SPC - Second Edition Statistical Process Control -

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh
tp o w
en
s Ma
ro ned e E gn
tec b xp aI
ted y t ira nte
an h e t io rn
d h Aut n: 3/ ati
o on
as m 31/ al
be otiv 20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Second Edition
SPC
Statistical Process Control
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h
an e A n: t i o rn
d h ut ati
o 3/3 on
as m 1 al
(SPC)

be otiv /20 Inc

en e 11

ISBN# 978-1-60534-108-8
wa Ind
CONTROL

Copyright © 1992, © 1995, © 2005

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
REFERENCE MANUAL

din p
gly
.
STATISTICAL PROCESS

Daimler Chrysler Corporation, Ford Motor Company, and General Motors Corporation
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h
an e A n: t i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 STATISTICAL PROCESS CONTROL
SPC
FOREWORD to Second Edition
This Reference Manual was developed by the Statistical Process Control (SPC) Work Group, sanctioned
by the DaimlerChrysler/Ford/General Motors Supplier Quality Requirements Task Force, and under the
auspices of the American Society for Quality (ASQ) and the Automotive Industry Action Group (AIAG).
The Work Group responsible for this Second edition was prepared by the quality and supplier assessment

din p
staffs at DaimlerChrysler Corporation, Delphi Corporation, Ford Motor Company, General Motors

.
or ou
gly
Corporation, Omnex, Inc. and Robert Bosch Corporation working in collaboration with the Automotive

cc Gr
Industry Action Group (AIAG).

Inc

d a ion
rke Act
The Task Force charter is to standardize the reference manuals, reporting formats and technical

al
on
nomenclature used by DaimlerChrysler, Ford and General Motors in their respective supplier assessment

ma ry
ati
systems. Accordingly, this Reference Manual can be used by any supplier to develop information

ter ust
en e 11
rn
responding to the requirements of either DaimlerChrysler’s, Ford’s or General Motors’ supplier
be otiv /20
wa Ind
nte

assessment systems. This second edition was prepared to recognize the needs and changes within the
automotive industry in SPC techniques that have evolved since the original manual was published in
1
aI

3/3

1991.
gn

an e A n:
Ma

as m

The manual is an introduction to statistical process control. It is not intended to limit evolution of SPC
o
o
i
d h ut
t
to:

methods suited to particular processes or commodities. While these guidelines are intended to cover
ted y t ira

normally occurring SPC system situations, there will be questions that arise. These questions should be
xp
op en ite 426 sed

directed to your customer’s Supplier Quality Assurance (SQA) activity. If you are uncertain as to how to
h
ro ed E
n

contact the appropriate SQA activity, the buyer in your customer’s purchasing office can help.
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

The Task Force gratefully acknowledges: the leadership and commitment of Vice Presidents Peter
an is d ume be t is
5

Rosenfeld at DaimlerChrysler Corporation, Thomas K. Brown at Ford Motor Company and Bo

tec
Th Doc Num men

Andersson of General Motors Corporation; the assistance of the AIAG in the development, production
s c um S

and distribution of the manual; the guidance of the Task Force principals Hank Gryn (DaimlerChrysler
E- der ocu

Corporation), Russ Hopkins (Ford Motor Company), and Joe Bransky (General Motors Corporation).
gh
Or is d

Therefore this manual was developed to meet the specific needs of the automotive industry.
Th

This Manual is copyrighted by DaimlerChrysler Corporation, Ford Motor Company, and General Motors
Corporation, all rights reserved, 2005. Additional manuals can be ordered from AIAG and/or permission
to copy portions of this manual for use within supplier organizations may be obtained from AIAG at 248-
358-3570 or http://www.aiag.org.

i
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t

ii
an e A n: i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
ACKNOWLEDGEMENTS to Second Edition
The joint consensus on the contents of this document was effected through Task Team Subcommittee
Members representing DaimlerChrysler, Ford, and General Motors, respectively, whose approval
signatures appear below, and who gratefully acknowledge the significant contribution of Gregory Gruska
of Omnex Inc., Gary A. Hiner of Delphi Corporation, and David W. Stamps of The Robert Bosch Corp.

The latest improvements were updating the format to conform to the current AIAG/ ISO/ TS 16949:2002
documentation, more clarification and examples to make the manual more user friendly and additional
areas which where not included or did not exist when the original manual was written.

The current re-write subcommittee is chaired by Mike Down from General Motors Corporation and

din p
.
or ou
consists of Todd Kerkstra and Dave Benham from DaimlerChrysler Corporation, Peter Cvetkovski from

gly
Ford Motor Company, Gregory Gruska, as a representative of the Omnex Inc. and ASQ, Gary A. Hiner of

cc Gr
Delphi Corporation, and David W. Stamps of The Robert Bosch Corp.

Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Michael H. Down Todd Kerkstra

General Motors Corporation DaimlerChrysler Corporation

Peter Cvetkovski David R. Benham

Ford Motor Company DaimlerChrysler Corporation

iii
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t

iv
an e A n: i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 STATISTICAL PROCESS CONTROL
SPC
FOREWORD to First Edition
This Reference Manual was prepared by the quality and supplier assessment staffs at Chrysler, Ford and
General Motors, working under the auspices of the Automotive Division of the American Society for Quality
Control Supplier Quality Requirements Task Force, in collaboration with the Automotive Industry Action
Group.

din p
.
or ou
The ASQC/AIAG Task Force charter is to standardize the reference manuals, reporting formats and technical

gly
nomenclature used by Chrysler, Ford and General Motors in their respective supplier assessment systems:

cc Gr
Supplier Quality Assurance, Total Quality Excellence and Targets for Excellence. Accordingly, this Reference

Inc

d a ion
Manual can be used by any supplier to develop information responding to the requirements of either

rke Act
al
Chrysler’s, Ford’s or General Motors’ supplier assessment systems. Until now, there has been no unified

on
formal approach in the automotive industry on statistical process control. Certain manufacturers provided

ma ry
ati
methods for their suppliers, while others had no specific requirements. In an effort to simplify and minimize

ter ust
en e 11
rn
variation in supplier quality requirements, Chrysler, Ford, and General Motors agreed to develop and, through
be otiv /20
wa Ind
nte

AIAG, distribute this manual. The work team responsible for the Manual’s content was led by Leonard A.
Brown of General Motors. The manual should be considered an introduction to statistical process control. It is
1
aI

3/3

not intended to limit evolution of statistical methods suited to particular processes or commodities nor is it
gn

intended to be comprehensive of all SPC techniques. Questions on the use of alternate methods should be
an e A n:
Ma

as m

referred to your customer’s quality activity.

o
o
i
d h ut
t
to:

ted y t ira

The Task Force gratefully acknowledges: the senior leadership and commitment of Vice Presidents Thomas T.
xp
op en ite 426 sed

Stallkamp at Chrysler, Clinton D. Lauer at Ford, and Donald A. Pais at General Motors; the technical
h
ro ed E

competence and hard work of their quality and supplier assessment teams; and the invaluable contributions of
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

the Automotive Industry Action Group (under AIAG Executive Director Joseph R. Phelan) in the
b
en

development, production and distribution of this Reference manual. We also wish to thank the ASQC reading
an is d ume be t is

team led by Tripp Martin of Peterson Spring, who reviewed the Manual and in the process made valuable
5

tec

contributions to intent and content.

Th Doc Num men

o
s c um S

Bruce W. Pince
E- der ocu

Task Force Coordinator

gh
Or is d

Sandy Corporation
Troy, Michigan
Th

December, 1991

This Manual is copyrighted by Chrysler Corporation, Ford Motor Company, General Motors Corporation,
all rights reserved, 1991. Additional copies can be ordered from A.I.A.G., and/or permission to copy portions
of the Manual for use within supplier organizations may be obtained from A.I.A.G. at (248) 358-3570.

v
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t

vi
an e A n: i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
ACKNOWLEDGEMENTS to First Edition
The joint consensus on the contents of this document was effected through Task Team Subcommittee
Members representing General Motors, Ford, and Chrysler, respectively, whose approval signatures appear
below, and who gratefully acknowledge the significant contribution of Pete Jessup of the Ford Motor
Company, who was responsible for developing the majority of the material found in Chapters I, II, and III, and
the Appendix of this document.

