quality management and quality planning

Quality Management and

Quality PlanningBy

Amartya Talukdar

Cost of Quality – 4 Categories

Early detection/prevention is less costly (Maybe by a factor of 10)

Ways of Improving Quality Plan-Do-Study-Act Cycle (PDSA)

Also called the Deming Wheel after originator

Circular, never ending problem solving process

Seven Tools of Quality Control Tools typically taught to problem solving

teams Quality Function Deployment

Used to translate customer preferences to design

PDSA Details Plan

Evaluate current process Collect procedures, data, identify problems Develop an improvement plan, performance

objectives Do

Implement the plan – trial basis Study

Collect data and evaluate against objectives Act

Communicate the results from trial If successful, implement new process

Seven Tools of Quality Control1. Cause-and-Effect Diagrams2. Flowcharts3. Checklists4. Control Charts5. Scatter Diagrams6. Pareto Analysis7. Histograms

Cause-and-Effect Diagrams Called Fishbone Diagram Focused on solving identified quality problem

Flowcharts Used to document the detailed steps in

a process Often the first step in Process Re-

Engineering

Control Charts Important tool used in Statistical

Process Control The UCL and LCL are calculated limits

used to show when process is in or out of control

Scatter Diagrams A graph that shows how two variables

are related to one another Data can be used in a regression analysis

to establish equation for the relationship

Pareto Analysis Technique that displays the degree of importance for

each element Named after the 19th century Italian economist; often

called the 80-20 Rule Principle is that quality problems are the result of only a

few problems e.g. 80% of the problems caused by 20% of causes

Histograms A chart that shows the frequency distribution

of observed values of a variable like service time

at a bank drive-up window Displays whether the distribution is

symmetrical (normal) or skewed

Product Design - Quality Function Deployment

Critical to ensure product design meets customer expectations

Useful tool for translating customer specifications into technical requirements is Quality Function Deployment (QFD)

QFD encompasses Customer requirements Competitive evaluation Product characteristics Relationship matrix Trade-off matrix Setting Targets

Quality Function Deployment(QFD) Details

Process used to ensure that the product meets customer specifications

Voice of theengineer

Voice of the

customer

Customer-basedbenchmarks

QFD - House of Quality

Adding trade-offs, targets & developing product specifications

Trade-offs

Targets

TechnicalBenchmarks

Reliability – critical to quality

Reliability is the probability that the product, service or part will function as expected

No product is 100% certain to function properly

Reliability is a probability function dependent on sub-parts or components


Reliability of a system is the product of component reliabilities

RS = (R1) (R2) (R3) . . . (Rn)RS = reliability of the product or

systemR1 = reliability of the components Increase reliability by placing

components in parallel


Increase reliability by placing components in parallel

Parallel components allow system to operate if one or the other fails

RS = R1 + (R2* Probability of needing 2nd component)

Process Management & Managing Supplier Quality

Quality products come from quality sources

Quality must be built into the process Quality at the source is belief that it is

better to uncover source of quality problems and correct it

TQM extends to quality of product from company’s suppliers

ISO Standards ISO 9000 Standards:

Certification developed by International Organization for Standardization

Set of internationally recognized quality standards

Companies are periodically audited & certified

ISO 9000:2000 QMS – Fundamentals and Standards ISO 9001:2000 QMS – Requirements ISO 9004:2000 QMS - Guidelines for

Performance More than 40,000 companies have been

certified ISO 14000:

Focuses on a company’s environmental responsibility

Why TQM Efforts Fail Lack of a genuine quality

culture Lack of top management

support and commitment Over- and under-reliance on

SPC methods

TQM Within OM TQM is broad sweeping organizational change TQM impacts

Marketing – providing key inputs of customer information

Finance – evaluating and monitoring financial impact Accounting – provides exact costing Engineering – translate customer requirements into

specific engineering terms Purchasing – acquiring materials to support product

development Human Resources – hire employees with skills

necessary Information systems – increased need for accessible

information

Quality Planning It is important to design in quality and

communicate important factors that directly contribute to meeting the customer’s requirements

