Six Sigma

1

ISI, SQC UNIT, BANGALOREISI, SQC UNIT, BANGALORE SSSS--DEFINE PHASEDEFINE PHASE

THE MANY FACETS OF SIX SIGMATHE MANY FACETS OF SIX SIGMA

�Metrics

� Benchmark

� Vision

� Philosophy

�Method

� Tool

� Symbol

� Goal

� Value

�Metrics

� Benchmark

� Vision

� Philosophy

�Method

� Tool

� Symbol

� Goal

� Value

S

I

X

S

I

G

M

A

2


WHAT IS THIS SIX SIGMA ?WHAT IS THIS SIX SIGMA ?

• A Philosophy

• A Statistical Measurement

• A Metric

• A Business Strategy

• make fewer mistakes in all that we

do.

• helps gage adequacy of product,

process and services

• A measuring system

• good quality reduces cost

3


SIGMA RATING AND PPM ARE CORRELATEDSIGMA RATING AND PPM ARE CORRELATED

Sigma Rating PPM

2 308,537

3 66,807

4 6,210

5 233

6 3.4

Process Defects per

Capability Million Opportunities

Sigma Rating PPM

2 308,537

3 66,807

4 6,210

5 233

6 3.4

Process Defects per

Capability Million Opportunities

SIGMA RATING A PERFORMANCE MEASURE

4


• TO REDUCE VARIATION

• TO REDUCE DEFECTS

• TO IMPROVE YIELD

• TO ENHANCE CUSTOMER SATISFACTION

• TO IMPROVE THE BOTTOMLINE

THE GOALS OF SIX SIGMATHE GOALS OF SIX SIGMA

5


To get results should we focus on Y or X?

• Y

• Dependent variable

• Output of the process

• Effect

• Symptom

• It is monitored

• X1, X2, …….., Xn

• Independent variable

• Input to the process

• Cause

• Problem

• It is controlled

Y = f(X)

THE FOCUS OF SIX SIGMATHE FOCUS OF SIX SIGMA

6


APPROACH TO DEFINEAPPROACH TO DEFINE

Develop Business

Case and

Team Charter

Develop Business

Case and

Team Charter

Map the

current

process

Map the

current

process

Listen to the

Voice of the

Customer

Listen to the

Voice of the

Customer

7


PREVIEW OF DEFINE STEPPREVIEW OF DEFINE STEP

• The subsequent discussions will take you through

– Selecting a project and drafting a charter

– A basic SIPOC analysis

– Gathering/confirming the “Voice of the Customer”

(Effectiveness goals e.g. On time delivery) and/or the

“Voice of the Business” (Efficiency goals e.g. reduction

of inventory).

8


GOALSGOALS– Understand the organizational context for process

improvement projects.

– Understand how to select a DMAI²C project and be able

to apply the criteria to your own project.

– Know the basic elements of the project charter and be able

to develop a charter for your own project.

– Understand the basics of estimating the business impact of

a project and be able to relate them to your own project.

– Be able to identify key players and stakeholders and

incorporate them into the communication plan for your

own project.

9


ORGANIZATIONAL CONTEXT FOR PROJECTSORGANIZATIONAL CONTEXT FOR PROJECTS

Champions/

Sponsors

Black Belt

Green Belt

Team Members

Master

Black BeltCoordinators

10


DMAI²C PROJECT SELECTION CRITERIADMAI²C PROJECT SELECTION CRITERIA

• Projects will likely have greater success if…

– The problem is related to a key business issue.

– The problem is linked to a clearly defined process

(you can identify the starting and ending points).

– You can identify the internal or external

customers who use or receive the output from this

process.

• You know or can find out how the customers

use the output.

• You know or can find out what is important to

the customer about the output.

11


– You can clearly identify what a defect is and count its

occurrence

– You can demonstrate how improvements could

enhance financial performance

– There is appropriate organizational support

• The champion cares about this work and wants to

make it happen

• The champion has the authority to commit time

and resources

• A process owner has been identified

DMAI²C PROJECT SELECTION CRITERIADMAI²C PROJECT SELECTION CRITERIA

12


COMMON PROBLEMS AND COMMON PROBLEMS AND

RECOMMENDATIONSRECOMMENDATIONS

• Many projects encounter one or more of the following problems:

– Not relevant to customers or to business needs

– Project scope too large; won’t have authority to commit time/resources or to make changes in the process; can’t easily identify starting and ending points

– Relevant data hard to collect or too much data

– Can’t define what a defect is

– Process doesn’t cycle often enough

– Problem written as solution

– Recent changes in the process

13


ELEMENTS OF A PROJECT CHARTERELEMENTS OF A PROJECT CHARTER

– Project Description

– Scope

– Goals and Measures (indicators)

– Expected Business Results

– Team Members

– Support Required

– Expected Customer Benefits

– Schedule

14


QUALITY AND PROCESSESQUALITY AND PROCESSES

– Quality is judged by customers based on the output of a process.

– Focusing on the work of the individual worker will not lead to greatly improved quality.

– To improve quality, the process must be improved.

– Simply focusing on or defining a process is NOT improvement—we have to make changes and use data to show the change really is an improvement.

