# Variation Reduction (DMAIC) Fundamentals

1

Module Objective
By the end of this module, the participant should be able to: • Discuss the fundamentals of DMAIC in Variation Reduction • Explain the benefits of DMAIC in Reducing Variation

2

Merrill Lynch Six Sigma is a System of Proven Methodologies…
DFSS

Growth

Service
Growth Engines Eliminate Redundancies

Eliminate Waste
LEAN

Develop Talent

Reduce Variability
DMAIC
3

Let‟s look at the second of these 3 Key Components . . .

Premise • The basic premise of DMAIC is that sources of variation can be: • Identified • Quantified • Eliminated or Controlled • Focused on strategic or core processes • Data driven • Measurements focused on right things Variation Is The Enemy! © 2002 ConceptFlow 4 .

The Goal: Breakthrough Performance Six Sigma Breakthrough Defects and Waste Current State Improvement Period Breakthrough State Time © 2002 ConceptFlow 5 .

98% 99.210 233 3.4 69.33 years 6 sigma .you'd miss 1 putt per round 4 sigma .38% 99.807 6.Process Capability “Sigma Level” 2 3 4 5 6 Defects per Million Opportunities Long Term Yield 308.you'd miss 6 putts per round 3 sigma .32% 99.99966% If you played 100 rounds of golf per year.you'd miss 1 putt every 9 rounds 5 sigma .you'd miss 1 putt every 2.15% 93.you'd miss 1 putt every 163 years! © 2002 ConceptFlow 6 . and played at: • • • • • 2 sigma .537 66.

Simple Definition of Variation Accuracy = Centering Precision = Spread Defects Defects Client target Client target Reduce spread Center on target Off center and spread = process variation = Defects © 2002 ConceptFlow 7 .

„Voice of the Process‟ Versus „Voice of the Client‟ LSL USL Defects Acceptable Defects Voice of the Client Voice of the Process Inadequate Process Capability Inadequate Design Margin Unstable Voice of the Process is independent of the Voice of the Client © 2002 ConceptFlow 8 .

Goals of Six Sigma LSL USL LSL USL Defects Defects Client target Client target Eliminate Defects Reduce Variation LSL USL Center Process on Target Reduce Variation GOAL Client target Meet Client Target and Specifications © 2002 ConceptFlow 9 .

Dev.5 By shifting a 3 sigma process 1. = 2 Tolerance = 12 Shift the average 1.5 By shifting a 6 sigma process 1.5  we create 66.807 defects per million opportunities -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 6 Sigma Process Average = 0 St. Client Satisfaction and Business Objectives are robust to shifts due to process or product variation! © 2002 ConceptFlow 10 .4 defects per million opportunities -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 In a six sigma process.Six Sigma As a Statistical Measure 3 Sigma Process Average = 0 St. = 1 Tolerance = 12 LSL USL Shift the average 1.5  we create 3.Dev.

Can We Tolerate Variability? • There will always be variability present in any process • We can tolerate variability if: • The process is on target • The total variability is relatively small compared to the process specifications • The process is stable over time LSL Nom USL Cost Acceptable Traditional View LSL Nom USL Taguchi Loss Function Cost (New View) © 2002 ConceptFlow 11 .

000 incorrect surgical operations per week Two short or long landings at most major airports each day 200.99966% Good (6 Sigma) Seven articles lost per hour Unsafe drinking water for almost 15 minutes each day 5.000 wrong drug prescriptions each year No electricity for almost seven hours each month © 2002 ConceptFlow One unsafe minute every seven months 1.8 Sigma) 20.Is 99% Good Enough? 99% Good (3.000 lost articles of mail per hour 99.7 incorrect operations per week One short or long landing every five years 68 wrong prescriptions per year One hour without electricity every 34 years 12 .

The Magnitude of Six Sigma Sigma Area Spelling 1.000 indebtedness per \$1 billion in assets Time 3 1/2 months per century Distance Coast-to-coast trip 3 4 Floor space of a small hardware store Floor space of a typical living room 2 1/2 days per century 45 minutes of freeway driving(in any direction A trip to the local gas station 4 steps in any direction 5 6 Size of the bottom of your telephone Size of a typical diamond 1 misspelled word in a set of encyclopedias 1 misspelled word in all of the books contained in a small library \$570 indebtedness per \$1 billion in assets \$2 indebtedness per \$1 billion assets 30 minutes per century 6 seconds per century © 2002 ConceptFlow 13 .7 million indebtedness per \$1 billion in assets \$63.5 misspelled words per page in a book 1 misspelled word per 30 pages in a book Money \$2.

