07 cm0471 module 7

8/22/2019 07 CM0471 Module 7

http://slidepdf.com/reader/full/07-cm0471-module-7 1/35

Six Sigma Black Belt

Cert. Prep. Course:ImproveImproveImproveImproveModule VII

©2009 ASQ 2

Agenda

This module consists of seven lessons:

1. Design of experiments (DOE)

2. Waste elimination3. Cycle-time reduction

4. Kaizen and kaizen blitz5. Theory of constraints (TOC)

6. Implementation

7. Risk analysis and mitigation

©2009 ASQ 3

Lesson 1 – Measuring and ModelingRelationships Between VariablesVII. A.1 Terminology

Define basic DOE terms, including independent anddependent variables, factors and levels, response,

treatment, error, etc. (Understand)

8/22/2019 07 CM0471 Module 7


©2009 ASQ 4

DOE Terminology

• Certain terms used with Design of Experiments

need to be defined clearly.

– Independent Variable

• A variable that can contribute to theexplanation of the outcome of an experiment.

• This is also known as a “predictor variable.”

– Dependent Variable

• A variable representing the outcome of anexperiment or response.

• The response is often referred to as the

“output variable” or “response variable.”

©2009 ASQ 5

– Factors

• A predictor (or independent) variable that isvaried with the intent of assessing its effect

on a response (or dependent) variable.

• Most often referred to as an “input variable.”

• A factor may be quantitative such as atemperature in degrees F or cycle time

in pieces per minute.• A factor may also be qualitative, such as a

different production machine or a differentmold.

DOE Terminology

©2009 ASQ 6

– Levels

• A specific setting for a factor.

• In DOE, levels are frequently set as high and lowfor each factor.

• A level is a potential setting, value, or assignment

of a factor of the value of the predictor variable.

• If the experiment is to be performed at twodifferent temperatures, then the “factor”

temperature has two levels.

• For example, if the factor is time, then thelow level may be 50 minutes, and the high

level may be 70 minutes.

DOE Terminology

8/22/2019 07 CM0471 Module 7


©2009 ASQ 7

– Response

• The output(s) of a process. Sometimes calleddependent variable(s).

• It shows the observed result or value of anexperimental treatment.

– Treatment

• The specific setting of factor levels for anexperimental unit.

• For example, a level of temperature at 65°Cand a level of time at 45 minutes describe atreatment as it relates to an output of yield.

DOE Terminology

©2009 ASQ 8

– Experimental Run• Is a single performance of the experiment for

a specific set of treatment combinations

• n = Number of experimental runs that will beconducted in a given experiment

• If all of the factors have the same number oflevels:

– n = Lf

– n = number of runs – L = number of levels – f = number of factors

• If the factors have different numbers of levels: – n = (L1)(L2)(L3) … (Lf)

DOE Terminology

©2009 ASQ 9

– Error

• An error from an experiment reveals variationin the outcome of identical tests.

• It is the unexplained variation in a collectionof observations.

• An error is variation in the response variablebeyond that accounted for by the factors orother assignable sources while conducting

an experiment.

• Many times referred to as “experimentalerror.”

DOE Terminology

8/22/2019 07 CM0471 Module 7


©2009 ASQ 10

Progress Check

Match each DOE term to the appropriate definition

The output of a process.Factors

Unexplained variation in the processTreatment

Process inputs an investigator

manipulates to cause a changein the output

Response

A specific combination of factor levelsError

DefinitionTerm

©2009 ASQ 12

Progress Check

Which of the descriptions in the table below is anexample of DOE terminology?

©2009 ASQ 14

Define and apply DOE principles, including power andsample size, balance, repetition, replication, order,efficiency, randomization, blocking, interaction,

confounding, resolution, etc. (Apply)

VII.A.2 Design principles

Lesson 1 – Measuring and ModelingRelationships Between Variables

8/22/2019 07 CM0471 Module 7


©2009 ASQ 15

Design Principles

Sample not rejected Sample rejected

Sample should not be rejected

CorrectDecision

Type II orconsumer’s risk

β = P(Type II)

Sample should be rejected

Type I orproducer’s risk

α = P(Type I)

Correct

Decision

Action Taken

State of Nature

α is the risk of finding a difference when there really isn’t one

β is the risk of not finding a difference when there really is one.

Power and Sample Size: This was covered in the Analyze PhaseAs the power of the test increases, the probability of a Type II error β decreases

©2009 ASQ 16

Design Principles

Replication:Performing the entire

experiment more thanonce for a given set of

independent variables

Speed Pressure MPG

fast High 23.0

slow Low 23.5

slow High 25.0

fast Low 22.0

fast High 23.4

slow Low 23.8

slow High 25.5

fast Low 22.5

fast High 22.8

slow Low 23.0

slow High 24.8

fast Low 21.7

Replication is good for producing enoughdata to investigate error of the experiment.