15


WORK AS A PROCESSWORK AS A PROCESS

– Applies to all kinds of work, whether repetitive in

nature or “one-of-a-kind.”

– Having a high-level view of a process helps to:

• define project boundaries (starting and ending points);

• describe where to collect data.

S U P P L I ER S

C U S T O M E R S

OutputsInputs Process

16


SIPOC EXAMPLESIPOC EXAMPLESuppliers Inputs Process Outputs Customers

Copier

Toner

Electricity

OriginalYourself

Power Company

Office Supply

Company

CopiesManufacturer

Making a

photocopy

You

File

Others

Put original

on glass

Put original

on glass Close LidClose Lid Adjust

Settings

Adjust

SettingsPress

START

Press

START

Remove

originals

and copies

Remove

originals

and copies

Process Steps

Paper

17


TRANSLATING VOC/VOB INTO CTQsTRANSLATING VOC/VOB INTO CTQs

• (CTQ = Critical to Quality)

need

VOC/VOB CTQ Tree

I want

CTQ

CTQ

CTQ

CTQ

CTQ

CTQ

CTQ

CTQ

18


EXAMPLE: CTQ TREEEXAMPLE: CTQ TREE

Need Drivers CTQs

General Specific

Hard to measure Easy to measure

Ease of Operation

and Maintenance

Operation

Low qualification of operator

Easy to setup

(training / documentation)

Digital Control

MaintenanceMTBF

Mistake Proofing

Documentation

Minimum special tools /

equipment required

Modification

Skills and Knowledge

Mistake Proofing

19


APPROACH TO MEASUREAPPROACH TO MEASURE

Collect

baseline

data on

defects

and

possible

causes

Develop a

Sampling

Strategy

Validate

your

measurement

system

using

Gage R&R

Analyze

patterns

in data

Determine

process

capability

20


• What is Data ?

Data is a numerical expression of an activity.

Conclusions based on facts and data are necessary for

any improvement.

-K. Ishikawa

If you are not able to express a phenomenon in

numbers, you do not know about it adequately.

-Lord Kelvin

DATA GATHERINGDATA GATHERING

21


TYPES OF DATATYPES OF DATA

CONTINUOUS DISCRETE

Measurable

e.g. :Length, Temperature

Subjective Assessment

e.g. :Score in a beauty

contest

Countable

e.g. :Number of defects

Data if properly collected

• Least influenced by individual biases

• Could be subject to critical analysis

• Generally beyond language barriers and therefore universal in

expression.

22


WHICH OF THE BELOW ARE CONTINUOUS WHICH OF THE BELOW ARE CONTINUOUS

AND DISCRETE DATA?AND DISCRETE DATA?• Width of sheet

• No. of liners thinned

• Tubes rejected by Go- Nogo Gauge

• Diameter of Piston

• Height of a Man

• Sheet thickness

• Out of 100 sheets the numbers that meet the thickness

4 ± 0.9

• Time taken to process a purchase order

• No. of bugs in a program

ISI, SQC UNIT, BANGALOREISI, SQC UNIT, BANGALORE SSSS--MEASURE PHASEMEASURE PHASE

24


WHAT IS THE MEASUREOF CENTRAL TENDENCY OF A SET OF

NUMBERS?

WHAT IS THE MEASUREWHAT IS THE MEASUREOF CENTRAL TENDENCY OF A SET OF OF CENTRAL TENDENCY OF A SET OF

NUMBERS? NUMBERS?

• There are three ways in which Central

Tendency of Numbers can be measured.

• These are the 3 M’s

�MEAN

�MEDIAN

�MODE

25


MEASURES OF DISPERSIONMEASURES OF DISPERSIONMEASURES OF DISPERSION

� The extent of the spread of the values from the

mean value is called Dispersion.

� The measures of Dispersions are– Range (R)– Standard Deviation (s)– Variance (s2)– Co-efficient of Variation (CV)

�Standard deviation is the most commonly used measure of dispersion.

26


• No two things in nature are alike.

• This is also true for manufactured products.

• This dissimilarity between two products for the same characteristic is called variation.

• The variation may be or can be made to be so small so as to make the product SEEM similar.

•When we say that 2 things are similar we actually mean that it is not possible to measure the variation present within the accuracy of the existing measuring equipment.

• Variation between 2 products are compared for SIMILAR features or characteristics.

WHAT IS VARIATION ?WHAT IS VARIATION ?

27


• Variations among pieces at the same

time

• Variations across time

TYPES OF VARIATIONTYPES OF VARIATION

28


GRAPHICAL DISPLAY OF GRAPHICAL DISPLAY OF VARIABILITIESVARIABILITIES

InherentVariability

Assignable Variability


TOTAL

VAR I A B I L I T Y



Aimed Value

CASE I

CASE II CASE III

(Black noise)

29


GENERIC CAUSES OF VARIATIONGENERIC CAUSES OF VARIATION

Machines

Materials

Methods

Measurements

Mother Nature

People

P

R

O

C

E

S

S

PP

RR

OO

CC

EE

SS

SS

30


THE NORMAL DISTRIBUTIONTHE NORMAL DISTRIBUTION

95.46

68.26

99.73-3 -2 +3-1 +1 +2

31


IMPORTANT SPC RATIOS USEDIMPORTANT SPC RATIOS USED

This compares the requirement of the process output vis-a-vis the

inherent variability of the process. Higher value than 1 implies that the

process has got the capability to give the product within the set limits.