Harvey Goltzer of the Chrysler Corporation contributed concepts relative to process capability and capability
studies, found in the introduction section of Chapter I. Jack Herman of Du Pont contributed some of the
concepts relative to capability and performance indices and the importance of measurement variability, found
in portions of Chapters II and IV, respectively.

din p
.
or ou
gly
The General Motors Powertrain Division contributed the discussion and examples relative to subgrouping and

cc Gr
process over-adjustment. The section in Chapter II which provides understanding of process capability and

Inc

d a ion
related issues was developed by the General Motors Corporate Statistical Review Committee. This committee
also contributed to the development of Chapter IV, Process Measurement Systems Analysis, as well as to some

rke Act
al
on
Appendix items.

ma ry
ati

ter ust
en e 11
Finally, valuable input to all sections of the manual was provided by ASQC representatives Gregory Gruska,
rn

Doug Berg, and Tripp Martin. be otiv /20

wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Leonard A. Brown, Victor W. Lowe, Jr David R. Benham,

G.M. Ford Chrysler

vii
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o

viii
an e A n: rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 TABLE OF CONTENTS
CHAPTER I ................................................................................................................................................ 1
Continual Improvement and Statistical Process Control......................................................................... 1
Introduction ........................................................................................................................................... 3
Six Points ............................................................................................................................................... 4
CHAPTER I – Section A ............................................................................................................................ 7
Prevention Versus Detection ..................................................................................................................... 7
CHAPTER I – Section B ............................................................................................................................ 9
A Process Control System......................................................................................................................... 9
CHAPTER I – Section C .......................................................................................................................... 13
Variation: Common................................................................................................................................. 13
and Special Causes .................................................................................................................................. 13

din p
.
or ou
gly
CHAPTER I – Section D .......................................................................................................................... 17

cc Gr
Local Actions And Actions On The System ........................................................................................... 17

Inc

d a ion
CHAPTER I – Section E .......................................................................................................................... 19

rke Act
al
Process Control and Process Capability .................................................................................................. 19

on
Control vs. Capability.......................................................................................................................... 19

ma ry
ati
Process Indices .................................................................................................................................... 21

ter ust
en e 11
rn

CHAPTER I – Section F........................................................................................................................... 25

be otiv /20
wa Ind
nte

The Process Improvement Cycle and Process Control............................................................................ 25

1
aI

CHAPTER I – Section G .......................................................................................................................... 29

3/3
gn

Control Charts: Tools For Process Control and Improvement ............................................................... 29

an e A n:
Ma

as m

How do they work? .............................................................................................................................. 30

o
o

Approach: ............................................................................................................................................ 32
i
d h ut
t
to:

ted y t ira

CHAPTER I – Section H .......................................................................................................................... 37

xp
op en ite 426 sed

Effective Use and Benefits of Control Charts ......................................................................................... 37

h
ro ed E
n

CHAPTER II............................................................................................................................................. 41
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Control Charts ......................................................................................................................................... 41

en

Introduction:........................................................................................................................................ 43
an is d ume be t is
5

tec

Variables Control Charts .................................................................................................................... 45

Th Doc Num men

Attributes Control Charts .................................................................................................................... 47

s c um S

Elements of Control Charts ................................................................................................................. 48

E- der ocu

gh

CHAPTER II - Section A ......................................................................................................................... 53

Or is d

Control Chart Process.............................................................................................................................. 53

Preparatory Steps ................................................................................................................................ 53
Th

Control Chart Mechanics .................................................................................................................... 55

Establish Control Limits .................................................................................................................. 59
Interpret for Statistical Control ........................................................................................................ 60
Final Comments ............................................................................................................................... 63
Extend Control Limits for Ongoing Control.................................................................................... 65
CHAPTER II - Section B ......................................................................................................................... 69
Defining “Out-of-Control” Signals ......................................................................................................... 69
Point Beyond a Control Limit.............................................................................................................. 69
Patterns or Trends Within the Control Limits ..................................................................................... 70
Special Cause Criteria......................................................................................................................... 75
Average Run Length (ARL).................................................................................................................. 76
CHAPTER II - Section C ......................................................................................................................... 79
Control Chart Formulas........................................................................................................................... 79

ix
 Variables Control Charts .................................................................................................................... 79
Average and Range Charts ( X , R ) ............................................................................................. 79
Average and Standard Deviation Charts ( X , s ) ........................................................................... 83

Median and Range Charts ( X , R ) ................................................................................................. 85

Individuals and Moving Range Charts ( X , MR ) ........................................................................ 87
Attributes Control Charts .................................................................................................................... 89
Control Charts for Nonconforming Items ........................................................................................ 89
Proportion Nonconforming (p Chart)............................................................................................... 89
Number of Nonconforming Chart (np Chart) .................................................................................. 93
Number of Nonconformities per Unit Chart (u Chart)..................................................................... 95

din p
.
or ou
gly
Number of Nonconformities Chart (c Chart) ................................................................................... 97

cc Gr
CHAPTER III ........................................................................................................................................... 99

Inc

d a ion
Other Types of Control Charts ................................................................................................................ 99
Introduction ....................................................................................................................................... 101

rke Act
al
Probability Based Charts .................................................................................................................. 101
on

ma ry
Short-Run Control Charts.................................................................................................................. 107
ati

ter ust
en e 11
Charts for Detecting Small Changes ................................................................................................. 109
rn

be otiv /20
wa Ind
Non-Normal Charts ........................................................................................................................... 113
nte

Multivariate ....................................................................................................................................... 116

1
aI

3/3

Other Charts ...................................................................................................................................... 117

gn

Regression Control Charts ............................................................................................................. 117

an e A n:
Ma

as m

Residual Charts .............................................................................................................................. 118

o
o
i

Autoregressive Charts .................................................................................................................... 118

d h ut
t
to:

ted y t ira

Zone Charts .................................................................................................................................... 121

xp
op en ite 426 sed

CHAPTER IV.......................................................................................................................................... 125

h
ro ed E
n

Understanding Process Capability......................................................................................................... 125

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

and Process Performance for Variables Data ........................................................................................ 125

en

Introduction ....................................................................................................................................... 127

an is d ume be t is
5

tec

CHAPTER IV - Section A ...................................................................................................................... 131

Th Doc Num men

Definitions of Process Terms ................................................................................................................ 131

s c um S

Process Measures for Predictable Processes ......................................................................................... 132

E- der ocu

gh

Indices – Bilateral Tolerances.......................................................................................................... 132

Or is d

Indices – Unilateral Tolerances ....................................................................................................... 137

Th

CHAPTER IV - Section B ...................................................................................................................... 139

Description of Conditions ..................................................................................................................... 139
Handling Non-Normal and Multivariate Distributions ......................................................................... 140
Relationship of Indices and Proportion Nonconforming................................................................... 140
Non-Normal Distributions Using Transformations........................................................................... 140
Non-Normal Distributions Using Non-Normal Forms ...................................................................... 142
Multivariate Distributions ................................................................................................................. 144
CHAPTER IV - Section C ...................................................................................................................... 147
Suggested Use of Process Measures ..................................................................................................... 147
The Loss Function Concept ............................................................................................................... 148
Alignment of Process to Customer Requirements.............................................................................. 153
APPENDIX A .......................................................................................................................................... 157
Some Comments on Sampling .............................................................................................................. 157
Effects of Subgrouping....................................................................................................................... 157

x
 Autocorrelated Data .......................................................................................................................... 157
Multiple Stream Process Example..................................................................................................... 162
Effects of Sample Size on Indices....................................................................................................... 168
APPENDIX B .......................................................................................................................................... 171
Some Comments on Special Causes ..................................................................................................... 171
Over-Adjustment ................................................................................................................................ 171
Time Dependent Processes ................................................................................................................ 173
Repeating Patterns............................................................................................................................. 175
APPENDIX C .......................................................................................................................................... 177
Selection Procedure for the Use of the Control Charts Described in This Manual............................... 177
APPENDIX D .......................................................................................................................................... 179
Relationship Between Cpm and Other Indices ....................................................................................... 179
APPENDIX E .......................................................................................................................................... 181

din p
.
Table of Constants and Formulas for Control Charts............................................................................ 181

or ou
gly
cc Gr
APPENDIX F .......................................................................................................................................... 185
Capability Index Calculations Example ................................................................................................ 185

Inc

d a ion
Data Set: ................................................................................................................................................ 186

rke Act
al
Analysis................................................................................................................................................. 187
on
Diameter Statistics: ........................................................................................................................... 188

ma ry
ati
Conclusion:........................................................................................................................................ 190

ter ust
en e 11
rn

APPENDIX G.......................................................................................................................................... 191

be otiv /20
wa Ind
nte

Glossary of Terms and Symbols ........................................................................................................... 191

1
aI

Terms Used in This Manual............................................................................................................... 191

3/3
gn

Symbols as Used in This Manual...................................................................................................... 204

an e A n:
Ma

as m

APPENDIX H.......................................................................................................................................... 211

o
o
i
d h ut
t

References and Suggested Readings ..................................................................................................... 211

to:

ted y t ira

APPENDIX I ........................................................................................................................................... 215

xp
op en ite 426 sed

Standard Normal Tables........................................................................................................................ 215

ro ed E
n

t p wn se
yri t is Lic 2

INDEX...................................................................................................................................................... 217
d i oc nt r: 4 lice

b
en

S.P.C. Manual User Feedback Process ................................................................................................. 221

an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

xi
 LIST OF ILLUSTRATIONS
Figure I.1: A Process Control System........................................................................................................... 8
Figure I.2: Variation: Common Cause and Special Cause.......................................................................... 12
Figure I.3: Process Control and Process Capability.................................................................................... 18
Figure I.4: The Process Improvement Cycle .............................................................................................. 24
Figure I.5: Control Charts ........................................................................................................................... 28
Figure II.1: Variables Data ......................................................................................................................... 44
Figure II.2: Attributes Data......................................................................................................................... 46
Figure II.3: Elements of Control Charts...................................................................................................... 49
Figure II.4a: Sample Control Chart (Front side)........................................................................................ 51
Figure II.4b: Sample Control Chart (back side) – Event Log ..................................................................... 52
Figure II.5: Extending Control Limits ........................................................................................................ 56

din p
Figure II.6: Control Limits Recalculation................................................................................................... 61