Replication

©2009 ASQ 17

Repeated Measures is also known as Repetition:Is a repeat of the measurements taken for the samerun

Repeat Measures help determine the inherentvariability in the measurement system

Design Principles

8/22/2019 07 CM0471 Module 7


©2009 ASQ 18

Confounding:

• When factors or interactions are not distinguishablefrom one another

• Two or more effects that cannot beunambiguously attributed to a single factor or interaction. This isalso referred to as aliasing.

• Confounding indicates that the value of a maineffect comes from both the main effect itself andalso from interactions.

Design Principles

©2009 ASQ 19

Order:

• Refers to the sequence in which the runs in theexperiment will be conducted

Run Order: Typically a random order inwhich the experiment was run.

Standard Order: In a standard order, thefirst factor alternates between the low andhigh setting for each run. The second factor

alternates between low and high settings.

Design Principles

©2009 ASQ 20

Randomization:

• A schedule for conducting treatment combinations ina DOE such that the conditions in one run neither

depend on the conditions of the previous run norpredict the conditions in the subsequent runs.

• Randomization in an experiment reduces the effects

of factors (variables) outside the experiment. Whenoutside factors (called “noise”) affect an experiment,we may draw false conclusions regarding the factorsin the experiment.

Design Principles

8/22/2019 07 CM0471 Module 7


©2009 ASQ 21

Blocking:

Blocking allows unwantedeffects to be separated from

desired effects!

• Is a group ofhomogeneousexperimental units.

• Variation betweenblocks should begreater than variation

within blocks.In this example, the twoblocks might represent Day 1and Day 2. Note: each factor

level appearstwice withineach block.

Blocks Speed Pressure Temp

2 slow high cold

2 slow low hot

2 fast high hot

2 fast low cold

1 fast high cold

1 slow high hot

1 fast low hot

1 slow low cold

See CSSBB HB, page 299

Design Principles

©2009 ASQ 22

T = 50 T = 70 P = 1 P = 3

50 1 45 45 45

70 1 66 66 66

50 3 29 29 29

70 3 88 88 88

50 1 43 43 43

70 1 63 63 63

50 3 32 32 32

70 3 91 91 91

37.3 77.0 54.3 60.0Factor Level Mean Response or Effect >>

Factor Level

Temperature Pressure

Yield

(Response)

7050

80

70

60

50

40

31

Temperature

M e a n

Pressure

Main Effects Plot for YieldData Means

An effect is the differencebetween two means

See CSSBBHB, page 300

Main Effects

Design Principles

©2009 ASQ 23

50 1 45 45

70 1 66 66

50 3 29 29

70 3 88 88

50 1 43 43

70 1 63 63

50 3 32 32

70 3 91 91

44.0 30.5 64.5 89.5

T = 70 andP =3

Mean Response >>

Temperature PressureYield

(Response)T = 50 and

P = 1T = 50 and

P = 3T = 70 and

P =1

31

90

80

70

60

50

40

30

Pressure

M e a n

50

70

Temperature

InteractionPlotfor YieldData Means

31

90

75

60

45

30

7050

90

75

60

45

30

Temperature

Pressure

50

70

Temperature

1

3

Pressure

InteractionPlotfor YieldData Means

See CSSBB HB, page 303

Interactions

Design Principles

8/22/2019 07 CM0471 Module 7


©2009 ASQ 24

Graphical representations of possibilities

Magnitude of effect

( ) ( ) above)figure(See ] ∆- ∆[2

1 =Effect(X x Y)

scoordinateCartesianarepairs XY thewhere

)] Y X - Y (X -) Y X - Y [(X 2

1 =Effect(X x Y)

21

LOLOLOHIHILOHIHI

∆∆∆∆

1∆∆∆∆

2

Strength of Interaction

Design Principles

©2009 ASQ 25

1-1

5

4

3

2

1

Y

M e a n

-1

1

X

Interaction Plot for RESPONSEData Means

Effect = [(4 – 2 ) – (5 – 1)] /2 = -1

1-1

5

4

3

2

1

Y

M e a n

-1

1

X

Interaction Plot for RESPONSEData Means

Effect = [(2 – 4 ) – (5 – 1)] /2 = -3

This value is negative!

When the Lines Cross!

Design Principles

©2009 ASQ 26

Balanced designs are mucheasier to analyze!

Balanced Design

Balanced Design:Each experimental level for any one factor isrepeated the same number of times for all possible

combinations involving the levels of the other factors.Balanced Design

Run Order Speed Pres sure

1 Slow Low

2 Slow High

3 Fast Low

4 Fast High

Unbalanced DesignRun Order Speed Pressure

1 Slow Low

2 Slow High

3 Slow Low

4 Fast High

See CSSBB HB, Page 306

Design Principles

8/22/2019 07 CM0471 Module 7


©2009 ASQ 27

Fraction Runs Resolution

Full 64 Full 26

½ 32 VI 2(6-1)

¼ 16 IV 2(6-2)

1/8 8 III 2(6-3)

Designation

Resolution:

• It is the level of confounding in a fractional

factorial design.