Design of experiments helps identify which variables have the most influence on the overall outcome of a process

Many scope aspects of IT projects affect quality like functionality, features, system outputs, performance, reliability, and maintainability

Quality Assurance Quality assurance includes all the

activities related to satisfying the relevant quality standards for a project

Another goal of quality assurance is continuous quality improvement

Benchmarking can be used to generate ideas for quality improvements

Quality audits help identify lessons learned that can improve performance on current or future projects

Quality Assurance Plan

Pareto Analysis Pareto analysis involves identifying

the vital few contributors that account for the most quality problems in a system

Also called the 80-20 rule, meaning that 80% of problems are often due to 20% of the causes

Pareto diagrams are histograms that help identify and prioritize problem areas

Sample Pareto Diagram

Statistical Sampling and Standard Deviation

Statistical sampling involves choosing part of a population of interest for inspection

The size of a sample depends on how representative you want the sample to be

Sample size formula:Sample size = .25 X (certainty Factor/acceptable

error)2

Commonly Used Certainty Factors

Desired Certainty Certainty Factor

95% 1.960

90% 1.645

80% 1.281

95% certainty: Sample size = 0.25 X (1.960/.05) 2 = 38490% certainty: Sample size = 0.25 X (1.645/.10)2 = 6880% certainty: Sample size = 0.25 X (1.281/.20)2 = 10

Six Sigma DefinedSix Sigma is “a comprehensive and flexible system for achieving, sustaining and maximizing business success. Six Sigma is uniquely driven by close understanding of customer needs, disciplined use of facts, data, and statistical analysis, and diligent attention to managing, improving, and reinventing business processes.”**Pande, Peter S., Robert P. Neuman, and Roland R. Cavanagh, TheSix Sigma Way. New York: McGraw-Hill, 2000, p. xi

Basic Information on Six Sigma The target for perfection is the

achievement of no more than 3.4 defects per million opportunities

The principles can apply to a wide variety of processes

Six Sigma projects normally follow a five-phase improvement process called DMAIC

DMAIC Define: Define the problem/opportunity,

process, and customer requirements Measure: Define measures, collect,

compile, and display data Analyze: Scrutinize process details to

find improvement opportunities Improve: Generate solutions and ideas

for improving the problem Control: Track and verify the stability of

the improvements and the predictability of the solution

How is Six Sigma Quality Control Unique? It requires an organization-wide

commitment Six Sigma organizations have the ability

and willingness to adopt contrary objectives, like reducing errors and getting things done faster

It is an operating philosophy that is customer-focused and strives to drive out waste, raise levels of quality, and improve financial performance at breakthrough levels

Examples of Six Sigma Organizations Motorola, Inc. pioneered the adoption of

Six Sigma in the 1980s and saved about $14 billion

Allied Signal/Honeywell saved more than $600 million a year by reducing the costs of reworking defects and improving aircraft engine design processes

General Electric uses Six Sigma to focus on achieving customer satisfaction

Six Sigma and Project Management

Joseph M. Juran stated that “all improvement takes place project by project, and in no other way”

It’s important to select projects carefully and apply higher quality where it makes sense

Six Sigma projects must focus on a quality problem or gap between current and desired performance and not have a clearly understood problem or a predetermined solution

After selecting Six Sigma projects, the project management concepts, tools, and techniques described in this text come into play, such as creating business cases, project charters, schedules, budgets, etc.

