Six Sigma Overview

1

Prepared by Dr. Leonard R. HeppAll Rights Reserved

Simply Left Mouse Click to advance through animations & slides.

2


A Methodology … For Continuous Improvement

Six Sigma is a highly disciplined data-based methodology of problem solving leveraging tools & techniques where appropriate.

What is Six Sigma?

Six Sigma follows two rigorous approaches:

DMAIC Methodology …for improving EXISTING processes

DMADOV Methodology …for CREATING a new product or process

Let’s Look At Each Method

Define Measure Analyze Improve Control

VerifyDefine Measure Analyze Design Optimize

3


Six Sigma: Key Terms

Critical to Quality Customer Performance Requirements

Characteristic (CTQ) of a Product or Service

Defect Any Event That Does NOT Meet the

Performance Specifications of a CTQ

Defect Opportunity Any Event Which Can Be Measured

That Provides a Chance of Not Meeting

a Customer Requirement

Concept of Defects is at the Core of Six Sigma!Concept of Defects is at the Core of Six Sigma!

Concept Definition

4


Target

CustomerSpecification

X XXX X XX XXXX

XX

XXXXX

XX

XX

XX

X

XXX

XX

XX XXX

XXXX X

X

X

X

Every Human Activity Has Variability...

1

defects

Variation & Defects are the Enemy

5


Target CustomerSpecification

1

2

3

A 3 process because 3 standard deviations fit between target and spec

Target

CustomerSpecification

12

34

56

6“No Defects”

Reliability thru Variance Reduction

6


And … What About Discrete Data?

We have an Invoice Payment Process of 45 Days or less and recent data for last month shows that of 100 invoices, 92 were paid in 45 days or less.

92% “On Time”

Question…How many were paid in 30 days? Between 40-45 Days? What was my shortest payment cycles?

With Discrete Data you’d have to go back and re-measure! With Discrete Data you’d have to go back and re-measure! And you can’t model new performance limits. And…And you can’t model new performance limits. And…

7


Average CustomerSpecification

63.4 Defects per

Million Opportunities6

_________________________________________________________________________

3

AverageCustomer

Specification

67,000 Defects per Million Opportunities3

Considerations

• Customer Specification

• Average

• Variation

Six Sigma - A Stretch GoalFor many processes

BUTNot Good Enough for Some!

What is Six Sigma Quality?

8


Sigma is a statistical unit of measure that reflects

process capability

DPMODPMO

6 3.4 99.9997%

5 233 99.98%

4 6,210 99%

3 66,807 93%

2 308,537 69%

%%

ProcessCapability

Defects Per MillionOpportunities

PercentageGood

Increase Requires Exponential DPMO Reduction

What is Six Sigma Quality?

9


1,000,000

10,000

1,000

100

10

1

Sigma Scale of Measure

Restaurant Bills

Doctor Prescription Writing

Order Write-up

Domestic AirlineFatality Rate(0.43 PPM)

Airline Baggage Handling

“Average”IndustrialCompany

Best-in-ClassIndustrial Company

Defectsper

Million

3 4 5 6 72

“Typical” Service Industry Processes are 1.5 to 3

IRS Tax Advice(phone in)

Sigma Quality Level - Examples

10


MeasureMeasure

AnalyzeAnalyze

ImproveImprove

DefineDefine

Six Sigma DMAIC Process

ControlControl

6

Characterization

Optimization

666

CharacterizationCharacterization

OptimizationOptimization

11


MeasureMeasure

AnalyzeAnalyze

ImproveImprove

DefineDefine

ControlControl

Six Sigma DMAIC Overview

Practical Problem: Low Yield

Statistical Problem: Mean Off Target

Statistical Solution: Isolate Key Variables

Practical Solution: Install Automatic Controller

Practical SolutionPractical Solution

Statistical SolutionStatistical Solution

Practical ProblemPractical ProblemProblem Solving

FlowNeed

Need

Need

Need

Statistical ProblemStatistical Problem

Do

Do

Do

Practical ProblemWhat’s The Problem?