.
or ou
gly
Figure II.7: Extend Control Limits for Ongoing Control............................................................................ 64

cc Gr
Figure II.8: Process Variation Relative to Specification Limits ................................................................. 67

Inc

d a ion
Figure II.9: Points Beyond Control Limits ................................................................................................. 70
Figure II.10: Runs in an Average Control Chart......................................................................................... 71

rke Act
al
Figure II.11: Runs in a Range Control Chart .............................................................................................. 72
on

ma ry
Figure II.12: Nonrandom Patterns in a Control Chart................................................................................. 74
ati

ter ust
en e 11
Figure II.13: Average and Range Charts .................................................................................................... 78
rn

be otiv /20
wa Ind
nte

Figure II.14: Average and Standard Deviation Charts................................................................................ 82

Figure II.15: Median and Range Charts...................................................................................................... 84
1
aI

3/3

Figure II.16: Individual and Moving Range Charts .................................................................................... 86

gn

Figure II.17: Proportion Nonconforming Chart .......................................................................................... 88

an e A n:
Ma

as m

Figure II.18: Number of Nonconforming Chart.......................................................................................... 92

o
o
i

Figure II.19: Number of Nonconforming per Unit Chart ........................................................................... 94

d h ut
t
to:

ted y t ira

Figure II.20: Number of Nonconformities Chart ........................................................................................ 96

xp
op en ite 426 sed

Figure III.1: Control Charts ...................................................................................................................... 100

h
ro ed E

Figure III.2: Stoplight Control .................................................................................................................. 102

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Figure III.3: Pre-Control ........................................................................................................................... 105

en

Figure III.4: DNOM Control Chart........................................................................................................... 108

an is d ume be t is
5

Figure III.5: CUSUM Chart with V-Mask................................................................................................ 109

tec
Th Doc Num men

Figure III.6: X, MR Chart .......................................................................................................................... 110

Figure III.7: EWMA Chart of Viscosity ................................................................................................... 112
s c um S
E- der ocu

Figure III.8: X, MR Chart of Viscosity..................................................................................................... 112

gh

Figure IV.1: Within- and Between-Subgroup Variation........................................................................... 130

Or is d

Figure IV.2: Cpk and Ppk Comparison ........................................................................................................ 133

Th

Figure IV.3: Comparison between a Predictable and Immature Process.................................................. 135

Figure IV.4: Cpk and Ppk Values Produced by a Predictable and Immature Process .............................. 136
Figure IV.5: “Goal Post” vs. Loss Function ............................................................................................. 148
Figure IV.6: Comparison of Loss Function and Specifications ................................................................ 150
Figure IV.7: Comparison of Loss Functions............................................................................................. 151
Figure IV.8: A Process Control System.................................................................................................... 152
Figure IV.9: Process Alignment to Requirements .................................................................................... 154

xii
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t

1
an e A n: i o rn

and
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
CHAPTER I

wa Ind
ter ust
ma ry
rke Act
Continual Improvement

d a ion
Statistical Process Control

cc Gr
or ou
din p
gly
.
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t

2
an e A n: i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER I
Continual Improvement and Statistical Process Control

Introduction
To prosper in today’s economic climate, we – automotive manufacturers,
suppliers and dealer organizations – must be dedicated to continual
improvement. We must constantly seek more efficient ways to produce
products and services. These products and services must continue to
improve in value. We must focus upon our customers, both internal and
external, and make customer satisfaction a primary business goal.
To accomplish this, everyone in our organizations must be committed to
improvement and to the use of effective methods. This manual describes
several basic statistical methods that can be used to make our efforts at
improvement more effective. Different levels of understanding are
needed to perform different tasks. This manual is aimed at practitioners

din p
.
or ou
and managers beginning the application of statistical methods. It will

gly
also serve as a refresher on these basic methods for those who are now

cc Gr
using more advanced techniques. Not all basic methods are included

Inc

d a ion
here. Coverage of other basic methods (such as check sheets, flowcharts,

rke Act
al
Pareto charts, cause and effect diagrams) and some advanced methods
on
(such as other control charts, designed experiments, quality function

ma ry
ati
deployment, etc.) is available in books and booklets such as those

ter ust
en e 11
rn

referenced in Appendix H.
be otiv /20
wa Ind
nte

The basic statistical methods addressed in this manual include those

1
aI

3/3

associated with statistical process control and process capability analysis.

gn

an e A n:

Chapter I provides background for process control, explains several

Ma

as m
o
o

important concepts such as special and common causes of variation. It

i
d h ut
t
to:

ted y t ira

also introduces the control chart, which can be a very effective tool for
xp

analyzing and monitoring processes.

op en ite 426 sed

h
ro ed E

Chapter II describes the construction and use of control charts for both
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

variables1 data and attributes data.

en
an is d ume be t is

Chapter III describes other types of control charts that can be used for
5

tec

specialized situations – probability based charts, short-run charts, charts

Th Doc Num men

for detecting small changes, non-normal, multivariate and other charts.

s c um S
E- der ocu

Chapter IV addresses process capability analysis.

gh
Or is d

The Appendices address sampling, over-adjustment, a process for

selecting control charts, table of constants and formulae, the normal
Th

table, a glossary of terms and symbols, and references.

1
The term “Variables”, although awkward sounding, is used in order to distinguish the difference
between something that varies, and the control chart used for data taken from a continuous variable.

3
CHAPTER I
Continual Improvement and Statistical Process Control

Six Points
Six points should be made before the main discussion begins:

1) Gathering data and using statistical methods to interpret them are not
ends in themselves. The overall aim should be increased
understanding of the reader’s processes. It is very easy to become
technique experts without realizing any improvements. Increased
knowledge should become a basis for action.
2) Measurement systems are critical to proper data analysis and they
should be well understood before process data are collected. When

din p
such systems lack statistical control or their variation accounts for a

.
or ou
gly
substantial portion of the total variation in process data,

cc Gr
inappropriate decisions may be made. For the purposes of this

Inc

d a ion
manual, it will be assumed that this system is under control and is
not a significant contributor to total variation in the data. The reader

rke Act
al
is referred to the Measurement Systems Analysis (MSA) Manual
on

ma ry
available from AIAG for more information on this topic.
ati

ter ust
en e 11
rn

3) The basic concept of studying variation and using statistical signals

be otiv /20
wa Ind
nte

to improve performance can be applied to any area. Such areas can

1
aI

be on the shop floor or in the office. Some examples are machines

3/3
gn

(performance characteristics), bookkeeping (error rates), gross sales,

an e A n:

waste analysis (scrap rates), computer systems (performance

Ma

as m
o
o

characteristics) and materials management (transit times). This

i
d h ut
t
to:

ted y t ira

manual focuses upon shop floor applications. The reader is

xp

encouraged to consult the references in Appendix H for

op en ite 426 sed

h
ro ed E

administrative and service applications.

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

4) SPC stands for Statistical Process Control. Historically, statistical

en

methods have been routinely applied to parts, rather than processes.

an is d ume be t is
5

Application of statistical techniques to control output (such as parts)

tec
Th Doc Num men

should be only the first step. Until the processes that generate the
o

output become the focus of our efforts, the full power of these
s c um S
E- der ocu

methods to improve quality, increase productivity and reduce cost

gh

may not be fully realized.

Or is d

5) Although each point in the text is illustrated with a worked-out

Th

example, real understanding of the subject involves deeper contact

with process control situations. The study of actual cases from the
reader’s own job location or from similar activities would be an
important supplement to the text. There is no substitute for hands-on
experience.
6) This manual should be considered a first step toward the use of
statistical methods. It provides generally accepted approaches,
which work in many instances. However, there exist exceptions
where it is improper to blindly use these approaches. This manual
does not replace the need for practitioners to increase their
knowledge of statistical methods and theory. Readers are
encouraged to pursue formal statistical education. Where the
reader’s processes and application of statistical methods have

4
 CHAPTER I
Continual Improvement and Statistical Process Control

advanced beyond the material covered here, the reader is also

encouraged to consult with persons who have the proper knowledge
and practice in statistical theory as to the appropriateness of other
techniques. In any event, the procedures used must satisfy the
customer's requirements.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

5
 CHAPTER I – Section A

Th
Prevention Versus Detection

Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t

6
an e A n: i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
THE NEED FOR

ma ry
rke Act
Prevention – Avoids Waste
d a ion
Detection – Tolerates Waste
PROCESS CONTROL

cc Gr
or ou
din p
gly
.
 CHAPTER I – Section A
Prevention Versus Detection

CHAPTER I – Section A
Prevention Versus Detection
In the past, Manufacturing often depended on Production to make the
product and on Quality Control to inspect the final product and screen
out items not meeting specifications. In administrative situations, work
is often checked and rechecked in efforts to catch errors. Both cases
involve a strategy of detection, which is wasteful, because it allows time
and materials to be invested in products or services that are not always
usable.

din p
.
It is much more effective to avoid waste by not producing unusable

or ou
gly
output in the first place – a strategy of prevention.

cc Gr
Inc
A prevention strategy sounds sensible – even obvious – to most people.

d a ion
It is easily captured in such slogans as, “Do it right the first time”.

rke Act
al
However, slogans are not enough. What is required is an understanding
on
of the elements of a statistical process control system. The remaining

ma ry
ati

ter ust
seven subsections of this introduction cover these elements and can be
en e 11
rn

viewed as answers to the following questions:

be otiv /20
wa Ind
nte

•
aI

What is meant by a process control system?