LSL x USL

- 3 + 3

s 6

LSL-USL

s 6

Tolerance

process of variation Normal

sticcharacteri ofrange allowable Maximum Cp

========

====

32


This gives us the positioning of the mean vis-a-vis the USL and the relationship between the two.

This gives us the positioning of the mean vis-a-vis the LSL and the relationship between the two.

Cpk - Process Performance Index. This is important

Cpk = Minimum of (Cpu and Cpl) ; for bilateral tolerances

= Cpu ;for unilateral tolerance on upper side i.e..

= Cpl ;for unilateral tolerance on lower side i.e..

X+Y-O

X+O-Y

s3

X-USL Cpu ====

s3

LSLX Cpl

−−−−====

33


• Every possibility of making an error is called an opportunity

• The total opportunities available for an error to take place areNos.Chkd. x Opp

• If there are more than 1 opp. The sigma can be calculated by finding DPMO.

• Knowing DPMO we refer to the Normal Dist. Table to get the Sigma value

• One could inflate the opp. and hence get an enhanced Sigma But the opp.are limited to what exactly is checked for.

E.g. a sheet is checked for thickness, length & width and can berejected for either.Hence the opp. is 3

THE CONCEPT OF OPPORTUNITYTHE CONCEPT OF OPPORTUNITY

DPMO = Defects x 106

Units chkd. x Opp.

DPMO = Defects x 106

Units chkd. x Opp.

34


•Defect : Any non-conformity in a product or service

– e.g. Late delivery or no. of tubes rejected

•Units : The nos. checked or inspected

– 100 deliveries were monitored for being late, no. of units are 100

– 1000 tubes were checked for oversize dia., no. of units are 1000

•Opportunity : Anything that you measure or check for.

– Finished refrigerator is checked for 25 defects at final inspection, the no. of opportunities is 25

A FEW TERMINOLOGIESA FEW TERMINOLOGIES

35


• This Z value is known as the SIGMA RATING

• Knowing the sigma rating we can establish the rejections that can be expected from the process

• The average, the standard deviation & the specification limits are required to get the sigma rating

• There are 2 kinds of sigma ratings.The short term Sigma & the long term Sigma (to be seen in detail later)

• One should not confuse between the sigma rating & the standard deviation

THE SIGMA RATINGTHE SIGMA RATING

36


Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO

0.01 931888 0.26 892512 0.51 838913 0.76 770350 1.01 687933 1.26 594835 1.51 496011 1.76 397432

0.02 930563 0.27 890651 0.52 836457 0.77 767305 1.02 684386 1.27 590954 1.52 492022 1.77 393580

0.03 929219 0.28 888767 0.53 833977 0.78 764238 1.03 680822 1.28 587064 1.53 488033 1.78 389739

0.04 927855 0.29 886860 0.54 831472 0.79 761148 1.04 677242 1.29 583166 1.54 484047 1.79 385908

0.05 926471 0.30 884930 0.55 828944 0.80 758036 1.05 673645 1.30 579260 1.55 480061 1.80 382089

0.06 925066 0.31 882977 0.56 826391 0.81 754903 1.06 670031 1.31 575345 1.56 476078 1.81 378281

0.07 923641 0.32 881000 0.57 823814 0.82 751748 1.07 666402 1.32 571424 1.57 472097 1.82 374484

0.08 922196 0.33 878999 0.58 821214 0.83 748571 1.08 662757 1.33 567495 1.58 468119 1.83 370700

0.09 920730 0.34 876976 0.59 818589 0.84 745373 1.09 659097 1.34 563559 1.59 464144 1.84 366928

0.10 919243 0.35 874928 0.60 815940 0.85 742154 1.10 655422 1.35 559618 1.60 460172 1.85 363169

0.11 917736 0.36 872857 0.61 813267 0.86 738914 1.11 651732 1.36 555670 1.61 456205 1.86 359424

0.12 916207 0.37 870762 0.62 810570 0.87 735653 1.12 648027 1.37 551717 1.62 452242 1.87 355691

0.13 914656 0.38 868643 0.63 807850 0.88 732371 1.13 644309 1.38 547758 1.63 448283 1.88 351973

0.14 913085 0.39 866500 0.64 805106 0.89 729069 1.14 640576 1.39 543795 1.64 444330 1.89 348268

0.15 911492 0.40 864334 0.65 802338 0.90 725747 1.15 636831 1.40 539828 1.65 440382 1.90 344578

0.16 909877 0.41 862143 0.66 799546 0.91 722405 1.16 633072 1.41 535856 1.66 436441 1.91 340903

0.17 908241 0.42 859929 0.67 796731 0.92 719043 1.17 629300 1.42 531881 1.67 432505 1.92 337243

0.18 906582 0.43 857690 0.68 793892 0.93 715661 1.18 625516 1.43 527903 1.68 428576 1.93 333598

0.19 904902 0.44 855428 0.69 791030 0.94 712260 1.19 621719 1.44 523922 1.69 424655 1.94 329969