43 PPM) 7 Sigma Scale of Measure © 2002 ConceptFlow 14 .000. Today‟s Performance How does your industry rank? 1.000 10.000 100 10 1 Phone In Tax Advice (140.000 1..Six Sigma..000 PPM) Restaurant Bills Doctor Prescription Writing Payroll Processing Order Writeup Journal Vouchers Wire Transfers Airline Baggage Handling • Purchased Material Lot Reject Rate Average Company 1 2 3 4 5 6 Best-in-Class Domestic Airline Flight Fatality Rate (0.000 DPMO PPM 100.

807 6.Can We Inspect Our Way to Six Sigma?  st PPMlt 308.4 Defects Per Million Opportunities 2 3 4 5 6 Process Capability Inspection can only get us so far © 2002 ConceptFlow 15 .537 66.210 233 3.

the farm owners feel they should carry on with the family tradition of training farm hands of first class farmers in the fatherly handling of farm live stock because they believe it is the basis of good fundamental farm management © 2002 ConceptFlow 16 . Since the forefathers of the farm owners trained the farm hands for first class farms in the fatherly handling of farm live stock.The Inspection Exercise Task: Count the number of times the 6th letter of the alphabet appears in the following text: The necessity of training farm hands for first class farms in the fatherly handling of farm live stock is foremost in the eyes of farm owners.

.. i. Pareto Analysis • Measurement Systems Analysis • Rational Sub-grouping • Capability Analysis • Hypothesis Testing • ANOVA (Analysis of Variance) • Regression • DOE (Design of Experiments) • Response Surface Design • SPC (Statistical Process Control) MINITAB™ statistical software is the tool box for most of these tools © 2002 ConceptFlow 17 ..DMAIC.e. The Analytical Toolset to Drive Data-driven Decisions • Some DMAIC Tools • CT Tree • Process Mapping • Graphical Techniques.

x 2 ....Overall Problem Solving Approach Measure Practical Problem Analyze Statistical Problem y  f ( x1 . x k ) Practical Solution Control Statistical Solution Improve © 2002 ConceptFlow 18 ..

. X2 .Process Focus of Six Sigma Process Y = f(x) X Y • Inputs (X1. Xn) • Independent • Cause • Control • Output • Dependent on Input • Effect • Monitor Determining the critical X‟s & controlling the X‟s to guarantee the Y‟s © 2002 ConceptFlow 19 . .

) X1 X2 X3 Xn Process Input Variables (PIV) Process Output Variables (POV) Critical To Satisfaction (CTS) Cost Quality Delivery Process Y (VOP) Client‟s Needs & Expectations (VOC) Control the INPUTS to achieve the desired OUTPUTS that satisfies our Clients‟ Expectations © 2002 ConceptFlow 20 .How Y=f(x) Relates to a Process “Correct” (In-Spec.

Dynamics of Execution Strategy .The Funnel Effect Define Phase Measure Phase Analyze Phase Improve Phase Control Phase 30-50 Inputs (X) 10-15 Xs 8-10 KPIVs 4-8 Key KPIVs 3-6 Key KPIVs Optimized Process © 2002 ConceptFlow 21 .

visions emerge. Managers and Black Belts will work on what you are interested in . . direction becomes apparent • To lead change. gets done © 2002 ConceptFlow 22 . Black Belts have to be able to ask critical questions • Of the organization • Of process experts • Of process authority • Questions may often drive behavior Our Process Owners. Ask Relevant Questions • To better understand our process. . Leaders need to ask questions • Same questions… same answers • Changing the result means we have to change the question • As questions arise. What gets measured.Leaders Need to Speak a New Language.

their needs and requirements • Quantify the gap(s) between process outputs and client requirements • Define the performance standards or measures • Establish project objectives • Ensure resources are in place for the improvement project Primary responsibility of Project Champion and Black Belt © 2002 ConceptFlow 23 .Purpose of the Define Phase • Identify product and/or process to be improved • Identify the clients.