3/3
gn

• How does variation affect process output?

an e A n:
Ma

as m
o
o

• How can statistical techniques tell whether a problem is local in

i
d h ut
t
to:

ted y t ira

nature or involves broader systems?

xp
op en ite 426 sed

• What is meant by a process being in statistical control?

h
ro ed E
n

What is meant by a process being capable?

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

• What is a continual improvement cycle, and what part can process

an is d ume be t is
5

control play in it?

tec
Th Doc Num men

• What are control charts, and how are they used?

s c um S
E- der ocu

• What benefits can be expected from using control charts?

gh
Or is d
Th

As this material is being studied, the reader may wish to refer to the
Glossary in Appendix G for brief definitions of key terms and symbols.

7
CHAPTER I – Section B
A Process Control System

PROCESS CONTROL SYSTEM MODEL

WITH FEEDBACK

VOICE

din p
OF THE

.
or ou
gly
PROCESS

cc Gr
Inc

d a ion
rke Act
al
STATISTICAL
METHODS on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

PEOPLE
1
aI

3/3

EQUIPMENT THE WAY

PRODUCTS
gn

MATERIAL WE WORK/
an e A n:
Ma

OR CUSTOMERS
as m

METHODS BLENDING OF
o
o

SERVICES
i
d h ut
t
to:

RESOURCES
ted y t ira

MEASUREMENT
xp

ENVIRONMENT
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

IDENTIFYING
an is d ume be t is
5

CHANGING NEEDS
tec
Th Doc Num men

INPUTS PROCESS/SYSTEM OUTPUTS AND EXPECTATIONS

o
s c um S
E- der ocu

gh
Or is d

VOICE
OF THE
Th

CUSTOMER

Figure I.1: A Process Control System

8
 CHAPTER I – Section B
A Process Control System

CHAPTER I – Section B
A Process Control System
A process control system can be described as a feedback system. SPC is
one type of feedback system. Other such systems, which are not
statistical, also exist. Four elements of that system are important to the
discussions that will follow:
1. The Process – By the process, we mean the whole combination of
suppliers, producers, people, equipment, input materials, methods, and
environment that work together to produce output, and the customers

din p
.
who use that output (see Figure I.1). The total performance of the

or ou
gly
process depends upon communication between supplier and customer,

cc Gr
the way the process is designed and implemented, and on the way it is

Inc

d a ion
operated and managed. The rest of the process control system is useful

rke Act
al
only if it contributes either to maintaining a level of excellence or to

on
improving the total performance of the process.

ma ry
ati

ter ust
2. Information About Performance – Much information about the
en e 11
rn

actual performance of the process can be learned by studying the process

be otiv /20
wa Ind
nte

output. The most helpful information about the performance of a process

1
aI

comes, however, from understanding the process itself and its internal
3/3
gn

variability. Process characteristics (such as temperatures, cycle times,

an e A n:
Ma

as m

feed rates, absenteeism, turnover, tardiness, or number of interruptions)

o
o

should be the ultimate focus of our efforts. We need to determine the

i
d h ut
t
to:

ted y t ira

target values for those characteristics that result in the most productive
xp
op en ite 426 sed

operation of the process, and then monitor how near to or far from those
h
ro ed E

target values we are. If this information is gathered and interpreted

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

correctly, it can show whether the process is acting in a usual or unusual

b
en

manner. Proper actions can then be taken, if needed, to correct the

an is d ume be t is
5

process or the just-produced output. When action is needed it must be

tec
Th Doc Num men

timely and appropriate, or the information-gathering effort is wasted.

o
s c um S

3. Action on the Process – Action on the process is frequently most

E- der ocu

economical when taken to prevent the important characteristics (process

gh
Or is d

or output) from varying too far from their target values. This ensures the
stability and the variation of the process output is maintained within
Th

acceptable limits. Such action might consist of:

• Changes in the operations
operator training
changes to the incoming materials
• Changes in the more basic elements of the process itself
the equipment
how people communicate and relate
the design of the process as a whole – which may be vulnerable
to changes in shop temperature or humidity
The effect of actions should be monitored, with further analysis and
action taken if necessary.

9
CHAPTER I – Section B
A Process Control System

4. Action on the Output – Action on the output is frequently least

economical when it is restricted to detecting and correcting out-of-
specification product without addressing the underlying process problem.
Unfortunately, if current output does not consistently meet customer
requirements, it may be necessary to sort all products and to scrap or
rework any nonconforming items. This must continue until the
necessary corrective action on the process has been taken and verified.
It is obvious that inspection followed by action on only the output is a
poor substitute for effective process management. Action on only the
output should be used strictly as an interim measure for unstable or
incapable processes (see Chapter I, Section E). Therefore, the
discussions that follow focus on gathering process information and
analyzing it so that action can be taken to correct the process itself.
Remember, the focus should be on prevention not detection.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

10
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

11
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
CHAPTER I – Section B
A Process Control System
CHAPTER I – Section C
Variation: Common and Special Causes

PIECES VARY FROM EACH OTHER

SIZE SIZE SIZE SIZE

BUT THEY FORM A PATTERN THAT, IF STABLE, CAN BE DESCRIBED AS A DISTRIBUTION

din p
.
or ou
gly
cc Gr
Inc

d a ion
SIZE SIZE SIZE

rke Act
al
on
DISTRIBUTION CAN DIFFER IN:

ma ry
ati
LOCATION SPREAD

ter ust
SHAPE

en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m

SIZE SIZE SIZE

o
o
i
d h ut
t
to:

ted y t ira
xp

IF ONLY COMMON CAUSES OF VARIATION

op en ite 426 sed

GET
ARE PRESENT, THE OUTPUT OF A PROCESS TAR
h
ro ed E

FORMS A DISTRIBUTION THAT IS STABLE OVER

n

E
LIN
t p wn se

TIME AND IS PREDICTABLE:

yri t is Lic 2
d i oc nt r: 4 lice

b
en

PREDICTION
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S

E
TIM
E- der ocu

SIZE ? GET
? ?? TAR
gh

?
Or is d

IF SPECIAL CAUSES OF VARIATION ARE ? ?? LIN

E
? ??
PRESENT, THE PROCESS OUTPUT IS NOT
??
Th

STABLE OVER TIME:

? ??? ? ?? ????
UNABLE TO
PREDICT

SIZE

Figure I.2: Variation: Common Cause and Special Cause

12
 CHAPTER I – Section C
Variation: Common and Special Causes

CHAPTER I – Section C
Variation: Common and Special Causes
In order to effectively use process control measurement data, it is
important to understand the concept of variation, as illustrated in Figure
I.2.
No two products or characteristics are exactly alike, because any process
contains many sources of variability. The differences among products
may be large, or they may be immeasurably small, but they are always
present. The diameter of a machined shaft, for instance, would be
susceptible to potential variation from the machine (clearances, bearing

din p
.
wear), tool (strength, rate of wear), material (diameter, hardness),

or ou
gly
operator (part feed, accuracy of centering), maintenance (lubrication,

cc Gr
replacement of worn parts), environment (temperature, constancy of

Inc

d a ion
power supply) and measurement system. Another example is the time

rke Act
al
required to process an invoice could vary according to the people

on
performing various steps, the reliability of any equipment they were

ma ry
ati
using, the accuracy and legibility of the invoice itself, the procedures

ter ust
en e 11
rn

followed, and the volume of other work in the office.

be otiv /20
wa Ind
nte

Some sources of variation in the process cause short-term, piece-to-piece

1
aI

differences, e.g., backlash and clearances within a machine and its

3/3
gn

fixturing, or the accuracy of a bookkeeper’s work. Other sources of

an e A n:
Ma

as m

variation tend to cause changes in the output only over a longer period of
o
o

time. These changes may occur either gradually as with tool or machine
i
d h ut
t
to:

ted y t ira

wear, stepwise as with procedural changes, or irregularly as with

xp
op en ite 426 sed

environmental changes such as power surges. Therefore, the time period

h
ro ed E

and conditions over which measurements are made are critical since they
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

will affect the amount of the total variation that will be observed.
b
en

While individual measured values may all be different, as a group they

an is d ume be t is
5

tec

tend to form a pattern that can be described as a distribution (see Figure

Th Doc Num men

I.2). This distribution can be characterized by:

s c um S
E- der ocu

• Location (typical or “central” value)

gh
Or is d

• Spread (span or “width” of values from smallest to largest)

Th

• Shape (the pattern of variation – whether it is symmetrical, skewed,

etc.)
From the standpoint of minimum requirements, the issue of variation is
often simplified: parts within specification tolerances are acceptable,
parts beyond specification tolerances are not acceptable; reports on time
are acceptable, late reports are not acceptable. However, the goal should
be to maintain the location to a target value with minimal variability. To
manage any process and reduce variation, the variation should be traced
back to its sources. The first step is to make the distinction between
common and special causes of variation.
Common causes refer to the many sources of variation that consistently
acting on the process. Common causes within a process produce a stable
and repeatable distribution over time. This is called “in a state of

13
CHAPTER I – Section C
Variation: Common and Special Causes

statistical control,” “in statistical control,” or sometimes just “in control.”