0.20 903199 0.45 853141 0.70 788145 0.95 708840 1.20 617911 1.45 519939 1.70 420740 1.95 326355

0.21 901475 0.46 850830 0.71 785236 0.96 705402 1.21 614092 1.46 515953 1.71 416834 1.96 322758

0.22 899727 0.47 848495 0.72 782305 0.97 701944 1.22 610261 1.47 511967 1.72 412936 1.97 319178

0.23 897958 0.48 846136 0.73 779350 0.98 698468 1.23 606420 1.48 507978 1.73 409046 1.98 315614

0.24 896165 0.49 843752 0.74 776373 0.99 694974 1.24 602568 1.49 503989 1.74 405165 1.99 312067

0.25 894350 0.50 841345 0.75 773373 1.00 691462 1.25 598706 1.50 500000 1.75 401294 2.00 308538

Sigma and DPMO conversion table

37


Sigma and DPMO conversion tableSigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO

2.01 305026 2.26 223627 2.51 156248 2.76 103835 3.01 65522 3.26 39204 3.51 22216 3.76 11911

2.02 301532 2.27 220650 2.52 153864 2.77 102042 3.02 64256 3.27 38364 3.52 21692 3.77 11604

2.03 298056 2.28 217695 2.53 151505 2.78 100273 3.03 63008 3.28 37538 3.53 21178 3.78 11304

2.04 294598 2.29 214764 2.54 149170 2.79 98525 3.04 61780 3.29 36727 3.54 20675 3.79 11011

2.05 291160 2.30 211855 2.55 146859 2.80 96801 3.05 60571 3.30 35930 3.55 20182 3.80 10724

2.06 287740 2.31 208970 2.56 144572 2.81 95098 3.06 59380 3.31 35148 3.56 19699 3.81 10444

2.07 284339 2.32 206108 2.57 142310 2.82 93418 3.07 58208 3.32 34379 3.57 19226 3.82 10170

2.08 280957 2.33 203269 2.58 140071 2.83 91759 3.08 57053 3.33 33625 3.58 18763 3.83 9903

2.09 277595 2.34 200454 2.59 137857 2.84 90123 3.09 55917 3.34 32884 3.59 18309 3.84 9642

2.10 274253 2.35 197662 2.60 135666 2.85 88508 3.10 54799 3.35 32157 3.60 17864 3.85 9387

2.11 270931 2.36 194894 2.61 133500 2.86 86915 3.11 53699 3.36 31443 3.61 17429 3.86 9137

2.12 267629 2.37 192150 2.62 131357 2.87 85344 3.12 52616 3.37 30742 3.62 17003 3.87 8894

2.13 264347 2.38 189430 2.63 129238 2.88 83793 3.13 51551 3.38 30054 3.63 16586 3.88 8656

2.14 261086 2.39 186733 2.64 127143 2.89 82264 3.14 50503 3.39 29379 3.64 16177 3.89 8424

2.15 257846 2.40 184060 2.65 125072 2.90 80757 3.15 49471 3.40 28716 3.65 15778 3.90 8198

2.16 254627 2.41 181411 2.66 123024 2.91 79270 3.16 48457 3.41 28067 3.66 15386 3.91 7976

2.17 251429 2.42 178786 2.67 121001 2.92 77804 3.17 47460 3.42 27429 3.67 15003 3.92 7760

2.18 248252 2.43 176186 2.68 119000 2.93 76359 3.18 46479 3.43 26803 3.68 14629 3.93 7549

2.19 245097 2.44 173609 2.69 117023 2.94 74934 3.19 45514 3.44 26190 3.69 14262 3.94 7344

2.20 241964 2.45 171056 2.70 115070 2.95 73529 3.20 44565 3.45 25588 3.70 13903 3.95 7143

2.21 238852 2.46 168528 2.71 113140 2.96 72145 3.21 43633 3.46 24998 3.71 13553 3.96 6947

2.22 235762 2.47 166023 2.72 111233 2.97 70781 3.22 42716 3.47 24419 3.72 13209 3.97 6756

2.23 232695 2.48 163543 2.73 109349 2.98 69437 3.23 41815 3.48 23852 3.73 12874 3.98 6569

2.24 229650 2.49 161087 2.74 107488 2.99 68112 3.24 40929 3.49 23295 3.74 12545 3.99 6387

2.25 226627 2.50 158655 2.75 105650 3.00 66807 3.25 40059 3.50 22750 3.75 12224 4.00 6210