• They are numbered using Roman Numerals

Design Principles

©2009 ASQ 28

Resolution III Designs:Designs in which no main effects are aliased with any othermain effect…

...But main effects are aliased with two-factor interactions,and 2-factor interactions may be aliased with each other.

Resolution IV Designs:Designs in which no main effect is aliased with any other maineffect or with any two-factor interaction…

…But two-factor interactions are aliased with other

interactions.

Resolution V and above Designs:Designs in which no main effect or two-factor interactions arealiased with any other main effect or two-factor interaction…

…But two-factor interactions are aliased with three-factorinteractions.

Design Principles

©2009 ASQ 29

A B C AB AC BC ABC

1 -1 1 -1 1 -1 -1

1 -1 -1 -1 -1 1 1

-1 -1 -1 1 1 1 -1

1 1 -1 1 -1 -1 -1

-1 1 -1 -1 1 -1 1

-1 -1 1 1 -1 -1 1

1 1 1 1 1 1 1

-1 1 1 -1 -1 1 -1

23 Factorial Matrix Let ABC = D

D is said to be confounded or aliased with the ABC interaction

AB = AxB ABC = AxBxC

First, we will use a typical three-factor design and use thethree-way higher interaction to substitute for the fourth factor:

Resolution IV Designs

Design Principles

8/22/2019 07 CM0471 Module 7


©2009 ASQ 30

I

Generator Generator Generator Generator

I=ABCD

A=BCD

B=ACD

C=ABD

D=ABC

AB=CD

AC=BD

AD=BC

A B C AB AC BC D (ABC) AD BD CD ABD ACD BCD ABCD1 -1 1 -1 1 -1 -1 -1 1 -1 1 -1 1 1

1 -1 -1 -1 -1 1 1 1 -1 -1 -1 -1 1 1

-1 -1 -1 1 1 1 -1 1 1 1 -1 -1 -1 1

1 1 -1 1 -1 -1 -1 -1 -1 1 -1 1 1 1

-1 1 -1 -1 1 -1 1 -1 1 -1 -1 1 -1 1

-1 -1 1 1 -1 -1 1 -1 -1 1 1 -1 -1 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1

-1 1 1 -1 -1 1 -1 1 -1 -1 1 1 -1 1

Minitab Alias Structure

I + ABCD A + BCDB + ACDC + ABD

D + ABC AB + CD AC + BD AD + BC

A=I*A=A2BCD=BCD

AB=I*AB=A2B2CD=CD

Expanding the table...

Expanded Resolution IV Designs

Design Principles

©2009 ASQ 31

Progress CheckWhich of the descriptions in the table below are examplesof main effects or interactions? Check the appropriate boxat the right.

Description/Name ExampleNon-

Example

A two-factor DOE with two interactions.

A three-factor DOE with A x B, A x C, B x C, and A x B xC interactions.

A four-factor DOE with 2 x 4 = 8 main effects plots

A five-factor DOE with a highest order interaction ofA x B x C x D

A DOE with factors A, B, and C and four main effectsplots

A DOE with factors A and B and only an A x B interaction

Completing an analysis of variance to plot four-way

interactions

©2009 ASQ 33

Progress Check

8/22/2019 07 CM0471 Module 7


©2009 ASQ 34

Plan, organize, and evaluate experiments bydetermining the objective, selecting factors,responses and measurement methods, choosingthe appropriate design, etc. (Evaluate)

VII.A.3 Planning experiments


©2009 ASQ 35

Planning Experiments

Establish the Objective:

What do you want to discover by conducting the experiment?For example:

•Are you trying to establish the relationship between theinput factors (Xs) and the output (response-Y)?

•Are you trying to d istinguish the vital few Xs from thetrivial many (possible factors)?

•Are you interested in knowing if several input factors acttogether to influence the output (Y)?

•Are you trying to determine the optimal settings of theinput factors?

©2009 ASQ 36

Usually stated in terms of the effects of inputson outputs

Typical experimental objectives are to determine the...

• Effects of material variation on product reliability

• Sources of variation in a critical process

• Effects of less-expensive materials on product performance

• Impact of operator variation on the product

• Cause-effect relationships between process inputs andproduct characteristics

• Equation which models your process

• Alternate method that produces the best output


8/22/2019 07 CM0471 Module 7


©2009 ASQ 37

A factor is one of the controlled or uncontrolled

inputs to a process whose influence upon one ormore responses is being studied in the experiment.

• Quantitative factors are variables data(temperature in degrees, time in seconds)

• Qualitative factors are attribute data(different machines, different operators,clean or not clean, algorithm A or B)


©2009 ASQ 38

The levels of an input factor are the values of the input factor(X) being examined in the experiment (not to be confused with the output (Y).

• Quantitative factors (continuous data):If an experiment is to be conducted using two d ifferent

speeds, then the factor speed has two levels.

• Qualitative factors (attributes data):If an experiment is to be conducted using location(Seattle and Reno), then the factor location has twolevels.