Common causes yield a stable system of chance causes. If only common
causes of variation are present and do not change, the output of a process
is predictable.
Special causes (often called assignable causes) refer to any factors
causing variation that affect only some of the process output. They are
often intermittent and unpredictable. Special causes are signaled by one
or more points beyond the control limits or non-random patterns of
points within the control limits. Unless all the special causes of variation
are identified and acted upon, they may continue to affect the process
output in unpredictable ways. If special causes of variation are present,
the process output will not be stable over time.
The changes in the process distribution due to special causes can be

din p
.
either detrimental or beneficial. When detrimental, they need to be

or ou
gly
understood and removed. When beneficial, they should be understood

cc Gr
and made a permanent part of the process. With some mature processes2,

Inc

d a ion
the customer may give special allowance to run a process with a

rke Act
al
consistently occurring special cause. Such allowances will usually

on
require that the process control plans can assure conformance to

ma ry
ati
customer requirements and protect the process from other special causes

ter ust
en e 11
rn

(see Chapter I, Section E).

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

2
Processes that have undergone several cycles of continual improvement.

14
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

15
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Variation: Common and Special Causes
CHAPTER I – Section C
CHAPTER I – Section D
Local Actions And Actions On The System

LOCAL ACTIONS AND ACTIONS ON THE SYSTEM

din p
.
or ou
gly
cc Gr
Local Actions

Inc

d a ion
• Are usually required to eliminate special causes of variation

rke Act
al
• Can usually be taken by people close to the process
on

ma ry
• Can correct typically about 15% of process problems
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Actions on the System

1
aI

• Are usually required to reduce the variation due to common causes

3/3
gn

• Almost always require management action for correction

an e A n:
Ma

as m
o
o

• Are needed to correct typically about 85% of process problems

i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

16
 CHAPTER I – Section D
Local Actions And Actions On The System

CHAPTER I – Section D
Local Actions And Actions On The System
There is an important connection between the two types of variation just
discussed and the types of action necessary to reduce them.3
Simple statistical process control techniques can detect special causes of
variation. Discovering a special cause of variation and taking the proper
action is usually the responsibility of someone who is directly connected
with the operation. Although management can sometimes be involved to
correct the condition, the resolution of a special cause of variation

din p
.
or ou
gly
usually requires local action, i.e., by people directly connected with the

cc Gr
operation. This is especially true during the early process improvement
efforts. As one succeeds in taking the proper action on special causes,

Inc

d a ion
those that remain will often require management action, rather than local

rke Act
al
action.
on

ma ry
ati
These same simple statistical techniques can also indicate the extent of

ter ust
en e 11
common causes of variation, but the causes themselves need more
rn

be otiv /20
wa Ind
nte

detailed analysis to isolate. The correction of these common causes of

variation is usually the responsibility of management. Sometimes people
1
aI

3/3

directly connected with the operation will be in a better position to

gn

identify them and pass them on to management for action. Overall, the
an e A n:
Ma

as m

resolution of common causes of variation usually requires action on the

o
o
i
d h ut
t

system.
to:

ted y t ira
xp
op en ite 426 sed

Only a relatively small proportion of excessive process variation –

h
ro ed E

industrial experience suggests about 15% – is correctable locally by

n

t p wn se
yri t is Lic 2

people directly connected with the operation. The majority – the other
d i oc nt r: 4 lice

b
en

85% – is correctable only by management action on the system.

an is d ume be t is

Confusion about the type of action to take is very costly to the

5

tec
Th Doc Num men

organization, in terms of wasted effort, delayed resolution of trouble, and

o

aggravating problems. It may be wrong, for example, to take local action

s c um S
E- der ocu

(e.g., adjusting a machine) when management action on the system is

gh

required (e.g., selecting suppliers that provide consistent input

Or is d

materials).4 Nevertheless, close teamwork between management and

Th

those persons directly connected with the operation is a must for

enhancing reduction of common causes of process variation.

3
Dr. W. E. Deming has treated this issue in many articles; e.g., see Deming (1967).
4
These observations were first made by Dr. J. M. Juran, and have been borne out in Dr. Deming’s
experience.

17
CHAPTER I – Section E
Process Control and Process Capability

PROCESS CONTROL

IN CONTROL
(SPECIAL CAUSE ELIMINATED)

din p
.
or ou
gly
ME
TI

cc Gr
Inc

d a ion
SIZE

rke Act
al
on

ma ry
OUT OF CONTROL
ati

ter ust
en e 11
(SPECIAL CAUSES PRESENT)
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

PROCESS CAPABILITY
ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

IN CONTROL AND
b

CAPABLE
en

OF MEETING
an is d ume be t is
5

SPECIFICATIONS
tec

(VARIATION FROM COMMON

Th Doc Num men

CAUSES
HAS BEEN REDUCED)
s c um S
E- der ocu

gh
Or is d

SIZE
ME
TI
Th

IN CONTROL BUT NOT CAPABLE

OF MEETING SPECIFICATIONS
(VARIATION FROM COMMON CAUSES
IS EXCESSIVE)

Figure I.3: Process Control and Process Capability

18
 CHAPTER I – Section E
Process Control and Process Capability

CHAPTER I – Section E
Process Control and Process Capability
The process control system is an integral part of the overall business
management system.5 As such, the goal of the process control system is
to make predictions about the current and future state of the process. This
leads to economically sound decisions about actions affecting the
process. These decisions require balancing the risk of taking action when
action is not necessary (over-control or “tampering”) versus failing to
take action when action is necessary (under-control).6 These risks should
be handled, however, in the context of the two sources of variation -
special causes and common causes (see Figure I.3).

din p
.
or ou
gly
A process is said to be operating in statistical control when the only

cc Gr
sources of variation are common causes. One function of a process

Inc

d a ion
control system, then, is to provide a statistical signal when special causes

rke Act
al
of variation are present, and to avoid giving false signals when they are

on
not present. This allows appropriate action(s) to be taken upon those

ma ry
ati
special causes (either removing them or, if they are beneficial, making

ter ust
en e 11
rn

them permanent). be otiv /20

wa Ind
nte

The process control system can be used as a one-time evaluation tool but
1
aI

the real benefit of a process control system is realized when it is used as

3/3
gn

a continual learning tool instead of a conformance tool (good/bad,

an e A n:
Ma

as m

stable/not stable, capable/not capable, etc.)

o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E

Control vs. Capability

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is

When discussing process capability, two somewhat contrasting concepts

5

tec

need to be considered:
Th Doc Num men

•
s c um S

Process capability
E- der ocu

gh

• Process performance
Or is d

Process capability is determined by the variation that comes from

Th

common causes. It generally represents the best performance of the

process itself. This is demonstrated when the process is being operated
in a state of statistical control regardless of the specifications.
Customers, internal or external, are however more typically concerned
with the process performance; that is, the overall output of the process
and how it relates to their requirements (defined by specifications),
irrespective of the process variation.

5
See TS 16949.
6
See W. E. Deming, (1994), and W. Shewhart, (1931).

19
CHAPTER I – Section E
Process Control and Process Capability

In general, since a process in statistical control can be described by a

predictable distribution, the proportion of in-specification parts can be
estimated from this distribution. As long as the process remains in
statistical control and does not undergo a change in location, spread or
shape, it will continue to produce the same distribution of in-
specification parts.
Once the process is in statistical control the first action on the process
should be to locate the process on the target. If the process spread is
unacceptable, this strategy allows the minimum number of out-of-
specification parts to be produced. Actions on the system to reduce the
variation from common causes are usually required to improve the ability
of the process (and its output) to meet specifications consistently. For a
more detailed discussion of process capability, process performance and
the associated assumptions, refer to Chapter IV.

din p
.
or ou
gly
cc Gr
The process must first be brought into statistical control by detecting and

Inc

d a ion
acting upon special causes of variation. Then its performance is

rke Act
predictable, and its capability to meet customer expectations can be

al
on
assessed. This is a basis for continual improvement.

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Every process is subject to classification based on capability and control.

A process can be classified into 1 of 4 cases, as illustrated by the
1
aI

3/3

following chart:
gn

an e A n:
Ma

as m
o
o

Statistical Control
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

In-Control Out-of-Control
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Acceptable Case 1 Case 3

b
en

Capability
an is d ume be t is
5

Unacceptable Case 2 Case 4

tec
Th Doc Num men

o
s c um S
E- der ocu

To be acceptable, the process must be in a state of statistical control and

gh
Or is d

the capability (common cause variation) must be less than the tolerance.
The ideal situation is to have a Case 1 process where the process is in
Th

statistical control and the ability to meet tolerance requirements is

acceptable. A Case 2 process is in control but has excessive common
cause variation, which must be reduced. A Case 3 process meets
tolerance requirements but is not in statistical control; special causes of
variation should be identified and acted upon. In Case 4, the process is
not in control nor is it acceptable. Both common and special cause
variation must be reduced.
Under certain circumstances, the customer may allow a producer to run a
process even though it is a Case 3 process. These circumstances may
include:
• The customer is insensitive to variation within specifications (see
discussion on the loss function in Chapter IV).