12


Six Sigma DMAIC The 12+3 Step DMAIC Strategy

DMAIC – The 12 + 3 Steps

Formulating the Practical ProblemFormulating the Practical Problem

Changing to a Statistical ProblemChanging to a Statistical Problem

Developing a Statistical SolutionDeveloping a Statistical Solution

Implementing the Practical SolutionImplementing the Practical Solution

Step 0: Build the House of Quality

A. Identify Needs B. Team Charter C. Process/SIPOC

Step 1: Select the CTQ Characteristic

Step 2: Define Performance Standards

Step 3: Validate MSA and Data Collection

DMAIC

DMAIC

DMAIC

DMAIC

Step 4: Establish Process Capability

Step 5: Define Performance Objectives

Step 6: Identify Variation Sources

DMAIC

DMAIC

DMAIC

Step 7: Screen Potential Causes

Step 8: Discover Variable Relationships

Step 9: Establish Operating Tolerances

DMAIC

DMAIC

DMAIC

Step 10: Validate MSA on the Xs

Step 11: Determine Process Capability

Step 12: Implement Process Controls

DMAIC

DMAIC

DMAIC

How good am I today?

How good do I need to be?

What factors make a difference?

How good am I today?

How good do I need to be?

What factors make a difference?

How do my customers look at me?How do my customers look at me?

What do I want to improve?

What’s the best way to measure?

Can I trust the output data?

What do I want to improve?

What’s the best way to measure?

Can I trust the output data?

What’s at the root of the problem?

How can I predict the output?

How tight does the control have to be?

What’s at the root of the problem?

How can I predict the output?

How tight does the control have to be?

Can I trust the in-process data?

Have I reached my goal?

How can I sustain the improvement?

Can I trust the in-process data?

Have I reached my goal?

How can I sustain the improvement?

13


Define Measure Analyze Improve ControlDefine Measure Analyze Improve Control

MeasureMeasure ImproveImprove ControlControlAnalyzeAnalyze

A. Customer A. Customer FocusFocus--CTQsNeeds

B. Team B. Team CharterCharter

C. Process C. Process MappingMapping

DefineDefine

Define The ProblemDefine The Problem

• Collect VOC• Define Customer Needs

• Business Case• Problem & Goal

Statements• Project Timeline• Team Members

Project Scope

• 5-7 High Level Steps

• Validated by Process Owner

MeasureMeasure ImproveImprove ControlControlAnalyzeAnalyze

A. Customer A. Customer FocusFocus--CTQsNeeds

B. Team B. Team CharterCharter

C. Process C. Process MappingMapping

DefineDefine

Define The ProblemDefine The Problem

• Collect VOC• Define Customer Needs

• Business Case• Problem & Goal

Statements• Project Timeline• Team Members

Project Scope

• 5-7 High Level Steps

• Validated by Process Owner

DefineDefine MeasureMeasure ImproveImprove ControlControlAnalyzeAnalyze

1. Select CTQ 1. Select CTQ CharacteristicsCharacteristicsMeasurementMeasurement

2. Define 2. Define Performance Performance

StandardsStandards

3. Validate 3. Validate Measurement Measurement

SystemSystem

Frame as a Practical ProblemFrame as a Practical Problem

• Quality Function Deployment (QFD)

• Select Key Output Metric “Y”

• Customer Specification Limits

• Targets

• Measurement Systems Analysis

• Data Collection Plan

5. Define 5. Define Performance Performance ObjectivesObjectives

6. Identify 6. Identify Variation Variation SourcesSources

4. Establish 4. Establish Performance Performance

CapabilityCapability


Frame as a DataFrame as a Data--Based ProblemBased Problem

• Baseline Current Process Performance

• Set Improvement Goal

• Identify Potential Root Causes

• Hypothesis Testing and Detailed Process Analysis


7. Screen 7. Screen Potential Potential CausesCauses

9. Establish 9. Establish Operating Operating

TolerancesTolerances

8. Discover 8. Discover Variable Variable

RelationshipsRelationships

Frame as a DataFrame as a Data--Based SolutionBased Solution

• Determine Root Cause

• Design Improvement Solution

• Pilot Solution


10. Validate 10. Validate Measurement Measurement

System (on XSystem (on X’’s)s)

12. Implement 12. Implement Process Process ControlsControls

11. Determine 11. Determine New Process New Process

CapabilityCapability

Frame as a Practical SolutionFrame as a Practical Solution

• Measurement System Analysis

• Improved Process Performance

• Control Plans & Dashboards

• Documentation• Process

Management Plan

14


Yes

Yes

AnalyzeMeasure Improve ControlDefine

Is the Improvement

a New or Redesigned

Product/Service?