©2009 ASQ 39

A difference

10

20

30

40

50

60

70

80

0 100 200 300 400 500 600

B difference

B - l o w

A - l o w

A - h i g h

B - h i g h

R e s p o n s e V a r i a b l e ( Y )

Factor Settings

Lo(-)

Hi(+) Factor Settings

Experimental Effect

True EffectY

Lo(-)

Hi(+)

Factor Settings

Experimental Effect

True EffectY

Setting Factor Levels


8/22/2019 07 CM0471 Module 7


©2009 ASQ 40

Higher Levels of Experimentation Increase Knowledge of the Process

Selecting the Type of Experiment Design

• Response Surface Methods

• Full Factorials with Replication

• Full Factorials with Repetition

• Full Factorials without Replicationor Repetition

• Screening or Fractional Designs

• OFAT (One Factor At a Time)

Optimize

Characterize

Screen


©2009 ASQ 41

• Noise factors

• Confounding

• Blocking

• Randomization

• Cost

• Time etc…

• Different Operators

• Different Suppliers

• Different Shifts

• Machines

• Speed

• Form type

• Training

• Tool version

ProcessControllableInputs(KPIV)

Key ProcessOutputs(KPOV)

Noise Inputs(Continuous)

Noise Inputs(Discrete)

Different design considerations:


©2009 ASQ 42

Progress Check

Scenario:

• Your team is entering a bowling competition

• Your team wants to set up a DOE to win thecompetition

Task: as a group plan an experiment:

• Discuss the Experimental Objective

• Factors and levels• Appropriate design• Different design considerations – noise,

confounding, blocking, randomization,

resolution, cost, and time

8/22/2019 07 CM0471 Module 7


©2009 ASQ 44

Design and conduct completely randomized,randomized block, and Latin square designsand evaluate their results. (Evaluate)

See CSSBB HB, pages311 to 319

VII.A.4 One-factor experiments


©2009 ASQ 45

One-Factor Experiments

• As the name implies, one-factor experimentsinvolve only one factor or input variable.

• In a one-factor experiment, the project team selects

a starting point or baseline set of levels for the factor.

• This is a single-factor analysis of varianceexperiment (one-way ANOVA) – covered in Analyze

module

See CSSBB HB, example 25.12,pages 256 to 258,

©2009 ASQ 46

Randomized Complete Block Design (RCBD):

• A completely randomized block design is usedwhen we want to make experimental error as smallas possible.

• We want to remove the variability between a factorthat is not in question from the factor being tested.


See CSSBB HB, example 28.3, pages299 to 305,

8/22/2019 07 CM0471 Module 7


©2009 ASQ 47

For example: In testing a tip for hardness on a hardness testingmachine, a metal coupon is used that provides a measurement ofdepth made by pressing the tip into the coupon.

• If we decide to obtain four observations for each tip, a completelyrandomized design would consist of randomly assigning each ofthe 4 x 4 = 16 runs to an experimental unit – that is, a metal coupon.

• Thus. 16 different metal coupons would be required – one for eachrun in the design.

Because we are only interested in the

hardness of the tip, we must removethe variability that may be inherent inthe coupon.To do this, we would test each tip onetime on the same coupon, thusreducing possible experimental error.


©2009 ASQ 48

Latin Square:

• A Latin square design is often used to reduce theimpact of two blocking factors by balancing out

their contributions.

• A basic assumption is that these block factors donot interact with the factor of interest or with each

other.


See CSSBB HB, pages 311 to 319

©2009 ASQ 49

Design, analyze, and interpret these types ofexperiments, and describe how confounding affectstheir use. (Evaluate)

VII.A.5 Two-level fractional factorial experiments


8/22/2019 07 CM0471 Module 7


©2009 ASQ 50

Two-Level Fractional FactorialExperimentsFractional factorial designs:

• If we have a six-factor, two-level experiment, we would need64 runs required for a Full Factorial Experiment, which maynot be desirable.

• Thus, when there is a need to investigate only the main

effects and low-order interactions, we can study the problemwith far fewer than 2k runs.

• We accomplish this by performing a “fraction” of theruns required by the full factorial experiment.

See CSSBB HB, pages 319 to 325

©2009 ASQ 51

Two-Level Fractional FactorialExperiments

©2009 ASQ 52

Two-Level Fractional FactorialExperiments

8/22/2019 07 CM0471 Module 7


©2009 ASQ 53

Design, conduct, and analyze full factorialexperiments. (Evaluate)

VII.A.6 Full factorial experiments


©2009 ASQ 54

Full Factorial Experiments

• A full factorial design tests all combinations of thefactor levels. The number of runs will be 2k wherek is the number of factors.

• Full factorial designs are typically used in thecharacterization phase of experimentation.