38


Sigma and DPMO conversion table

Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO Sigma DPMO

4.01 6037 4.26 2890 4.51 1306 4.76 557 5.01 224 5.26 85 5.51 30.4 5.76 10.2

4.02 5868 4.27 2803 4.52 1264 4.77 538 5.02 216 5.27 82 5.52 29.1 5.77 9.8

4.03 5703 4.28 2718 4.53 1223 4.78 519 5.03 208 5.28 78 5.53 27.9 5.78 9.4

4.04 5543 4.29 2635 4.54 1183 4.79 501 5.04 200 5.29 75 5.54 26.7 5.79 8.9

4.05 5386 4.30 2555 4.55 1144 4.80 483 5.05 193 5.30 72 5.55 25.6 5.80 8.5

4.06 5234 4.31 2477 4.56 1107 4.81 467 5.06 185 5.31 70 5.56 24.5 5.81 8.2

4.07 5085 4.32 2401 4.57 1070 4.82 450 5.07 179 5.32 67 5.57 23.5 5.82 7.8

4.08 4940 4.33 2327 4.58 1035 4.83 434 5.08 172 5.33 64 5.58 22.5 5.83 7.5

4.09 4799 4.34 2256 4.59 1001 4.84 419 5.09 165 5.34 62 5.59 21.6 5.84 7.1

4.10 4661 4.35 2186 4.60 968 4.85 404 5.10 159 5.35 59 5.60 20.7 5.85 6.8

4.11 4527 4.36 2118 4.61 936 4.86 390 5.11 153 5.36 57 5.61 19.8 5.86 6.5

4.12 4397 4.37 2052 4.62 904 4.87 376 5.12 147 5.37 54 5.62 19.0 5.87 6.2

4.13 4269 4.38 1988 4.63 874 4.88 362 5.13 142 5.38 52 5.63 18.1 5.88 5.9

4.14 4145 4.39 1926 4.64 845 4.89 350 5.14 136 5.39 50 5.64 17.4 5.89 5.7

4.15 4025 4.40 1866 4.65 816 4.90 337 5.15 131 5.40 48 5.65 16.6 5.90 5.4

4.16 3907 4.41 1807 4.66 789 4.91 325 5.16 126 5.41 46 5.66 15.9 5.91 5.2

4.17 3793 4.42 1750 4.67 762 4.92 313 5.17 121 5.42 44 5.67 15.2 5.92 4.9

4.18 3681 4.43 1695 4.68 736 4.93 302 5.18 117 5.43 42 5.68 14.6 5.93 4.7

4.19 3573 4.44 1641 4.69 711 4.94 291 5.19 112 5.44 41 5.69 14.0 5.94 4.5

4.20 3467 4.45 1589 4.70 687 4.95 280 5.20 108 5.45 39 5.70 13.4 5.95 4.3

4.21 3364 4.46 1538 4.71 664 4.96 270 5.21 104 5.46 37 5.71 12.8 5.96 4.1

4.22 3264 4.47 1489 4.72 641 4.97 260 5.22 100 5.47 36 5.72 12.2 5.97 3.9

4.23 3167 4.48 1441 4.73 619 4.98 251 5.23 96 5.48 34 5.73 11.7 5.98 3.7

4.24 3072 4.49 1395 4.74 598 4.99 242 5.24 92 5.49 33 5.74 11.2 5.99 3.6

4.25 2980 4.50 1350 4.75 577 5.00 233 5.25 88 5.50 32 5.75 10.7 6.00 3.4

39


• Brainstorm on X’s

• Find change of which X’s affect Y and in what

manner

• Ultimately find which X’s are critical to move the

Y in the desired direction

IN THE ANALYSIS PHASE YOU WILL...IN THE ANALYSIS PHASE YOU WILL...

IN MEASURE PHASE, WE DEALT WITH Y’s.

IN ANALYSIS PHASE, WE WILL DISCOVER

& DEAL WITH X’s.

IN MEASURE PHASE, WE DEALT WITH Y’s.

IN ANALYSIS PHASE, WE WILL DISCOVER

& DEAL WITH X’s.

40


To get results should we focus on Y or X?

• Y

• Dependent variable

• Output of the process

• Effect

• Symptom

• It is monitored

• X1, X2,…….., Xn

• Independent variable

• Input to the process

• Cause

• Problem

• It is controlled

Y = f(X)

THE FOCUS OF SIX SIGMATHE FOCUS OF SIX SIGMA

41


B e com e A w a r e o f th e P r o b lem

S e le c t L ik e ly C a u se s

Id e n t if y P o ten t ia l C a u se s

D e f in e a n d V e r ify R o o t C a u se s

D e s c r ib e th e P r o b lem

Id en t if y A lt e rn a t iv e S o lu t io n s

I s th e P o te n t ia l C a u se a R o o t C a u se ?

C o n g r a tu la t e y o u r T e am

Im p lem en t p e rm a n a n t a c t io n s

V e r i fy C o r r e c t io v e A c t io n s

P r e v e n t R e c u r r e n c e

Y e s

42


High Petrol

Consumption

Procedure Driver Vehicle

MaterialsMaintenanceRoad

Restrictions

No turnOne way

Circuitous

Road

Frequent

stops

Crossings

Traffic

Speed Breakers

Steep

Poor

condition

Potholes Irregular

servicing

False

economy

Negligence

Clogged

filters

Low pressureIgnorance

Faulty

pressure

Tyres

Petrol

Oil

Not changed

Low level

Incorrect viscosity

ImpuritiesIncorrect

Octane no.

Additives

SparesSpurious

Inferior

Impatience

Craze

Always

late

Riding on

clutch

Lack of

awareness

Poor

anticipation

Wrong

gears

Poor

skill

Wrong

culture

Bad

attitude

Inexperience

Body

Heavy

Shape

Technical

details

Carburetor

Spark plugs

LifeContacts

Fuel mix

Engine

High H.P

Cylinders

Cause and Effect Diagram for High Petrol ConsumptionCause and Effect Diagram for High Petrol Consumption

43


Validation of CausesValidation of Causes

⌦GEMBA (Work Place) Investigation

⌦ List each cause and verify them through workplace

observations.