Six Sigma and Statistics The term sigma means standard

deviation Standard deviation measures how much

variation exists in a distribution of data Standard deviation is a key factor in

determining the acceptable number of defective units found in a population

Six Sigma projects strive for no more than 3.4 defects per million opportunities

Standard Deviation A small standard deviation means

that data cluster closely around the middle of a distribution and there is little variability among the data

A normal distribution is a bell-shaped curve that is symmetrical about the mean or average value of a population

Normal Distribution and Standard Deviation

Six Sigma as a PhilosophyInternal &ExternalFailure Costs

Prevention &Appraisal

Costs

Old Belief4sC

osts

Internal &ExternalFailure Costs

Prevention &Appraisal

Costs

New BeliefCos

ts4s

5s

6s

Quality

Quality

Old Belief

High Quality = High Cost

New Belief

High Quality = Low Cost

s is a measure of how much variation exists in a process

Focus: The End User Customer: Internal or External Consumer: The End User

the “Voice of the Consumer” (Consumer Cue)must be translated into

the “Voice of the Engineer” (Technical Requirement)

Six Sigma as a Metric

CENSO

RED

1)( 2

n

xxiU

sSigma = s = Deviation ( Square root of variance )

-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7

Axis graduated in Sigma

68.27 %

95.45 %

99.73 %

99.9937 %

99.999943 %

99.9999998 %

result: 317300 ppm outside (deviation)

45500 ppm

2700 ppm

63 ppm

0.57 ppm

0.002 ppm

between + / - 1sbetween + / - 2sbetween + / - 3sbetween + / - 4sbetween + / - 5sbetween + / - 6s

s =

Six Sigma as a ToolProcess Mapping Tolerance Analysis

Structure Tree Components Search

Pareto Analysis Hypothesis Testing

Gauge R & R Regression

Rational Subgrouping DOE

Baselining SPC

Many familiar quality tools applied in a structured methodology

Distinguish “Vital Few”from “Trivial Many”

Y = Dependent Variable Output, Defectx = Independent Variables Potential Causex* = Independent Variable Critical Cause

Define the Problem / Defect Statement

Y = f ( x1*, x2, x3, x4

*, x5. . . Xn)

Process(Parameters)

Material

Methods

People

Environment

OutputMachine

Measurements

Strategy by Phase - Phase

Measure(What)

Analyze(Where, When, Why)

Improve(How)

Control(Sustain, Leverage)

Step

What is the frequency of Defects?• Define the defect• Define performance standards• Validate measurement system• Establish capability metric

Where, when and why do Defects occur?• Identify sources of variation• Determine the critical process parameters

How can we improve the process?• Screen potential causes• Discover relationships• Establish operating tolerancesWere the improvements effective?• Re-establish capability metric

How can we maintain the improvements?• Implement process control mechanisms • Leverage project learning's• Document & Proceduralize

Focus

YYYY

XVital X

XVital XVital X

Y, Vital X

Y, Vital X

Process Characterization

Process Optimization

Measure

Improve

Ana

lyze

Control

Measure

Improve

Ana

lyze

Control

Measure

Improve

Ana

lyze

Control

Measure

Improve

Ana

lyze

Control

Measure

Ana

lyze

Improvement

Control

A Black Belt has…, and will…

Leadership

Stat

istica

l,

Qualit

y Skil

l

Interpersonal Skill

Driving the UseTh

e new

of S

ix Si

gma

Mentoring

http://www.mot.com/

Black Belt TrainingTask

Time on Consulting/Training

Mentoring Related Projects

Green Belt

Utilize Statistical/Quality technique

2%~5%Find one new green belt

2 / year

Black Belt

Lead use of technique and communic-ate new ones

5%~10% Two green belts 4 / year

Master Black Belt

Consulting/Mentoring/Training

80~100% Five Black Belts 10 / year

Barrier Breakthrough Plan

1.00

10.00

100.00

J94

DPM

Op

6 Sigma

SIGMA

J95 J96 J97

6

5.65

5.6

5.5

5.4

5.3

MY95 MY96 MY97 MY98

Pareto, Brainstorming, C&E, BvC8D, 7D, TCS Teams, SPC

DOE, DFM, PCRenewBlack Belt Program (Internal Motorola)

Black Belt Program (External Suppliers)

Proliferation of Master Black Belts

http://www.mot.com/

MeasureCharacterize Process

Understand ProcessEvaluate

Improve and Verify ProcessImprove

Maintain New ProcessControl

DefineProblem

UnderstandProcess

CollectData

ProcessPerformance

Process Capability- Cp/Cpk- Run Charts

Understand Problem(Control or Capability)

Data Types- Defectives- Defects- Continuous

Measurement Systems Evaluation (MSE)

Define Process- Process Mapping

Historical Performance

Brainstorm Potential Defect Causes

Defect Statement

Project Goals

Understand the Process and Potential Impact

Measure Phase

ReduceComplaints (int./ext.)