20
 CHAPTER I – Section E
Process Control and Process Capability

• The economics involved in acting upon the special cause exceed the
benefit to any and all customers. Economically allowable special
causes may include tool wear, tool regrind, cyclical (seasonal)
variation, etc.
• The special cause has been identified and has been documented as
consistent and predictable.
In these situations, the customer may require the following:
• The process is mature.
• The special cause to be allowed has been shown to act in a consistent
manner over a known period of time.
• A process control plan is in effect which will assure conformance to

din p
.
specification of all process output and protection from other special

or ou
gly
causes or inconsistency in the allowed special cause.

cc Gr
Inc
See also Appendix A for a discussion on time dependent processes.

d a ion
Process Indices

rke Act
al
on

ma ry
ati
The accepted practice in the automotive industry is to calculate the

ter ust
en e 11
rn

capability (common cause variation) only after a process has been

be otiv /20
wa Ind
nte

demonstrated to be in a state of statistical control. These results are used

1
aI

as a basis for prediction of how the process will perform. There is little
3/3
gn

value in making predictions based on data collected from a process that

an e A n:

is not stable and not repeatable over time. Special causes are responsible
Ma

as m
o
o

for changes in the shape, spread, or location of a process distribution, and

i
d h ut
t
to:

ted y t ira

thus can rapidly invalidate prediction about the process. That is, in
xp

order for the various process indices and ratios to be used as

op en ite 426 sed

h
ro ed E

predictive tools, the requirement is that the data used to calculate

n

t p wn se
yri t is Lic 2

them are gathered from processes that are in a state of statistical

d i oc nt r: 4 lice

b
en

control.
an is d ume be t is
5

Process indices can be divided into two categories: those that are
tec
Th Doc Num men

calculated using within-subgroup estimates of variation and those using

s c um S

total variation when estimating a given index (see also chapter IV).
E- der ocu

gh

Several different indices have been developed because:

Or is d

1) No single index can be universally applied to all processes, and

Th

2) No given process can be completely described by a single index.

For example, it is recommended that Cp and Cpk both be used (see
Chapter IV), and further that they be combined with graphical techniques
to better understand the relationship between the estimated distribution
and the specification limits. In one sense, this amounts to comparing (and
trying to align) the “voice of the process” with the “voice of the
customer” (see also Sherkenbach (1991)).
All indices have weaknesses and can be misleading. Any inferences
drawn from computed indices should be driven by appropriate
interpretation of the data from which the indices were computed.

21
CHAPTER I – Section E
Process Control and Process Capability

Automotive companies have set requirements for process capability. It is

the reader’s responsibility to communicate with their customer and
determine which indices to use. In some cases, it might be best to use no
index at all. It is important to remember that most capability indices
include the product specification in the formula. If the specification is
inappropriate, or not based upon customer requirements, much time and
effort may be wasted in trying to force the process to conform. Chapter
IV deals with selected capability and performance indices and contains
advice on the application of those indices.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

22
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

23
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
This page intentionally left blank

wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Process Control and Process Capability
CHAPTER I – Section E
 CHAPTER I – Section F
The Process Improvement Cycle and Process Control

STAGES OF THE CONTINUAL PROCESS IMPROVEMENT CYCLE

1. ANALYZE THE PROCESS 2. MAINTAIN THE PROCESS

- What should the process be doing? - Monitor process performance

din p
- Detect special cause

.
- What can go wrong?

or ou
gly
- What is the process doing? variation and act upon it.

cc Gr
- Achieve a state of statistical control.

Inc

d a ion
- Determine capability

rke Act
al
on
PLAN DO PLAN DO

ma ry
ati

ter ust
en e 11
rn

1 be otiv /20
2
wa Ind
nte

1
aI

3/3

ACT ACT
gn

STUDY STUDY
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

PLAN DO
en
an is d ume be t is
5

3
tec
Th Doc Num men

3. IMPROVE THE PROCESS

s c um S
E- der ocu

ACT - Change the process to better

STUDY
gh

understand common cause

Or is d

variation.
Th

- Reduce common cause

variation.

Figure I.4: The Process Improvement Cycle

24
 CHAPTER I – Section F
The Process Improvement Cycle and Process Control

CHAPTER I – Section F
The Process Improvement Cycle and Process Control
In applying the concept of continual improvement to processes, there is a
three-stage cycle that can be useful (see Figure I.4). Every process is in
one of the three stages of the Improvement Cycle.

1. Analyze the Process

A basic understanding of the process is a must when considering process
improvement. Among the questions to be answered in order to achieve a

din p
better understanding of the process are:

.
or ou
gly
cc Gr
• What should the process be doing?

Inc
What is expected at each step of the process?

d a ion
What are the operational definitions of the deliverables?

rke Act
al
on
• What can go wrong?

ma ry
ati
What can vary in this process?

ter ust
en e 11
rn

What do we already know about this process’ variability?

be otiv /20
wa Ind
nte

What parameters are most sensitive to variation?

1
aI

•
3/3

What is the process doing?

gn

Is this process producing scrap or output that requires rework?

an e A n:
Ma

as m

Does this process produce output that is in a state of statistical

o
o
i

control?
d h ut
t
to:

ted y t ira

Is the process capable?

xp
op en ite 426 sed

Is the process reliable?

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

Many techniques discussed in the APQP Manual7 may be applied to gain

o

a better understanding of the process. These activities include:

s c um S
E- der ocu

gh

• Group meetings
Or is d

• Consultation with people who develop or operate the process

Th

(“subject matter experts”)

• Review of the process’ history
• Construction of a Failure Modes and Effects Analysis (FMEA)
Control charts explained in this manual are powerful tools that should be
used during the Process Improvement Cycle. These simple statistical
methods help differentiate between common and special causes of
variation. The special causes of variation must be addressed. When a
state of statistical control has been reached, the process’ current level of
long-term capability can be assessed (see Chapter IV).

7
Chrysler, Ford, and General Motors, (1995).

25
CHAPTER I – Section F
The Process Improvement Cycle and Process Control

2. Maintain (Control) the Process

Once a better understanding of the process has been achieved, the
process must be maintained at an appropriate level of capability.
Processes are dynamic and will change. The performance of the process
should be monitored so effective measures to prevent undesirable change
can be taken. Desirable change also should be understood and
institutionalized. Again, the simple statistical methods explained in this
manual can assist. Construction and use of control charts and other tools
will allow for efficient monitoring of the process. When the tool signals
that the process has changed, quick and efficient measures can be taken
to isolate the cause(s) and act upon them.
It is too easy to stop at this stage of the Process Improvement Cycle. It is
important to realize that there is a limit to any company’s resources.

din p
.
or ou
gly
Some, perhaps many, processes should be at this stage. However, failure

cc Gr
to proceed to the next stage in this cycle can result in a significant
competitive disadvantage. The attainment of “world class” requires a

Inc

d a ion
steady and planned effort to move into the next stage of the Cycle.

rke Act
al
on

ma ry
ati
3. Improve the Process

ter ust
en e 11
rn

Up to this point, the effort has been to stabilize the processes and
be otiv /20
wa Ind
nte

maintain them. However, for some processes, the customer will be

1
aI

3/3

sensitive even to variation within engineering specifications (see Chapter

gn

IV). In these instances, the value of continual improvement will not be

an e A n:
Ma

as m

realized until variation is reduced. At this point, additional process

o
o
i

analysis tools, including more advanced statistical methods such as

d h ut
t
to:

ted y t ira

designed experiments and advanced control charts may be useful.

xp
op en ite 426 sed

Appendix H lists some helpful references for further study.

h
ro ed E
n

t p wn se

Process improvement through variation reduction typically involves

yri t is Lic 2
d i oc nt r: 4 lice

purposefully introducing changes into the process and measuring the

en
an is d ume be t is

effects. The goal is a better understanding of the process, so that the

5

tec

common cause variation can be further reduced. The intent of this

Th Doc Num men

reduction is improved quality at lower cost.

s c um S
E- der ocu

When new process parameters have been determined, the Cycle shifts
gh

back to Analyze the Process. Since changes have been made, process
Or is d

stability will need to be reconfirmed. The process then continues to

Th

move around the Process Improvement Cycle.

26
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

27
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
This page intentionally left blank

wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
The Process Improvement Cycle and Process Control
CHAPTER I – Section F
CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

CONTROL CHARTS

din p
Upper Control Limit

.
or ou
gly
cc Gr
Inc

d a ion
Center Line

rke Act
al
on

ma ry
ati

ter ust
Lower Control Limit

en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

1. Collection
xp
op en ite 426 sed

• Gather Data and plot on a chart.

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

2. Control
b
en

• Calculate trial control limits from process data.

an is d ume be t is
5

tec
Th Doc Num men

• Identify special causes of variation and act upon them.

o
s c um S

3. Analysis and Improvement

E- der ocu

gh

• Quantify common cause variation; take action to reduce it.