Causes Specifications/ Desired

states

Observations Remarks

Bunching

of car

Maximum 2 at a

time

6 out of 10

moments

May be a potential

root cause for

Delay in servicing

44


CTQ(y)

Type

Cause(x)

Type

Question to

Answer

Type of

Analysis

Single

Continuous

(Noise)

Discrete in

Two categories

(Bearing-1 & 2)

Do bearing have any effect on

Variation in Noise?2 Variance test

(F-test)

Single

Continuous

(Noise)

Discrete in two

categories

(Bearing-1& 2)

Do bearing have any effect on

Noise level?2-sample t-test.

Data Based validation of Causes

Single

Continuous

(Cycle time)

Discrete in more

than two categories

(Machine-1,2& 3)

Do machine have any effect

on Cycle time?

Analysis of Variance/

Analysis of Means

Single

Continuous

(Cycle time)

Discrete in more

than two categories

(Machine-1,2& 3)

Do machine have any effect

on variation in Cycle time?

Bartlett’s test for

equal variances.

Single

Discrete

(defectives)

Discrete in two

categories

(Month -1& 2)

Do time have any effect on

defectives?

2-proportions test

Single

Discrete

(Error/defect)

Discrete in more

than two/more

categories

(Type of jobs)

Do type of jobs have any effect

on error level?χχχχ2 test (Chi- square

test)

45


CTQ(y)Type

Cause(x)Type

Question toAnswer

Type of

AnalysisSingle

Continuous

(Cycle time)

Single continuous

(Time taken in

Process step 1)

Do time taken in process step 1

Influence the cycle time?

Simple correlation and

Regression analysis.

Data Based validation of Causes (Contd.)

Single

Discrete

(defectives)

Single

Continuous

(Cycle time)

Multiple continuous

(Time taken in

Process steps 1,2&3)

Multiple

Regression analysis.

Single continuous

(machine speed).

Logistic

Regression

Which process steps

Influence the cycle time

maximum?Does the machine speed

influence defectives?

Single

Discrete

(defectives)

Single continuous

(machine speed).Does the machine speed

influence defectives?

Multiple Logistic

Regression

Single

Continuous

On same

Subject

(Noise level at

Two different

speed)

Discrete in two

categories

(machine speed).

Does the machine speed

influence noise level?Paired t-test

46


CTQ(y)Type

Cause(x)Type

Question toAnswer

Type of

AnalysisSingle discrete in

two categories

(Two supplier)

Do the supplier have different

cycle time?

Wilcoxon signed rank

test

Data Based validation of Causes (Contd.)

Single

Continuous

but non normal

(Cycle time)

Single

Continuous

but non normal

(Cycle time)

Single discrete in

More than two

categories

(Supplier 1,2 & 3)

Do the supplier have different

cycle time?Kruskal-wallis test

47


� Learn, how to select the solution

� How to pilot the solutions

� How to assess the risks

� How to implement the solutions

IN THE IMPROVE PHASE YOU WILL…IN THE IMPROVE PHASE YOU WILL…

48


GENERATING SOLUTION IDEASGENERATING SOLUTION IDEAS– Review what you know about the process and the verified cause.

– Brainstorm solution ideas; use creativity techniques.

– Combine ideas into solutions.

49


A PATH TO BETTER SOLUTIONSA PATH TO BETTER SOLUTIONS• In order to get better solutions, we will follow these steps:

– Generate Criteria

– Weight Criteria

– Evaluate Ideas

• The tool we will use for this is the Solution Prioritization Matrix.

Solution

A

B

C

D

0.2 1.25 0.3 1.65 0.6

Easy Quick Tech Hi Impact Customers

SUM

2.6

3.0 2.4

2.7

19.8

8.25

8.4

4.8

10

13.75

43.6

1.2 2.7 19.8 5.47.5 36.6

32.1

4.2 18.15 5.418.75 47.71.2

Criteria and Weights

50


WHY PILOT?WHY PILOT?

– Improve the solution

– Understand risks

– Validate expected results

– Smooth implementation

– Facilitate buy-in

– Identify previously unknown performance

problems

51


WHEN TO PILOTWHEN TO PILOT

Pilot when:

– You need to confirm the expected results and

practicality of the solution.

– You want to reduce the risk of failure.

– The scope of the change is large, and reversing the

change would be difficult.

– Implementing the change will be costly.

– Changes would have far-reaching, unforeseen

consequences.

52


EVALUATE THE PILOTEVALUATE THE PILOT

– Compare results against the original definition of a

defect and against CTQs.

– Recalculate process sigma and process capability.

– Analyze causal relationships and process conditions.

53


PLANNING GRIDS DEFINITIONPLANNING GRIDS DEFINITION• A planning grid helps to identify the resources and outcomes for each step in the process.