ReduceCost

ReduceDefects

Problem Definitions need to be based onquantitative facts supported by analytical data.

What are the Goals?

What do you want to improve? What is your ‘Y’?

Problem Definition

Baselining: Quantifying the goodness (or badness!) of the current process, before ANY improvements are made, using sample data. The key to baselining is collecting representative sample data

Sampling Plan- Size of Subgroups- Number of Subgroups- Take as many “X” as possible into consideration

Time

How do we know our process?

Process Map

Historical Data

Fishbone

RATIONAL SUBROUPING Allows samples to be taken that include only white noise, within the samples. Black noise

occurs between the samples.

RATIONAL SUBGROUPSMinimize variation within subroups

Maximize variation between subrgoups

TIME

PR

OC

ES

S

RE

SP

ON

SE

WHITE NOISE (Common Cause Variation)

BLACK NOISE (Signal)

Visualizing the Causes

s st + sshift = stotal

Time 1

Time 2Time 3

Time 4

Within Group

• Called s short term (sst)• Our potential – the best

we can be• The s reported by all 6

sigma companies• The trivial many

Visualizing the Causes

Between Groups

s st + sshift = stotal

Time 1

Time 2Time 3

Time 4 • Called sshift (truly a measurement in sigmas of how far the mean has shifted)

• Indicates our process control

• The vital few

Assignable Cause Outside influences Black noise Potentially controllable How the process is actually performing

over time

Fishbone

Common Cause Variation Variation present in every process Not controllable The best the process can be within the

present technology

Data within subgroups (Z.st) will contain only Common Cause Variation

s2Total = s2

Part-Part + s2R&R

Recommendation:

Resolution £10% of tolerance to measureGauge R&R £ 20% of tolerance to measure

• Repeatability (Equipment variation)Variation observed with one measurement device when used several times by one operator while measuring the identical characteristic on the same part.• Reproducibility (Appraised variation)Variation Obtained from different operators using the same device when measuring the identical characteristic on the same part.• Stability or DriftTotal variation in the measurement obtained with a measurement obtained on the same master or reference value when measuring the same characteristic, over an extending time period.

Gauge R&R

Part-PartR&R

To understand where you want to be, you need to know how to get there.v

Map the Process

Measure the Process

Identify the variables - ‘x’

Understand the Problem -’Y’ = function of variables -’x’

Y=f(x)

In many cases, the data sample can be transformed so that it is approximately normal. For example, square roots, logarithms, and reciprocals often take a positively skewed distribution and convert it to something close to a bell-shaped curve

What do we Need?LSL USL LSL USL

LSL USL

Off-Target, Low VariationHigh Potential DefectsGood Cp but Bad Cpk

On TargetHigh VariationHigh Potential DefectsNo so good Cp and Cpk

On-Target, Low VariationLow Potential DefectsGood Cp and Cpk

Variation reduction and process centering create processes with less potential for defects.

The concept of defect reduction applies to ALL processes (not just manufacturing)

Eliminate “Trivial Many”

Qualitative Evaluation Technical Expertise Graphical Methods Screening Design of Experiments Identify “Vital Few”

Pareto Analysis Hypothesis Testing Regression Design of Experiments

Quantify Opportunity

% Reduction in Variation Cost/ Benefit Our Goal:

Identify the Key Factors (x’s)

Graph>Box plot

75%

50%

25%

Graph>Box plot

Without X values

DBP

Box plots help to see the data distribution

Day

DBP

109

104

99

94

109

104

99

94Operator

DBP

109

104

99

94

Shift

DBP

109

104

99

94

Statistical Analysis

0.0250.0200.0150.0100.0050.000

7

6

5

4

3

2

1

0

New Machine

Freq

uenc

y

0.0250.0200.0150.0100.0050.000

30

20

10

0

Machine 6 mths

Freq

uenc

y

Is the factor really important?