©2009 ASQ 55

Progress CheckMatch each Design to the appropriate Statement

A four-factor, two-levelexperiment has 16 runs

Latin Square

We want to remove thevariability between a factorthat is not in question fromthe factor being tested

Fractional Factorial Design

Cannot be used whenestimation of interactionis required

Randomized CompleteBlock Design

Confounding takes placeFull Factorial Design

StatementDesign

8/22/2019 07 CM0471 Module 7


©2009 ASQ 57

Lesson 2 – Waste Elimination

Select and apply tools and techniques for eliminatingor preventing waste, including pull systems, kanban,

5S, standard work, poka-yoke, etc. (Analyze)

VII.B Waste elimination

©2009 ASQ 58

Waste Elimination

Eight categories of Muda

1. Overproduction above demand

2. Waiting for processing, use, work

3. Transport of products/materials

4. Over-processing

5. Inventory

6. Unnecessary motion7. Defective parts/products

8. Underutilized people

Waste elimination is a goal of Lean and Six Sigma.

©2009 ASQ 59

Kanban:

• Another tool used to create a lean enterprise isa kanban system. Kanban is a system used to

minimize WIP (work in progress).

• This lean tool creates a pull system, which places a

cap on WIP, while maintaining focus on minimizing

the process lead-time.

• A Kanban can be a mechanical or electronic

mechanism to establish a “sign” for the release of

material and a “visible” record.

Waste Elimination

8/22/2019 07 CM0471 Module 7


©2009 ASQ 60

• Kanban is a Japanese term: kan meaning “card,”

ban meaning “signal.”

• The kanban system works by signaling the need to

replenish stock or materials, or to produce more ofan item.

• Kanban can be done using cards as the signaling

component.

• In a simple kanban system, an empty box, container,or pallet can signal the need for more supplies.

• The supplier or warehouse should only deliver

components to the production line when signaled.

Waste Elimination

©2009 ASQ 61

5S:

• Japanese originally, 5S stands for five “s” words.

• The 5S method assists in the organization of theworkplace and standardization of work procedures.

• Sorting (Seiri) – Keep only what is necessary in

the work area.

• Storage/Set in Order (Seiton) – Organize the way

necessary items are kept, making it easier to findand utilize.

Waste Elimination

©2009 ASQ 62

5S (continued)

• Shining (Seiso) – Cleanliness of the work environmentand the equipment to facilitate a quality process andproduct.

• Standardizing (Seiketsu) – Tasks, procedures,schedules, and the persons responsible for helpingkeep the workplace in a clean and organized mannerare parts of the control plan for the business unit ordepartment.

• Sustaining (Shitsuke) – Indoctrinate the practice of 5Sinto your organizatio’'s culture until it becomes part ofyour standard operating procedures.

• Note: Several companies have revised the Japaneseterm, 5S, to 6S. The sixth “S” is for safety.

Waste Elimination

8/22/2019 07 CM0471 Module 7


©2009 ASQ 63

Standard Work:

• Identification and agreement on the optimal way to performeach task/step in a process becomes the standard operatingprocedure or standard work procedure.

• Standard work contributes to process control by minimizing thevariation in the product flowing through the process. There arethree basic elements involved:

– Takt time – matches the time to deliver a service, producea part or finished product to the pace of sales, and is thebasis for allocating work among workers.

– Standard in-process inventory – the minimum number ofitems or parts, including units in machines, required tokeep a cell or process moving.

– Sequence – the order in which associates perform tasks atvarious processes.

Waste Elimination

©2009 ASQ 6464

Takt Time

TaktTime Volume (Daily demand requirement)

Time (Available time per working day)=

Sets pace of work tomatch customer demand

©2009 ASQ 65

Poka-yoke:

• Poka-yoke is a Japanese term that means “to avoid

inadvertent errors.”

• “Poka-yoke” is often referred to as “mistake-proofing.”

• A poka-yoke device is one that prevents incorrectparts from being made or assembled, or easilyidentifies a flaw or error, and helps to eliminatevariations in process.

• Example: A financial institution’s loan bookingsystem requires all data entry fields on a screen tobe populated before allowing the associate to move

to the next screen, preventing an incomplete accountset-up.

Waste Elimination

8/22/2019 07 CM0471 Module 7


©2009 ASQ 66

Progress Check

1. In a pull system, who is doing the pulling?

a. Customerb. Cycle time

c. Takt time

2. Identify the type of Non-Value-Added activitywhich would directly ensue as a result of poor

quality:

a. Reworking a product

b. Unnecessary motion

c. Transportation waste

©2009 ASQ 67

Lesson 3 – Cycle-Time Reduction

VII.C Cycle-time reduction

Use various tools and techniques for reducing cycletime, including continuous flow, single-minute

exchange of die (SMED), etc. (Analyze)

©2009 ASQ 6868

Cycle-Time Reduction

“One of the most noteworthy accomplishments in keeping theprice of products low is the gradual shortening of the cycle

time. The longer an article is in the process and the more it ismoved about, the greater is its ultimate cost.”

Henry Ford, 1926

• Time that elapses from beginning to end of process

• Ultimate objective or goal of Lean processes is to reduce cycletime by eliminating waste

Work Errors, waiting, transportation, movement etc…..

Total Cycle Time

8/22/2019 07 CM0471 Module 7


©2009 ASQ 69

• Cycle time is the amount of time needed to

complete a single task or activity for the productor service.