Due Date

Respon-sibility

DeliverableStepWhom to

Involve

Budget/ Cost

Other Topics

Step Number

1

2

4

3

Develop new wiring

schematic

Wiring PlanCalvin,Max,Sheryl

6/12 N/ADept.staff,Maria

Conduct ChangesMaria 6/12 N/A

Dept.staff

Revalidate SystemMaria 6/16 N/A

Update

documentationCalvin 6/8 N/A

Otherdeptsupv

Machine

Changes

Documentation

Documentation

54


PLANNING GRID FEATURESPLANNING GRID FEATURESPerson(s)responsibleidentified

Steps listedsequentially

Outcome of eachstep clearlyidentified

Timing andother factors

tracked

Due Date

Respon-sibility

DeliverableStepWhom to

Involve

Budget/ Cost

Other Topics

Step Number

1

2

4

3

Develop new wiring

schematic

Wiring PlanCalvin,Max,Sheryl

6/12 N/ADept.staff,Maria

Conduct ChangesMaria 6/12 N/A

Dept.staff

Revalidate SystemMaria 6/16 N/A

Update

documentationCalvin 6/8 N/A

Otherdeptsupv

Machine

Changes

Documentation

Documentation

55


THE PEOPLE SIDETHE PEOPLE SIDE

– Too often we spend all our time on the “technical” side of a

change—what has to happen by when, etc.

– We ignore the people side of the change—how to help those

who will have to change their ways of doing things in order

to realize the new change.

– There are three elements of the people side:

• Communication: the exchange of information both from you to

others and from others to you.

• Participation: involving people in the planning and execution of a

change so they can develop shared ownership and commitment.

• Education: providing people with what they will need to know

before they successfully implement the desired changes.

56


BENEFITS OF PEOPLEBENEFITS OF PEOPLE--SIDE PLANNINGSIDE PLANNING

– Increased understanding �� decreased confusion

– Increased commitment �� decreased resistance

– Increased capability �� decreased fear of failure

57


APPROACH TO CONTROLAPPROACH TO CONTROL

Quality Control &

Process Change

Management

Quality Control &

Process Change

Management

Standardize and

Document

Effective

Methods

Standardize and

Document

Effective

Methods

Establish

Ongoing

Project

Monitoring

Establish

Ongoing

Project

Monitoring

Evaluating

Results

Evaluating

Results

Hand Off,

Summarize Key

Learnings and

Draft Future Plans

Hand Off,

Summarize Key

Learnings and

Draft Future Plans

58


CONTROL PHASECONTROL PHASE

Till now you may have

• Described your ‘Y’ and got the direction of action.

• Isolated your critical Xs.

• Optimized the process by deciding on the level of these Xs.

IN THE CONTROL PHASE YOU DEAL

WITH HOLDING THE GAINS THROUGH

CONTROL OF Xs.

59


SOME WAYS TO MAINTAIN/CONTROL SOME WAYS TO MAINTAIN/CONTROL

YOUR NEW PROCESS CAN BE BYYOUR NEW PROCESS CAN BE BY

�Standardizing : Put it in audit checklist, quality

plan, documented procedure etc.

�Fool-Proofing : To ensure that it can be done in

no other way except how you want it to be done.

�Have Advance Warning Systems : If

something goes wrong, it tells that to you NOW

What you would as such come to know of it over

a period of time.

e.g. : Control Charts

60


PROCESSES CAN BE CONTROLLED BY PROCESSES CAN BE CONTROLLED BY

BUT NOT LIMITED TOBUT NOT LIMITED TO

• SPC and Advance Warning Systems

• Inspection & Audits

• 100% checking of output

• Fool-Proofing

• Sampling Plans

• Control / Quality Plans

• Quality Systems

61


• Different processes would be sensitive to

different parameterse.g.: Man, Machine

• Hence different processes would require

different controls which in turn would

depend on that specific parameter to which

it is sensitivee.g.: Brazing may be welder dependent. Hence

training would be an important control parameter.

Thus leading to concept of brazer qualification.

62


IN A NUT SHELL IN A NUT SHELL Your Process may be:

AFFECTED BY CAN BE MAINTAINED BY

� Initial Setup

� Equipment/

machine settings

� Fixtures

� Tooling

� Man

� Material

� Maintenance

� Environment

� FPI in between the checks

� Parameters feedback system

� First approvals, monitor the assignable causes

� Tool approval, Fool-Proofing of mating parts, FPI,

Audit checks

� Training and Fool-Proofing

� Certification before use

� PM schedules, FPI check after each maintenance

� Control environment and PM of environment

controlling equipment

63


CONTROL PLANCONTROL PLAN Page 01 of 02

PROTOTYPE PRE-LAUNCH PRODUCTION

CONTROL PLAN NUMBER : CP-M13-02

KEY CONTACT / PHONE : Head, Quality & C.S. DATE (ORI.) : 08.09.99 DATE (REV.) : 27.06.00

PART NUMBER / LATEST CHANGE LEVEL : ML CORE TEAM : BSK, HA, GVS, KSK CUSTOMER ENGG. APPROVAL / DATE (IF REQD.) : N.A.

PART NAME / DESCRIPTION : Generic SUPPLIER / PLANT APPROVAL / DATE : 01.10.99 CUSTOMER QUALITY APPROVAL / DATE (IF REQD.) : N.A.