Do we understand the impact for the factor?

Has our improvement made an impact

What is the true impact?

Hypothesis Testing

Regression Analysis

5545352515 5

60

50

40

30

20

10

0

X

Y

R-Sq = 86.0 %Y = 2.19469 + 0.918549X

95% PI

Regression

Regression Plot

Compare

Sample

Means

& Variances

Identify

Relationships

Establish

Limits

Apply statistics to validate actions & improvements

A- Poor Control, Poor ProcessB- Must control the Process better, Technology is fineC- Process control is good, bad Process or technologyD- World Class

A B

C DTECHNOLOGY

1 2 3 4 5 6goodpoor

2.52.01.51.00.5CO

NTRO

L

Zshift

Z St

poor

good

Con-sumer Cue

Technical Requirement

Preliminary Drawing/Database

Identity CTQs

Identify Critical Process

Obtain Data on Similar Process

Calculate Z values

Rev 0 Drawings

Stop Adjust process & design

1st piece inspection

Prepilot Data

Pilot data

Obtain dataRecheck ‘Z’ levels

Stop Fix process & design

Z<3

Z>= Design Intent M.A.I.C

M.A.DSix Sigma Design Process

• #1 Define the customer Cue and technical requirement we need to satisfy

Consumer Cue: Blocks Cannot rattle and must not interfere with box

Technical Requirement: There must be a positive Gap

Reliability (Level)Time to installTotal Electricity Usage Req'd for SystemTime to RepairFloor space occupied NO INPUT IN THIS AREASensor Resolution Response time to power lossVoltagePowerTime to supply Backup Powerbackup power capacity (time)Sensor SensitivityFloor LoadingTime Between maintenanceTime between equipment replacementSafety Index RatingCost of investmentCost of maintenaceCost of installationYears in Mainstream marketCustomer Support RatingDependency on weather conditionsECO-ratingHours of training req'd

Preferred up dwn dwn dwn dwn dwn dwn tgt tgt dwn dwn tgt dwn up up

Engineering Metrics

Customer Requirements Cus

tom

er W

eigh

ts

Relia

bilit

y (L

evel

)

Tim

e to

insta

ll

Tota

l Ele

ctric

ity U

sage

Req

'd fo

r Sys

tem

Tim

e to

Rep

air

Floo

r spa

ce o

ccup

ied

Sens

or R

esol

utio

n

Resp

onse

tim

e to

pow

er lo

ss

Volta

ge

Pow

er

Tim

e to

sup

ply

Back

up P

ower

back

up p

ower

cap

acity

(tim

e)

Sens

or S

ensi

tivity

Floo

r Loa

ding

Tim

e Be

twee

n m

aint

enan

ce

Tim

e be

twee

n eq

uipm

ent r

epla

cem

ent

Safe

ty In

dex

Rat

ing

Cos

t of i

nves

tmen

t

Cos

t of m

aint

enac

e

Cos

t of i

nsta

llatio

n

Year

s in

Mai

nstr

eam

mar

ket

1 Fast Response 92 Long time of backup power supply 93 Low environmental impact 94 Safe to operate 95 Meet power requirements 96 Low investment cost 3 7 Occupies small floor space 38 Easy to upgrade 39 Low upgrading costs 3

10 Low time to implement 311 Cheap to maintain 312 Low recovery or cycle time 313 Long life cycle of the system/component 114 Cheap to operate 115 Cheap to install 116 Long Existing proven technology 1

QFD, F

MEA, R

TY

• #2 Define the target dimensions (New designs) or process mean (existing design) for all mating Parts