• Cycle times may vary by task; therefore, it isbeneficial to show a range and average on thevalue stream map.

• If cycle-time variation can be reduced, the processbecomes more predictable.


©2009 ASQ 70

• Often, cycle-time can be reduced by breaking down asingle task and analyzing the amount of time that it takesto complete each sub-activity of that task.

• Analyze task processing time

– Compare task processing times to Takt time – look

for steps with work time closest to or longer thanTakt time

– Analyze resources required to meet time constraints

• After this breakdown, it is easier to tell which sub-activitiesmay be contributing to a slower cycle time.

• Ultimately, non-value-added activities can be eliminated,and value-added activities can be performed more quickly

and efficiently.


©2009 ASQ 71

Progress Check

a. To please the customerb. To reduce internal wastec. To create a bottleneck

d. To increase capacity

e. To remain competitive

The reduction of cycle time is undertaken for which

of the following principle reasons? Identify all thatapply.

8/22/2019 07 CM0471 Module 7


©2009 ASQ 72

Lesson 4 – Kaizen and Kaizen Blitz

Define and distinguish between these two methodsand apply them in various situations. (Apply)

VII.D Kaizen and kaizen blitz

©2009 ASQ 73

Kaizen and Kaizen Blitz

“Means and methods to approach the target throughcycles of continuous improvement” (Kaizen)

Masaaki Imai: “Kaizen means gradual, unending

improvement, doing ‘little things’ better; setting andachieving ever-higher standards.”

©2009 ASQ 74

• Kaizen is a Japanese term that is translated tomean “continuous improvement.”

• Many companies have successfully used workshops

called kaizen “events” or “blitzes” to drive dramaticimprovements in cycle times, inventory levels,changeover times, and overall quality.

• Successful kaizen workshops require three keycomponents:

– Selecting the right project and boundaries

– Empowering the proper team

– Planning for follow-up

Kaizen and Kaizen Blitz

8/22/2019 07 CM0471 Module 7


©2009 ASQ 75

Progress Check

The key elements of kaizen are…… Check all that apply.a. Qualityb. Involvement of all employeesc. Poor morale

d. Willingness to changee. Indiscipline

f. Communication

©2009 ASQ 76

Lesson 5 – Theory of Constraints(TOC)

VII.E Theory of constraints (TOC)

Define and describe this concept and its uses.(Understand)

©2009 ASQ 77

Theory of Constraints (TOC)

• As opposed to variation reduction (Six Sigma) orwaste removal (Lean), theory of constraints (TOC)focuses on system improvement by first payingattention to the “weakest link” of the system.

• A constraint is anything that limits an organizationfrom moving toward its goal.

• A constraint can be physical and internal (such asa machine, facility, or policy) or non-physical andexternal (such as market conditions or demand fora product).

8/22/2019 07 CM0471 Module 7


©2009 ASQ 78

• The theory of constraints was created by Dr. Eliayhu Goldratt,

and is based on the assumption that every system has atleast one constraint limiting it from getting more of what itstrives for.

• If this were not true, then the system would produce infiniteoutput.

• The theory of constraints is both descriptive and prescriptive innature; it not only describes why system constraints happen,but also offers guidance on what to do about them.

• When applying TOC, a five-step progressive approach is usedto ensure a focus is placed on system-level improvements,

rather than efforts that would only sub-optimize systemcomponents.


©2009 ASQ 79

The Five Focusing Steps of the Theory of Constraints show howto achieve ongoing improvement by addressing theseconstraints in a continuous fashion:

1. IDENTIFY – the system’s constraint

2. EXPLOIT – use kaizen or other methods to improve the rateof the constraining process

3. SUBORDINATE – the rates of everything else to match thatof the constraining process

4. ELEVATE – the system’s constraint. Could mean training,additional equipment, or new technology

5. REPEAT – the systems rate can be further improved byrepeating these steps with a new constraint


©2009 ASQ 80

Drum-Buffer-Rope:

• It is named after the 3 essential elements of the solution

• The drum or constraint or weakest link,

• The buffer or material release duration,

• The rope or release timing.

• The aim of the solution is to protect the weakest link in the system,and therefore the system as a whole, against process dependency

and variation and thus maximize the systems’ overalleffectiveness.

• The outcome is a robust and dependable process that will allow usto produce more, with less inventory, less rework/defects, andbetter on-time delivery.