SUPPLIER PLANT : Nanjangud SUPPLIER CODE : N.A. OTHER APPROVAL / DATE (IF REQD.) : N.A. OTHER APPROVAL / DATE (IF REQD.) : N.A.

CHARACTERISTICS METHODS

SAMPLE

PART /

PROCESS

NUMBER

PROCESS NAME /

OPERATION

DESCRIPTION

MACHINE,

DEVICE, JIG,

TOOLS FOR

MFG.

NO. PRODUCT PROCESS

SPL.

CHAR.

CLASSPRODUCT /

PROCESS

SPEC. /

TOLERANCE

EVALUATION /

MEASUREMENT

TECHNIQUE SIZE FREQ.

CONTROL

METHOD

REACTION

PLAN

1 Drill hole

position

* + 50 µ - X

+ 50 µ - Y

True measure Every spindle SPC chart Check &

adjust

2 Hole quality * As per SOP Cross Section Once / week

All layer counts

Hole

quality

chart / log

book

Check &

adjust

3 Hole diameter As per traveler Pin gauge Bottom

panel

Every

lot

Visual Check &

adjust

4 Hole count As per

program

Drill check sheet Bottom

panel

Every

stack

Visual Check &

adjust

Machine setup

M13 Drilling Drilling

machines,

Drill bits,

Aluminium

Entry & backup

5 Scope drill target pad

distance

As per traveler Scope drill m/c. All

panels

100% Log book Quarantine

& adjust

previous

process

�

64


POKAPOKA--YOKE (JAPANESE TERM)YOKE (JAPANESE TERM)

• Fail Safe

– Methods used to either make the occurrence of a

nonconformity impossible or to ensure that it

does not pass undetected.

• However…

As we get better at fool-proofing,

we get better fools.

65


STANDARDIZATIONSTANDARDIZATION

– Making sure that important elements of a process are performed

consistently in the best possible way.

– Changes are made only when data shows that a new alternative

is better.

– Documentation is key

• Making sure documentation is up to date and used encourages ongoing

use of standardized methods.

– Discussion

• What images come to mind when you think of process standardization?

66


BENEFITS OF STANDARDIZATIONBENEFITS OF STANDARDIZATION

• Standardization helps us compete more

successfully in the marketplace by providing:

– Increased reliability

– Reduced costs

– Improved employee performance

– Increased safety

– Processes that remain in control

– Continuous improvement

– Flexible practices that allow for quick

response to customer needs

67


STANDARDIZATION = “STANDARD STANDARDIZATION = “STANDARD

PRACTICES AND PROCEDURES”PRACTICES AND PROCEDURES”

• A standard practice is…

“A definition of a work method wherein all

variables of the method have been specified

in detail.”

• It is a written agreement between the worker

and the company regarding how the job will

be done.

68


WHAT DEFINES A STANDARD PRACTICE?WHAT DEFINES A STANDARD PRACTICE?

•Purpose: To gather technology and process skill in written form and to make it easier for everyone to do their work.

•Contents:

– Written so even workers who are not fully trained can use them easily

– Specific to the job, written at an appropriate level of detail

– Describe how to prevent variation

– Provide clear instructions

– Considered priorities

69


TRAININGTRAINING

– When you have completed the documentation, you need to make sure that everyone using a common process is trained in the new methods.

– Even experienced employees need to be trained in the new methods.

70


� It is an important tool to give advance warnings

about the drift from the normal behavior.

�To draw control charts, for a process we need to

know the average and standard deviation

obtained from the process.

SHEWHART CONTROL CHARTSSHEWHART CONTROL CHARTS

71


DISPLAYING BEFORE & AFTER DATADISPLAYING BEFORE & AFTER DATA

• Add more data to an

existing run chart or

control chart.

• Prepare new Pareto charts

for those you created in

Step 2. Make scale and

dimensions the same so you

can more accurately judge

degree of improvement.

• Draw new frequency plots

on the same scale as the

original plots.

Before After

}Improvement

Target

}Remaining Gap

Good

Step 4 changesimplemented

} Improvement

Before After

A1 A2 A3 A4 A2 A1 A3 A4

Before After

72


RECALCULATE PROCESS SIGMARECALCULATE PROCESS SIGMA

Step BEFOREOpportunitiesUnits

1

2

3

4

5

6

7

8

9

10

11

12

40

40

2 x 40 = 80 61

37

113

34

142

38

69

143

37

64

72

35

1 x 40 = 40

3 x 40 = 120

1 x 40 = 40

4 x 40 = 160

1 x 40 = 40

2 x 40 = 80

4 x 40 = 160

1 x 40 = 40

2 x 40 = 80

2 x 40 = 80

1 x 40 = 40

40

40

40

40

40

40

40

40

40

40

960 846

AFTER

17

1

12

0

8

0

3

37

0

5

2

1

86

Yield = 1 –

= 1 – .88

= 12%

Sigma = 0.3

960

846

Yield = 1 –

= 1 – .09

= 91%

Sigma = 2.8

960

86

BEFORE

AFTER

Six Sigma

Documents

due sibility datecalvin

wiring schematic conduct

dependent variable output

approval date

wiring plan machine

data based validation

process effect symptom

adjust check