Or is d
Th

These three phases are repeated for continual process improvement

Figure I.5: Control Charts

28
 CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

CHAPTER I – Section G
Control Charts:
Tools For Process Control and Improvement
In his books8, Dr. W. E. Deming identifies two mistakes frequently made
in process control:
“Mistake 1. Ascribe a variation or a mistake to a special cause,
when in fact the cause belongs to the system (common causes).
Mistake 2. Ascribe a variation or a mistake to a system (common
causes), when in fact the cause was special.

din p
.
or ou
gly
Over adjustment [tampering] is a common example of mistake

cc Gr
No. 1. Never doing anything to try to find a special cause is a

Inc

d a ion
common example of mistake No.2.”

rke Act
al
on
For effective variation management during production, there must be an

ma ry
ati
effective means of detecting special causes. There is a common

ter ust
en e 11
rn

misconception that histograms can be used for this purpose. Histograms

be otiv /20
wa Ind
nte

are the graphical representation of the distributional form of the process

1
aI

variation. The distributional form is studied to verify that the process

3/3
gn

variation is symmetric and unimodal and that it follows a normal

distribution.
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Unfortunately normality does not guarantee that there are no special

causes acting on the process. That is, some special causes may change
xp
op en ite 426 sed

the process without destroying its symmetry or unimodality. Also a non-

ro ed E
n

normal distribution may have no special causes acting upon it but its
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

distributional form is non-symmetric.

en
an is d ume be t is

Time-based statistical and probabilistic methods do provide necessary

5

tec

and sufficient methods of determining if special causes exist. Although

Th Doc Num men

several classes of methods are useful in this task, the most versatile and
s c um S
E- der ocu

robust is the genre of control charts which were first developed and
implemented by Dr. Walter Shewhart of the Bell Laboratories9 while
gh
Or is d

studying process data in the 1920’s. He first made the distinction

between controlled and uncontrolled variation due to what is called
Th

common and special causes. He developed a simple but powerful tool to

separate the two – the control chart. Since that time, control charts have
been used successfully in a wide variety of process control and
improvement situations. Experience has shown that control charts
effectively direct attention toward special causes of variation when they
occur and reflect the extent of common cause variation that must be
reduced by system or process improvement.
It is impossible to reduce the above mistakes to zero. Dr. Shewhart
realized this and developed a graphical approach to minimize, over the
long run, the economic loss from both mistakes.

8
Deming (1989) and Deming (1994).
9
Shewhart (1931).

29
CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

If process control activities assure that no special cause sources of

variation are active10, the process is said to be in statistical control or “in
control.” Such processes are said to be stable, predictable, and consistent
since it is possible to predict11 the performance of the process.
The active existence of any special cause will render the process out of
statistical control or “out of control.” The performance of such unstable
processes cannot be predicted.

How do they work?

Control Limits

When Shewhart developed control charts he was concerned with the

din p
.
or ou
gly
economic control of processes; i.e., action is taken on the process only

cc Gr
when special causes are present. To do this, sample statistics are
compared to control limits. But how are these limits determined?

Inc

d a ion
Consider a process distribution that can be described by the normal form.

rke Act
al
on
The goal is to determine when special causes are affecting it. Another

ma ry
way of saying this is, “Has the process changed since it was last looked
ati

ter ust
en e 11
at it or during the period sampled?”
rn

be otiv /20
wa Ind
nte

Shewhart's Two Rules for the Presentation of Data:

1
aI

3/3

Data should always be presented in such a way that preserves the

gn

evidence in the data for all the predictions that might be made from these
an e A n:
Ma

as m

data.
o
o

Whenever an average, range, or histogram is used to summarize data,

i
d h ut
t
to:

ted y t ira

the summary should not mislead the user into taking any action that the
xp
op en ite 426 sed

user would not take if the data were presented in a time series.
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Since the normal distribution is described by its process location (mean)

and process width (range or standard deviation) this question becomes:
Has the process location or process width changed?
Consider only the location. What approach can be used to determine if
the process location has changed? One possibility would be to look at
10
This is done by using the process information to identify and eliminate the existence of special causes
or detecting them and removing their effect when they do occur.
11
As with all probabilistic methods some risk is involved. The exact level of belief in prediction of
future actions cannot be determined by statistical measures alone. Subject-matter expertise is
required.

30
 CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

every part produced by the process, but that is usually not economical.
The alternative is to use a sample of the process, and calculate the mean
of the sample.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
If the process has not changed, will the sample mean be equal to the
on
distribution mean?

ma ry
ati

ter ust
en e 11
The answer is that this very rarely happens. But how is this possible?
rn

be otiv /20
After all, the process has not changed. Doesn’t that imply that the
wa Ind
nte

process mean remains the same? The reason for this is that the sample
1
aI

3/3

mean is only an estimation of the process mean.

gn

To make this a little clearer, consider taking a sample of size one. The
an e A n:
Ma

as m
o

mean of the sample is the individual sample itself. With such random
o
i
d h ut
t
to:

ted y t ira

samples from the distribution, the readings will eventually cover the
xp

entire process range. Using the formula:

op en ite 426 sed

( )
ro ed E
n

t p wn se

1
yri t is Lic 2
d i oc nt r: 4 lice

Range of the distribution of means = Process Range

b
en

n
an is d ume be t is
5

for a sample of size four, the resulting range of sample averages will be
tec
Th Doc Num men

1 4 = 1 of the process range; for a sample of size 100 it will be

o

2
s c um S

of the process range.12

E- der ocu

1 100 = 1
10
gh
Or is d
Th

Shewhart used this sampling distribution to establish an operational

definition of “in statistical control.” First, start off with the assumption
that the process is in statistical control, i.e., innocent until proven guilty.
Then, compare the sample to the sampling distribution using the ±3
standard deviation limits13. These are called control limits. If the sample
falls outside these limits then there is reason to believe that a special
cause is present. Further, it is expected that all the (random) samples will
exhibit a random ordering within these limits. If a group of samples
shows a pattern there is reason to believe that a special cause is present.
(see Chapter I, Section C, and Chapter II, Section A).

12
See the Central Limit Theorem.
13
Shewhart selected the ±3 standard deviation limits as useful limits in achieving the economic control of
processes.

31
CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

Distribution of Averages Distribution of Individuals

take n samples

din p
calculate x

.
or ou
gly
cc Gr
LCL UCL

Inc

d a ion
due to Sampling Variation

rke Act
al
on
In general, to set up a control chart we calculate:

ma ry
ati
Centerline = average of the statistic being analyzed

ter ust
en e 11
rn

UCL = upper control limit = centerline + 3 x standard deviation of the averages

be otiv /20
wa Ind
nte

1
LCL = lower control limit = centerline - 3 x standard deviation of the averages
aI

3/3
gn

an e A n:
Ma

as m

Approach:
o
o
i
d h ut
t
to:

ted y t ira
xp

Since Control Charts provide the operational definition of “in statistical

op en ite 426 sed

control,” they are useful tools at every stage of the Improvement Cycle
ro ed E
n

(see Chapter I, Section F). Within each stage, the PDSA14 cycle should
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

be used.
an is d ume be t is
5

tec
Th Doc Num men

For analysis of existing data sets

o
s c um S
E- der ocu

For the Analysis and Improvement stages of the cycle:

gh
Or is d

• Review the data:

Is the metric appropriate; i.e., does it reflect a process attribute
Th

and tied to a key business factor?

Are the data consistent; i.e., is the same operational definition
used by all parties collecting the data?
Are the data reliable; i.e., is a planned data collection scheme
utilized?
Is the measurement system appropriate and acceptable?
• Plot the data:
Plot using the time order
Compare to control limits and determine if there are any points
outside the control limits

14
Plan-Do-Study-Act cycle; also known as the PDCA, (Plan-Do-Check-Act) cycle.

32
 CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

Compare to the centerline and determine if there are any non-

random patterns clearly discernible
• Analyze the data
• Take appropriate action
The data are compared with the control limits to see whether the
variation is stable and appears to come from only common causes. If
special causes of variation are evident, the process is studied to further
determine what is affecting it. After actions (see Chapter I, Section D)
have been taken, further data are collected, control limits are recalculated
if necessary, and any additional special causes are acted upon.
After all special causes have been addressed and the process is running in
statistical control, the control chart continues as a monitoring tool.

din p
.
or ou
gly
Process capability can also be calculated. If the variation from common

cc Gr
causes is excessive, the process cannot produce output that consistently
meets customer requirements. The process itself must be investigated,

Inc

d a ion
and, typically, management action must be taken to improve the system.

rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

For control be otiv /20

wa Ind
nte

• Review the data collection scheme before starting:

1
aI

3/3

Is the metric appropriate; i.e., does it reflect a process attribute

gn

and tied to a key business factor?

an e A n:
Ma

as m

Will the data be consistent; i.e., is the same operational

o
o
i
d h ut
t

definition used by all parties collecting the data?

to:

ted y t ira

Will the data be reliable; i.e., is a planned data collection scheme

xp
op en ite 426 sed

used?
h
ro ed E
n

Is the measurement system appropriate and acceptable?