8/22/2019 07 CM0471 Module 7


©2009 ASQ 81

TOC defines three operational measures to determine whether an

organization is moving toward its goal: – Increased throughput (selling price - cost of raw materials)

– Decreased inventory

– Decreased operating expenses

The following four measurements are used to identify results for theoverall organization:

– Net profit (NP) = throughput - operating expense (T-OE)

– Return on investment = net profit / inventory (NP/I)

– Productivity = throughput / operating expense (T/OE)

– Turnover = throughput / inventory (T/I)


©2009 ASQ 82

Progress Check

The goals of TOC are…… Check all that apply.a. Increased throughputb. Reduced inventoryc. Increased operational expenses

d. Reduced demand

©2009 ASQ 83

Lesson 6 – Implementation

VII. F Implementation

Develop plans for implementing the improvedprocess (i.e., conduct pilot tests, simulations, etc.),

and evaluate results to select the optimum solution.(Evaluate)

8/22/2019 07 CM0471 Module 7


©2009 ASQ 84

Implementation

Pilot run:

• It is a short run terminated by elapsed time or quantity produced

• Helps the team identify what went well and what was overlooked

Simulation:

• Usually performed with computer software

• Simulation allows one to implement solutions off-line anddetermine conclusions on how the process improvementwill operate when brought on-line

Model:

• Provides a 3-D version of what is proposed, and is an effectivevisual communication

Prototype:

• Representation of the actual product in terms of form, fit, and function

©2009 ASQ 85

Implementation Strategy Considerations:

• Infrastructure

– Includes everything to ensure successfulcompletion

• Communication Plan

– Who, what, when, where, how, and to whom it iscommunicated

• Resources – Time, people, money, energy, etc.

• Management Commitment

– This is very important. Without it, a project will notbe successful.

Implementation

©2009 ASQ 86

Progress Check

When creating a communication plan, consideration

is given to the following …

Check all that apply

a. Goals and desired results of the programb. Overall communications objectives, such as

reinforcing customer service, loyalty, etc.

c. Is sent only to the President of the organizationd. Audience you want to influencee. Methods in your plan that you can use to

measure and evaluate the program’s resultsf. Media for communications

8/22/2019 07 CM0471 Module 7


©2009 ASQ 87

Lesson 7 – Risk Analysis andMitigation

VII.G Risk analysis and mitigationUse tools such as feasibility studies, SWOT analysis(strengths, weaknesses, opportunities, and threats),PEST analysis (political, environmental, social, and

technological), and consequential metrics to analyze

and mitigate risk. (Apply)

©2009 ASQ 88

Risk Analysis and Mitigation

Feasibility studies:

• A feasibility study is an analysis of the viability ofan idea. The feasibility study focuses on helping

answer the essential question of “should weproceed with the proposed project idea?”

• All activities of the study are directed toward helping

answer this question.

• The feasibility study will be a major informationsource in making a go/no-go decision.

• This indicates the importance of a properlydeveloped feasibility study.

©2009 ASQ 89

Risk analysis:

• Is a technique to identify and assess factors thatmay jeopardize the success of a project from

achieving a goal.

• The outcome of the risk analysis would be the

creation of a risk register to identify and quantify risk

elements to the project and their potential impact.

• This technique also helps to define preventive

measures to reduce the risk.


8/22/2019 07 CM0471 Module 7


©2009 ASQ 90

The Strengths, Weaknesses, Opportunities, and

Threats (SWOT) analysis provides a framework toidentify elements that help or hinder an organization.


©2009 ASQ 91

• PEST analysis stands for “Political, Economic,

Social, and Technological analysis”

• Describes a framework of macro-environmentalfactors used in the strategic planning process

• Political Factors:

– Will government policy influence laws that

regulate or tax your business? – Is the government involved in trading agreements

such as EU, NAFTA, ASEAN, or others?


©2009 ASQ 92

• Economic Factors:

– Interest rates

– The level of inflation

– Employment level per capita

– Long-term prospects for the economy

– Gross Domestic Product (GDP) per capita, and so on.

• Sociocultural Factors:

– What are attitudes to foreign products and services?

– What are the roles of men and women within society?

– How long is the population living?

– Does the population have a strong/weak opinion onenvironmental issues?


8/22/2019 07 CM0471 Module 7


©2009 ASQ 93

• Technological Factors:

– Does technology allow for products and servicesto be made more cheaply and to a better

standard of quality?

– How is distribution changed by newtechnologies, e.g., books via the Internet,

flight tickets, auctions, etc?

– Does technology offer companies a new wayto communicate with consumers?


©2009 ASQ 94


TECHNOLOGICALSOCIO-CULTURAL

•Impact of Emerging Technologies

•Impact of Internet and Reduced

Communication Costs

•R&D Activity

•Impact of Technology Transfer

•Likely Technological Change

•Population Growth/Age Profile

•Heath, Education, Social Mobility

•Employment Patterns, Attitudes to Work

•Press, Public Opinion, Attitudes and

Taboos

•Lifestyle Choices

•Likely Socio-cultural Change

•Business Cycle Stage

•Growth, Inflation and Interest Rates

•Unemployment, Labor Supply, Labor

Costs

•Disposable Income Distribution

•Globalization

•Likely Economic Change

•Government Type

•Government Stability

•Freedom of Press, Rule of Law,

Bureaucracy, Corruption

•Regulation/Deregulation Trends

•Social/Employment Legislation

•Likely Political Change

ECONOMICPOLITICAL

PEST ANALYSIS FRAMEWORK

PEST Analysis Framework

©2009 ASQ 95

Unintended consequences:

• Understanding the mindset of business is crucial to the successof any quality project.