Is the Improvement

a New or Redesigned

Product/Service?

AnalyzeMeasure Design VerifyDefine

IsIncremental

Improvement Enough?

IsIncremental

Improvement Enough?

Does a Process Exist?

Does a Process Exist?

NoNo

Yes No

DMAIC/DMADOV Transition Points:

Optimize

When To Use DMADOV

15


Design For Six SigmaDesign For Six SigmaDesign For Six Sigma

DefineDefineDefine MeasureMeasureMeasure AnalyzeAnalyzeAnalyze DesignDesignDesign VerifyVerifyVerify

Develop Detailed Design And Control/Test Plan

Develop Detailed Design And Control/Test Plan

Implement Full-Scale Processes and Document Control Plans

Implement Full-Scale Processes and Document Control Plans

Initiate, Scope, And Plan The Project

Initiate, Scope, And Plan The Project

Understand Customer Needs And Specify CTQs

Understand Customer Needs And Specify CTQs

Develop Design Concepts And High-Level Design

Develop Design Concepts And High-Level Design

OptimizeOptimizeOptimize

Test Design, Optimize and validate performance to CTQs

Test Design, Optimize and validate performance to CTQs

VerifyDefine Measure Analyze Design Optimize VerifyDefine Measure Analyze Design Optimize

DMADOV – The 5 Phases and 14 Steps

16


1. Identify customer needs (CTQ’s) and set performance goals.

8. Generate and validate models - Identify transfer functions.9. Capability flow-up utilizing scorecards…watch for:

• Low Zst on scorecard.• Lack of transfer function.• Unknown process capability.

10. Optimize design• Statistical analysis of variance drivers• Robustness• Error proofing

11. Generate process specs and verify measurement system X’s

12. Statistically confirm predictions.13. Develop control plan for CTQ’s (mean and variance).14. Document the effort and results.

The DMADOV Methodology – 14 Steps

2. Perform QFD/CTQ flowdown…Needs to Design Requirements 3. Establish measurement system capability.

4. Develop conceptual designs5. Reliability Analysis of Designs6. Build Scorecard of Customer Needs (CTQ’s)7. Perform risk assessment

QUALITY BY DESIGN!QUALITY BY DESIGN!

DMADOV DMADOV FundamentalsFundamentals(Key Concepts)(Key Concepts)

QFD-CTQFlow-Down

FMEA

Business Model (Transfer Function)

Scorecards

AnalyzeAnalyze

DesignDesign

OptimizeOptimize

VerifyVerify

DefineDefine

MeasureMeasure

17


VerifyDefine Measure Analyze Design Optimize VerifyDefine Measure Analyze Design Optimize

Step 1: Identify Customer CTQ’s and Set Quality Goals

DEFINE

ID Service or Product Issue

Business Case

Problem/Goal Statement

Project Leadership/Team

ID Customers

Gather Customer Needs

Rank/Prioritize Needs

Project Scope Review

MGPP

Resources/ Team Assessment

Project Management Approach

Project Plan and Timelines

Organization Challenges

Plan & Manage the

ProjectInitiate the

Project

ID Customers & Gather

NeedsScope the

Project

Key Focus Areas

DMADOV

MSA Analysis• GR&R – Precision

• Accuracy

• Linearity

• Stability

Baseline-Sigma and/or Zscores

Legal and Regulatory

Organize and Prioritize Customer Needs

Determine Project CTQ’s

Flow-Down of CTQ’s by QFD

Perform CTQ Flow-Down Establish MSA and Baseline

Step 2: Perform QFD/CTQ Flowdown

Step 3: Establish Measurement System Capability & Baseline

MEASURE

Key Focus Areas

DMADOV

Step 4: Develop Conceptual Designs

Step 5: Reliability Analysis of Designs

Step 6: Build Scorecards of CTQ’s

Step 7: Perform Risk Analysis

High Level Design and Process Focus

Requirements for Process, IT, Facilities, HR, etc.