Gap Must Be T=.011, LSL=.001 and USL = .021

Gap

(-) (-) (-) (-)(+)

Step #3 • Gather process capability data.• Use actual or similar part data to calculate SS of largest contributors.• May use expert data for minimal contributors• Do not calculate s from current tolerances

Gap Requirements

mT = .010USL = .020 LSL = .001

(-) (-) (-) (-)(+)

mgap= mbox – mcube1 – mcube2 – mcube3 – mcube4

sgap = s2box + s2

cube1 + s2cube2 + s2

cube3 + s2cube4

Short Termmgap= 5.080 – 1.250 – 1.250 – 1.250 – 1.250.016

sgap = (.001)2 + (.001)2 + (.001)2 + (.001)2 + (.001)2 = .00224

Long Termsgap = (.0015)2 + (.0015)2 + (.0015)2 + (.0015)2 + (.0015)2 = .00335

From process:Average sst

Cube 1.250 .001Box 5.080 .001

Zshift = 1.6

Design of Experiments (DOE) To estimate the effects of independent

Variables on Responses.

Terminology Factor – An independent variable Level – A value for the factor. Response - Outcome

X YPROCESS

THE COFFEE EXAMPLE

Low HighCoffee Brand Maxwell House Chock Full o Nuts Water Spring TapCoffee Amount 1 2

LevelFactor

Main Effects: Effect of each individual factor on response

Bean ‘A’ Bean ‘B’

Tast

e

2.2

3.7

ME

Bean ‘A’Temp ‘X’

Bean ‘B’Temp ‘Y’

Tast

e

Interaction

Concept of Interaction

Why use DoE ? Shift the average of a process.

Reduce the variation.

Shift average and reduce variation

x1 x2

High LowAdhesion Area (cm2) 15 20Type of Glue Acryl UrethanThickness of Foam Styrene Thick ThinThickness of Logo Thick ThinAmount of pressure Short LongPressure application time Small BigPrimer applied Yes No

LevelFactor

Mini Case - NISSAN MOTOR COMPANY

++ +

+ +- -

-

++ -

- +- +

+

-- +

+ +- -

-

-- -

- +- +

+

12345678

9.8

Design ArrayA B C DNo Gluing Str

8.99.28.9

12.313

13.912.6

A - Adhesion Area (cm2)B - Type of GlueC - Thickness of Foam StyreneD - Thickness of Logo

`

A B C D+ 4.60 5.50 5.65 5.58- 6.48 5.58 5.43 5.50

Effect Tabulation

＋ - ＋ - ＋ - ＋ -

Gl u

ing

Str e

ngt h

AdhesionArea Type of Glue Thk of Foam

StyreneThk of logo

Factor Effect Plot

4.6

6.5

5.5

5.58 5.65

5.43

5.58

5

1. Define Objective.2. Select the Response (Y)3. Select the factors (Xs)4. Choose the factor levels5. Select the Experimental Design6. Run Experiment and Collect the Data7. Analyze the data8. Conclusions9. Perform a confirmation run.

STEPS IN PLANNING AN EXPERIMENT

CONTROL PHASE - SIX SIGMA

Control Phase Activities:

- Confirmation of Improvement- Confirmation you solved the practical problem- Benefit validation- Buy into the Control plan- Quality plan implementation- Procedural changes- System changes- Statistical process control implementation- “Mistake-proofing” the process- Closure documentation- Audit process

-Scoping next project

CONTROL PHASE - SIX SIGMA

How to create a Control Plan:

1. Select Causal Variable(s). Proven vital few X(s)2. Define Control Plan - 5Ws for optimal ranges of X(s)3. Validate Control Plan - Observe Y4. Implement/Document Control Plan5. Audit Control Plan6. Monitor Performance Metrics

CONTROL PHASE - SIX SIGMAControl Plan Tools:

1. Basic Six Sigma control methods. - 7M Tools: Affinity diagram, tree diagram, process decision program charts, matrix diagrams, interrelationship diagrams, prioritization matrices, activity network diagram.