Define Phase Deliverables • Product or process balanced against strategic business requirements • Client. goals and benefits of the six sigma project • Financial benefits approved by Finance Department • A high-level process map of major processing events occurring in the process • Project plan with milestones and deliverables States the need of the client (CTS) and the problem © 2002 ConceptFlow 24 . critical-to-satisfaction requirements • Linkage of client requirements to process outputs • Team formed with charter describing purpose.

50 Inputs (X) 10 .The Funnel Effect .Measure Phase Key Tools • Process Map • C&E Matrix • FMEA • Measurement System Analysis • Data Collection Plan • Baseline Control Charts • Process Capability Analysis • Graphic Techniques Eliminate Waste 30 .15 Xs Optimized Process © 2002 ConceptFlow 25 .

Measure Phase Deliverables • • • • • • • • • • • • Validated project definition Detailed process map with outputs and labeled inputs Clearly defined process output measures (Y‟s) Measurement System Analyses on the Y‟s Control charts on baseline. process performance Process capability and entitlement Assessment of current control plan C&E Matrix to determine focus within process FMEA to assess and prioritize process risk Low hanging fruit actions assigned to team Valid data in accordance with the data collection plan Graphical interpretation of data Determines baseline process performance © 2002 ConceptFlow 26 .

Analyze Phase Key Tools • Hypothesis Testing • Multi-Vari Studies • Variance Components • ANOVA • Correlation • Regression Eliminate Waste 30 .The Funnel Effect .15 Xs 8 .50 Inputs (X) 10 .10 KPIVs Optimized Process © 2002 ConceptFlow 27 .

Analyze Phase Deliverables • A prioritized list of potential sources of variation • Variation Component Studies • Measurement Analysis on the x‟s • Data collected to validate sources • Graphical and statistical analysis of data • p-value establishing level of significance and probability • Correlation and regression analysis to determine variable relationships • Reduced list of potential key input variables that affect the output(s) • Updated control charts. process map & FMEA • Results to data (compared to baseline) Statistically links key input variables with key output variable © 2002 ConceptFlow 28 .

15 Xs 8 .Improve Phase Key Tools • Screening DOEs • DOEs • Historical DOEs • Modeling DOEs Eliminate Waste 30 .The Funnel Effect .50 Inputs 10 .10 KPIVs 4-8 Key KPIVs 3-6 Key KPIVs Optimized Process © 2002 ConceptFlow 29 .

Improve Phase Deliverables • Systematically experiment with KPIVs • Determine optimum conditions of KPIVs to result in optimum KPOV to meet CTQ • Proposed optimum operating conditions and solution • Proposed operating tolerances • Continue to lean process • A new (improved) process performance baseline and capability • Validated „significance‟ of improvements • Confirmation plan or results Statistically validates optimum process operation conditions © 2002 ConceptFlow 30 .

50 Inputs (X) 10 .The Funnel Effect .Control Phase Key Tools • Control Plan • SPC • Mistake Proofing • Standard Operations Eliminate Waste 30 .15 Xs 8 .10 KPIVs 4-8 Key KPIVs 3-6 Key KPIVs Optimized Process © 2002 ConceptFlow 31 .

sustainable results • Documented project and process improvements • Statistically validated process performance monitoring vehicle Keep the Process Owner involved throughout the project! © 2002 ConceptFlow 32 .Control Phase Deliverables • Approved Control Plan encompassing documentation required to maintain improvements • Lock in optimum KPIVs • Monitor and control KPIVs to generate optimum KPOV • Continue to lean process • Meet CTQ consistently.

Dynamics of Execution Strategy .The Funnel Effect Define Phase Measure Phase 30-50 Inputs (X) 10-15 Xs 8-10 KPIVs 4-8 Key KPIVs 3-6 Key KPIVs Analyze Phase Improve Phase Control Phase Optimized Process © 2002 ConceptFlow 33 .

the participant should be able to: • Discuss the fundamentals of DMAIC in Variation Reduction • Explain the benefits of DMAIC in Reducing Variation © 2002 ConceptFlow 34 .Module Objective By the end of this module.

Reference Material Variation Reduction – Cheat Sheet and Key Questions to Ask 35 .