Assess Capability via CTQ Flow-Down & Flow-Up

Select Best-Fit Design

Customer Feedback

Risk & Patent Reviews

Functional Analysis

Alternate Design Reviews

Develop Design Concepts

Develop High Level Designs

Evaluate High Level Design Capability

Key Focus Areas

ANALYZE DMADOV

Step 8: Generate & Validate Models • Develop Transfer Function

Step 9: Capability Flow-Up Utilizing Scorecards• Low Zst on Scorecard• Lack of Transfer Function• Unknown Process Capability

Predict & Improve Design Capability

Design Reviews & Risk Assessment

Control Strategy Plan

Test & Validate Control Plan

Pilot Process Review

Design Elements to meet Functional CTQs

Detailed Process Elements

Determine & Measure CTPs

Develop Detailed Design

Evaluate Detailed Design Capability

Prepare Control & Verification Plan

DMADOV

Watch for These!

Key Focus Areas

DESIGN

Step 10: Optimize Design• Statistical Analysis of Variance Drivers• Robustness• Error Proofing

Step 11: Generate Process Specs & Verify MSA on the X’s

Business Model (Transfer Function)

MSA on System X’s

Pilot & Assess Variance

Robustness & Error Proofing

Pilot Process & Test to ScoreCard

Optimize Design Process Specs and MSA

Key Focus Areas

OPTIMIZE DMADOV

Step 12: Statistically Determine Predictions

Step 13: Develop Control; Plans for CTQs (Mean & Variance)

Step 14: Document Efforts and Results

Build Full Scale Process

Scale-Up & Test

Verify Performance

Turnover to Ops & Maintenance

Transition to Process Owners

Project Close

Build Pilot Scale Process

Pilot Test & Evaluate

Implement Planning

Execute Pilot and Analyze Results

Implement Production Process

Transition to Process Owners

Key Focus Areas

VERIFY DMADOV

18


Project Charter

Problem Statement & Goal

Project Scope

Project Milestones (with firm dates for DMA, targets for DOV)

High-Level CTQ’s

Project Team (Leader, Champion, Sponsor, Black Belt, Master Black Belt, Team Members, Other Resources)

Internal Communication Plan

Business Case

Project Risk Assessment (FMEA or written assessment)

Multi-Generational Plan

Cost-Benefit Analysis

Identify, segment & prioritize customer

User Profiling (how many, how often, from where)

Identify & prioritize CTQ’s

Interviews, surveys, or focus groups

Measurement Plan

Acceptance Criteria

As-Is Process Documentation

QFD to determine how to satisfy CTQ’s

Benchmarking (within your team, within company, external)

To-Be Process Map/High-Level Solution

Make vs. Buy Analysis

Vendor/Technology Selection

Detailed Functional Specification

Prototype (use to iteratively refine functional specification)

Define Test Cases

Final Project Schedule and Project Plan

Phased Rollout Schedule

End-User Communication and Marketing Plan

Technical Specification

Interface Design

Application Architecture

Information Architecture (DB)

Server Architecture

Code Reuse Strategy

System FMEA

Security Plan (engage SSO team)

Backup & DR Plan

Monitoring Solution

Peer Technical Reviews

Packaged Software Customization Review

Help Desk Strategy

Purchase Hardware and Software

Schedule Stress Test and Security Review

Code Application

Develop User & Training Documentation

System Documentation

Application Kit (for Production Support)

Help Desk Documentation

Unit Test

Integration Test

Browser Lab Test

Peer Code Review

Freeze Code

Performance & Load Test

Security Code Review

Data Migration

Production Deployment

Preliminary Acceptance Testing

Launch Monitoring Tools

User Training

Production Pilot

Update System FMEA

Transition to Production Support

Transition to Help Desk

GO LIVE

Performance and Usage Monitoring

Issues Log

Feedback Management

Bug Fixes and Further Optimization

Final Acceptance Testing and CTQ Measurement

Document & Share Best Practices

DMADOV Project Progress Overview

Define Measure Analyze Optimize Verify Design

Six Sigma Overview

Documents