2. Statistical Process Control (SPC) - Used with various types of distributions - Control Charts

• Attribute based (np, p, c, u). Variable based (X-R, X)• Additional Variable based tools

-PRE-Control-Common Cause Chart (Exponentially Balanced Moving Average (EWMA))

PRODUCT MANAGEMENT

OVERALL GOAL OF

SOFTWARE

KNOWLEDGE OF COMPETITORS

SUPERVISION

PRODUCT DESIGN

PRODUCT MANAGEMENT

PRODUCT DESIGN

PRODUCT MANAGEMENT

INNOVATION

OUTPUT

DIRECTORY ORGANIZATION

INTUITIVE ANSWERS

SUPPORT

METHODS TO MAKE EASIER FOR USERS

CHARACTERISTICS:• Organizing ideas into meaningful

categories• Data Reduction. Large numbers of qual.

Inputs into major dimensions or categories.

AFFINITY DIAGRAM

MATRIX DIAGRAM

Pat

ient

sch

edul

ed

Atte

ndan

t ass

igne

d

Atte

ndan

t arri

ves

Obt

ains

equ

ipm

ent

Tran

spor

ts p

atie

nt

Pro

vide

The

rapy

Not

ifies

of r

etur

n

Atte

ndan

t ass

igne

d

Atte

ndan

t arri

ves

Pat

ient

retu

rned

Arrive at scheduled time 5 5 5 5 1 5 0 0 0 0 0Arrive with proper equipment 4 2 0 0 5 0 0 0 0 0 0Dressed properly 4 0 0 0 0 0 0 0 0 0 0Delivered via correct mode 2 3 0 0 1 0 0 0 0 0 0Take back to room promptly 4 0 0 0 0 0 0 5 5 5 5

IMPORTANCE SCORE 39 25 25 27 25 0 20 20 20 20RANK 1 3 3 2 3 7 6 6 6 6

5 = high importance, 3 = average importance, 1 = low importance

HOWS

WHATS

RELATIONSHIP MATRIX

CUSTOMER IMPORTANCE MATRIX

Add

feat

ures

Mak

e ex

istin

g pr

oduc

t fas

ter

Mak

e ex

istin

g pr

oduc

t eas

ier t

o us

e

Leav

e as

-is a

nd lo

wer

pric

e

Dev

ote

reso

urce

s to

new

pro

duct

s

Incr

ease

tech

nica

l sup

port

budg

et

Out

arro

ws

In a

rrow

s

Tota

l arro

ws

Stre

ngth

Add features 5 0 5 45Make existing product faster 2 1 3 27Make existing product easier to use 1 2 3 21Leave as-is and lower price 0 3 3 21Devote resources to new products 1 1 2 18Increase technical support budget 0 2 2 18

(9) = Strong Influence

(3) = Some Influence

(1) = Weak/possible influence

Means row leads to column item

Means column leads to row item

COMBINATION ID/MATRIX DIAGRAM

CHARACTERISTICS:• Uncover patterns in

cause and effect relationships.

• Most detailed level in tree diagram. Impact on one another evaluated.

CONTROL PHASE - SIX SIGMAControl Plan Tools:

1. Basic Six Sigma control methods. - 7M Tools: Affinity diagram, tree diagram, process decision program charts, matrix diagrams, interrelationship diagrams, prioritization matrices, activity network diagram.

