owners 53 of 60 DMAICOn the Lookout for Special Cause • Common cause variation – Natural variability – Random – Inherent in the process • Special cause variation – May be caused by operator errors, adjusted machines, or  defective raw materials – Generally large when compared to the common cause variation – Considered an unacceptable level of process performance • Special causes tend to cause a process to shift out of  control where. The output does not meet the desired  specifications. 54 of 60 27
  • 28. DMAICWhat is a Process Control System?• A Process Control System (PCS) – strategy for maintaining the improved process performance over time – identifies the specific actions and tools required for sustaining the  process improvements or gains • A control system may incorporate y y p – Risk Management – Mistake‐proofing devices  – Statistical process control (SPC)  – Data collection plans  – Ongoing measurements  – Audit plans  – Response plans* Response plans – Product drawings  – Process documentation  – Process ownership 55 of 60 Statistical Process Control DMAIC Step 12Statistical -- Probability based decision rules.Process -- Any repetitive task or steps.Control -- Monitoring of process performance. SPC will signal when the process is “out-of- control”. Your Mission is to find out why and take corrective action! 56 of 60 28
  • 29. Control Chart Components DMAIC Step 12 Average Chart Statis (mean/defects) 0.065 Upper Control Limit measured Grand Average 0.060 Central Line stic Lower Control Limit 0.055 0 5 Sample / Subgroup (time ordered) 10 15 20 25 Monitors Shift Variation Chart e/Sigma) 0.010 Upper Control Limit redStatistic (Range measur 0.005 0 005 Average Range/Sigma 0.000 Central Line Lower Control Limit Sample / Subgroup (time ordered) Monitors Drift 57 of 60 Significance of 3s limit DMAIC Step 12 A Control Chart is a graphic display of a continuing two tailed test with HO and HA defined as: Ho: iHa: i /2 /2 UCLx X LCLx /2• For 3 limits,  = 0.00135. approximate confidence level is 99.7%.• 3  limits provide good sensitivity to change with low potential for over- reacting when the process is stable. 58 of 60 29
  • 30. Types of Control Chart DMAIC Step 12 Variable Chart (Continuous) Attribute Chart (Discrete)• Uses Measured Values • Defects: Number of non conformance in a part – Cycle Time, Lengths, • Defective: Pass/Fail, Diameters, Drops, etc. Good/Bad, Go/No-Go• Generally One Characteristic Per y Information Chart • Can Be Many Characteristics• More Expensive, But More Per Chart Information • Less Expensive, But Less Information High or Low Volume? Constant Variable Lot / Unit Size Lot / Unit Size Low L High Defects Poisson c u Individuals & X-Bar & Moving Range Defective np p Binomial Range 59 of 60 THANK YOU 60 of 60 30

×