• A system is defined as “a group of interacting, interrelated, orinterdependent elements forming a complex whole.”

• " The ASQ Glossary defines system as “a group of inter-dependent processes and people that together perform acommon mission.”

• This latter definition highlights an important aspect of systems,namely that a system operates in unity toward a unified purpose.

• Without a true understanding of a system’s purpose, elements,and interdependencies, it is difficult to know what improvementswould truly benefit the system as a whole, rather than benefiting

only one of its elements at the possible expense of others.


8/22/2019 07 CM0471 Module 7


©2009 ASQ 96

Progress Check

What are some of the reasons to do a FeasibilityStudy? Select all that apply.

a. Identifies new opportunities through theinvestigative process.

b. It is a waste of time.

c. Enhances the probability of success by addressingand mitigating factors early on that could affect

the project.

d. Provides quality information for decision-making.

e. Provides documentation that the business

venture was thoroughly investigated.

©2009 ASQ 97

Progress Check

• Your organization is planning to introduce a newlawnmower in Florida. Your boss has instructedyou as a part of market planning to conduct aPEST analysis.

• As a group, create a PEST analysis table and

discuss.

©2009 ASQ 98

Module Status

1. Design of experiments (DOE)2. Waste elimination3. Cycle-time reduction4. Kaizen and kaizen blitz

5. Theory of constraints (TOC)6. Implementation7. Risk analysis and mitigation

8/22/2019 07 CM0471 Module 7


©2009 ASQ 99

Module 7

Exercise Solutions

©2009 ASQ 100

Answer

Process inputs an investigatormanipulates to cause a changein the output

Factors

A specific combination of factor

levels

Treatment

The output of a processResponse

Unexplained variation in the

processError

DefinitionTerm

Match each DOE term to the appropriate definition.

©2009 ASQ 101

Answer

Which of the descriptions in the table below is anexample of DOE terminology?

8/22/2019 07 CM0471 Module 7


©2009 ASQ 102

Answer Which of the descriptions in the table below are examplesof main effects or interactions? Check the appropriate box

at the right.Description/Name Example Non-Example

A two-factor DOE with two interactions. X

A three-factor DOE with A x B, A x C, B x C, and

A x B x C interactions.

X

A four-factor DOE with 2 x 4 = 8 main effects plots X

A five-factor DOE with a highest order interaction ofA x B x C x D

X

A DOE with factors A, B, and C and four main effectsplots

X

A DOE with factors A and B and only an A x Binteraction

X

Completing an analysis of variance to plot four-way

interactions

X

©2009 ASQ 103

Answers

Answers:

1. B2. C3. C

©2009 ASQ 104

Answer Match each Design to the appropriate Statement

Cannot be used when estimationof interaction is required

Latin square

Confounding takes placeFractional Factorial Design

We want to remove the variabilitybetween a factor that is not inquestion from the factor beingtested.

Randomized CompleteBlock Design

A four-factor, two-level experimenthas 16 runs

Full factorial Design

StatementDesign

8/22/2019 07 CM0471 Module 7


©2009 ASQ 105

Answer

1. In a pull system, who is doing the pulling?

a. Customer

b. Cycle time

c. Takt t ime

2. Identify the type of Non-Value-Added activity which woulddirectly ensue as a result of poor quality:

a. Reworking a product

b. Unnecessary motion

c. Transportation waste

Answers:1. a2. a

©2009 ASQ 106

Answer

a. To please the customerb. To reduce internal waste

c. To create a bottleneckd. To increase capacity

e. To remain competitive

The reduction of cycle time is undertaken for which

of the following principle reasons? Identify all thatapply.

Answer:

a, b, d, and e

©2009 ASQ 107

Answer

The key elements of kaizen are…… Check all that apply.a. Qualityb. Involvement of all employeesc. Poor morale

d. Willingness to changee. Indisciplinef. Communication

Answer:a, b, d, and f

8/22/2019 07 CM0471 Module 7


©2009 ASQ 108

Answer

The goals of TOC are…… Check all that apply.a. Increased throughputb. Reduced inventoryc. Increased operational expenses

d. Reduced demand

Answer:a and b

©2009 ASQ 109

Answer

When creating a communication plan, considerationis given to the following …Check all that apply

a. Goals and desired results of the programb. Overall communications objectives, such as

reinforcing customer service, loyalty, etc.c. Is sent only to the President of the organizationd. Audience you want to influence

e. Methods in your plan that you can use tomeasure and evaluate the program’s results

f. Media for communications

Answer:a, b, d, e and f

©2009 ASQ 110

Answer

What are some of the reasons to do a FeasibilityStudy? Select all that apply.

a. Identifies new opportunities through the investigativeprocess.

b. It is a waste of time.

c. Enhances the probability of success by addressing andmitigating factors early on that could affectthe project.

d. Provides quality information for decision-making.

e. Provides documentation that the business

venture was thoroughly investigated.

Answer:a, c, d, and e

07 cm0471 module 7

Documents