2. Statistical Process Control (SPC) - Used with various types of distributions - Control Charts

• Attribute based (np, p, c, u). Variable based (X-R, X)• Additional Variable based tools

-PRE-Control-Common Cause Chart (Exponentially Balanced Moving Average (EWMA))

How do we select the correct Control Chart:

Type Data

Ind. Meas. or subgroups

Normally dist. data

Interest in sudden mean changes

Graph defects of defectives

Oport. Area constant from sample to sample

X, Rm

p, np

X - RMA, EWMA or CUSUM and Rm

u

C, u

Size of the subgroup constant

p

If mean is big, X and R are effective too

Ir neither n nor p are small: X - R, X - Rm are effective

More efective to detect gradual changes in long term

Use X - R chart with modified rules

VariablesAttributes

Measurement of subgroups

Individuals

Yes

No No

Yes

Yes

No

Yes

No

Defects Defectives

Additional Variable based tools:1. PRE-Control

• Algorithm for control based on tolerances• Assumes production process with measurable/adjustable

quality characteristic that varies.• Not equivalent to SPC. Process known to be capable of

meeting tolerance and assures that it does so.• SPC used always before PRE-Control is applied.• Process qualified by taking consecutive samples of individual

measurements, until 5 in a row fall in central zone, before 2 fall in cautionary. Action taken if 2 samples are in Cau. zone.

• Color coded

YELLOW ZONE

GREEN ZONE

YELLOW ZONE

RED ZONE

RED ZONE

1/4 TOL. 1/2 TOL. 1/4 TOL.

Low

Tole

ran

ce L

imt

Tole

ran

ce L

imt

High

Refe

renc

e Lin

e

PRE-

Cont

rol

DIM

ENSI

ONNO

MIN

AL

Refe

renc

e Lin

e

PRE-

Cont

rol

2. Common Causes Chart (EWMA).• Mean of automated manufacturing processes drifts because of

inherent process factor. SPC consideres process static.• Drift produced by common causes.• Implement a “Common Cause Chart”.• No control limits. Action limits are placed on chart.

• Computed based on costs• Violating action limit does not result in search for special

cause. Action taken to bring process closer to target value.

• Process mean tracked by EWMA • Benefits:

• Used when process has inherent drift• Provide forecast of where next process measurement will be.• Used to develop procedures for dynamic process control

• Equation: EWMA = y^t + s (yt - y^t) s between 0 and 1

EWMA chart of sand temperature

0

50

100

150

1 4 7

10 13 16 19 22 25 28

Observations

Degr

ees Sand

TemperatureEWMA

Sand Temperature EWMA Error

125 125.00 0.00123 125.00 -2.00118 123.20 -5.20116 118.52 -2.52108 116.25 -8.25112 108.83 3.17101 111.68 -10.68100 102.07 -2.0792 100.21 -8.21

102 98.22 3.78111 101.62 9.38107 110.60 -3.60112 107.30 4.70112 111.53 0.47122 111.95 10.05140 121.00 19.00125 138.00 -13.00130 126.31 3.69136 129.63 6.37130 135.36 -5.36112 130.54 -18.54115 113.85 1.15100 114.89 -14.89113 101.49 11.51111 111.85 -0.85

Project Closure

• Improvement fully implemented and process re-baselined.• Quality Plan and control procedures institutionalized.• Owners of the process: Fully trained and running the process.• Any required documentation done.• History binder completed. Closure cover sheet signed.• Score card developed on characteristics improved and reporting

method defined.

The Cost of Quality The cost of quality is

the cost of conformance or delivering products that meet requirements and fitness for use

the cost of nonconformance or taking responsibility for failures or not meeting quality expectations

Five Cost Categories Related to Quality

Prevention cost: the cost of planning and executing a project so it is error-free or within an acceptable error range

Appraisal cost: the cost of evaluating processes and their outputs to ensure quality

Internal failure cost: cost incurred to correct an identified defect before the customer receives the product

External failure cost: cost that relates to all errors not detected and corrected before delivery to the customer

Measurement and test equipment costs: capital cost of equipment used to perform prevention and appraisal activities

Maturity Models Maturity models are frameworks for

helping organization improve their processes and systems Software Quality Function Deployment

model focuses on defining user requirements and planning software projects

The Software Engineering Institute’s Capability Maturity Model provides a generic path to process improvement for software development

Several groups are working on project management maturity models, such as PMI’s Organizational Project Management Maturity Model (OPM3)

THANK YOU