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

• Plot each point as it is determined:

an is d ume be t is
5

Compare to control limits and determine if there are any points

tec
Th Doc Num men

outside the control limits

o

Compare to the centerline and determine if there are any non-

s c um S
E- der ocu

random patterns clearly discernible

gh
Or is d

• Analyze the data

Th

• Take appropriate action:

Continue to run with no action taken; or
Identify source of the special cause and remove (if unacceptable
response) or reinforce (if acceptable response); or
Continue to run with no action taken and reduce sample size or
frequency; or
Initiate a continual improvement action

Often it is found that although the process was aimed at the target
value during initial setup, the actual process location ( µ )15 may not

15
The Greek letter µ is used to indicate the actual process mean, which is estimated by the sample
mean X .

33
CHAPTER I – Section G
Control Charts: Tools For Process Control and Improvement

match this value. For those processes where the actual location
deviates from the target and the ability to relocate the process is
economical, consideration should be given to adjusting the process
so that it is aligned with the target (see Chapter IV, Section C). This
assumes that this adjustment does not affect the process variation.
This may not always hold true, but the causes for any possible
increase in process variation after re-targeting the process should be
understood and assessed against both customer satisfaction and
economics.
The long-term performance of the process should continue to be
analyzed. This can be accomplished by a periodic and systematic
review of the ongoing control charts. New evidence of special
causes might be revealed. Some special causes, when understood,
will be beneficial and useful for process improvement Others will

din p
.
or ou
gly
be detrimental, and will need to be corrected or removed.

cc Gr
Inc

d a ion
The purpose of the Improvement Cycle is to gain an understanding of the

rke Act
al
process and its variability to improve its performance. As this

on
understanding matures, the need for continual monitoring of product

ma ry
variables may become less – especially in processes where
ati

ter ust
en e 11
documented analysis shows that the dominant source of variation are
rn

more efficiently and effectively controlled by other approaches. For

be otiv /20
wa Ind
nte

example: in processes where maintenance is the dominant source of

1
aI

variation, the process is best controlled by preventive and predictive

3/3
gn

maintenance; for processes where process setup is the dominant source

an e A n:
Ma

of variation, the process is best controlled by setup control charts.

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

For a process that is in statistical control, improvement efforts will often

h
ro ed E

focus on reducing the common cause variation in the process. Reducing

n

t p wn se

this variation will have the effect of “shrinking” the control limits on the
yri t is Lic 2
d i oc nt r: 4 lice

b
en

control chart (i.e., the limits, upon their recalculation, will be closer
an is d ume be t is

together). Many people, not familiar with control charts, feel this is
5

tec

“penalizing” the process for improving. They do not realize that if a

Th Doc Num men

process is stable and the control limits are calculated correctly, the
s c um S
E- der ocu

chance that the process will erroneously yield an out-of-control point is

gh

the same regardless of the distance between the control limits (see
Or is d

Chapter I, Section E).

Th

One area deserving mention is the question of recalculation of control

chart limits. Once properly computed, and if no changes to the common
cause variation of the process occur, then the control limits remain
legitimate. Signals of special causes of variation do not require the
recalculation of control limits. For long-term analysis of control charts, it
is best to recalculate control limits as infrequently as possible; only as
dictated by changes in the process.
For continual process improvement, repeat the three stages of the
Improvement Cycle: Analyze the Process; Maintain (Control) the
Process; Improve the Process, see Figure I.4.

34
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

35
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
This page intentionally left blank

wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Control Charts: Tools For Process Control and Improvement
CHAPTER I – Section G
CHAPTER I – Section H
Effective Use and Benefits Of Control Charts

BENEFITS OF CONTROL CHARTS

Properly used, control charts can:
• Be used by operators for ongoing control of a process
• Help the process perform consistently and predictably
• Allow the process to achieve
– Higher quality
– Lower unit cost

din p
.
or ou
gly
– Higher effective capability

cc Gr
• Provide a common language for discussing the performance of the

Inc

d a ion
process

rke Act
• Distinguish special from common causes of variation, as a guide to

al
on
local action or action on the system.

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

36
 CHAPTER I – Section H
Effective Use and Benefits Of Control Charts

CHAPTER I – Section H
Effective Use and Benefits of Control Charts
Important benefits can be obtained from the effective use of control
charts. The gains and benefits from the control charts are directly related
to the following:
Management Philosophy: How the company is managed can directly
impact the effectiveness of SPC.
The following are examples of what needs to be present:
• Focus the organization on variation reduction.

din p
.
or ou
gly
• Establish an open environment that minimizes internal competition

cc Gr
and supports cross-functional teamwork.

Inc

d a ion
• Support and fund management and employee training in the proper

rke Act
al
use and application of SPC.
on

ma ry
ati
• Show support and interest in the application and resulting benefits of

ter ust
en e 11
rn

properly applied SPC. Make regular visits and asks questions in

be otiv /20
wa Ind
nte

those areas.
1
aI

• Apply SPC to promote the understanding of variation in engineering

3/3
gn

processes.
an e A n:
Ma

as m
o

• Apply SPC to management data and use the information in day-to-

o
i
d h ut
t
to:

ted y t ira

day decision making.

xp
op en ite 426 sed

The above items support the requirements contained in

ro ed E
n

ISO 9000:2000 and ISO/TS 16949:2002.

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec

Engineering Philosophy: How engineering uses data to develop designs

Th Doc Num men

can and will have an influence on the level and type of variation in the
s c um S

finished product.
E- der ocu

gh

The following are some ways that engineering can show effective use of
Or is d

SPC:
Th

• Focus the engineering organization on variation reduction throughout

the design process; e.g., number of design changes, design for
manufacturing and assembly, personnel moves, etc.
• Establish an open engineering environment that minimizes internal
competition and supports cross-functional teamwork.
• Support and fund engineering management and employees training
in the proper use and application of SPC.
• Apply SPC to promote the understanding of variation in engineering
processes.
• Require an understanding of variation and stability in relation to
measurement and the data that are used for design development.

37
CHAPTER I – Section H
Effective Use and Benefits Of Control Charts

• Support engineering changes proposed due to analysis of SPC

information to aid in the reduction of variation.
Manufacturing: How manufacturing develops and operates machines
and transfer systems can impact the level and type of variation in the
finished product:
• Focus the manufacturing organization on variation reduction; e.g.,
number of different processes, impact of multi-fixture and multi-tool
processes, tool and machine maintenance, etc.
• Establish an open engineering environment that minimizes internal
competition and supports cross-functional teamwork.
• Support and fund manufacturing management and employees
training in the proper use and application of SPC.

din p
.
or ou
gly
• Apply SPC in the understanding of variation in the manufacturing

cc Gr
processes.

Inc

d a ion
• Require an understanding of variation and stability in relation to

rke Act
al
measurement and the data that are used for process design
on

ma ry
development.
ati

ter ust
en e 11
rn

• Use the analysis of SPC information to support process changes for

be otiv /20
wa Ind
nte

the reduction of variation.

1
aI

3/3

• Do not release control charts to operators until the process is stable.

gn

The transfer of responsibility for the process to production should

an e A n:
Ma

as m

occur after the process is stable.

o
o
i
d h ut
t
to:

ted y t ira

• Assure proper placement of SPC data for optimum use by the

xp
op en ite 426 sed

employees.
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Quality Control: The Quality function is a critical component in

b
en

providing support for an effective SPC process:

an is d ume be t is
5

tec

• Support SPC training for management, engineering, and employees

Th Doc Num men

in the organization.
s c um S
E- der ocu

• Mentor key people in the organization in the proper application of

gh

SPC.
Or is d

• Assist in the identification and reduction of the sources of variation.

Th

• Ensure optimum use of SPC data and information.

Production: Production personnel are directly related to the process and

can affect process variation. They should:
• Be properly trained in the application of SPC and problem solving.
• Have an understanding of variation and stability in relation to
measurement and the data that are used for process control and
improvement.
• Be alert to and communicate when conditions change.
• Update, maintain and display control charts within area of
responsibility.

38
 CHAPTER I – Section H
Effective Use and Benefits Of Control Charts

• Interact and learn about the process from the information collected.
• Use the SPC information in real time to run the process.

Application of the concepts outlined above will result in the proper

environment for the understanding and reduction of variation. Then the
Plan-Do-Study-Act process can be used to further improve the process.
At a minimum, the use of SPC for process monitoring will result in the
process being maintained at its current performance level. However, real
improvements can be achieved when SPC is used to direct the way
processes are analyzed.
Proper use of SPC can result in an organization focused on improving the

din p
quality of the product and process.

.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

39
 CHAPTER I – Section H

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
Effective Use and Benefits Of Control Charts

s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

40
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

41
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
CHAPTER II
Control Charts
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

42
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER II
Control Charts

Introduction:

Control charts can be used to monitor or evaluate a process. There are

basically two types of control charts, those for variables data and those
for attributes data. The process itself will dictate which type of control
chart to use. If the data derived from the process are of a discrete nature
(e.g., go/no-go, acceptable/not acceptable) then an attributes type of chart
would be used. If the data derived from the process are of a continuous
nature (e.g., diameter, length) then a variables type of chart would be
used. Within each chart type there are several chart combinations that
can be used to further evaluate the process.

din p
.
Some of the more common chart types, Average ( X ) and Range ( R )

or ou
gly
charts, Individuals ( I ) chart, Moving Range ( MR ) chart, etc., belong to

cc Gr
the variables chart family. Charts based on count or percent data (e.g., p,

Inc

d a ion
np, c, u) belong to the attributes chart family.

rke Act
al
When introducing control charts into an organization, it is important to
on

ma ry
prioritize problem areas and use charts where they are most needed.
ati

ter ust
Problem signals can come from the cost control system, user complaints,
en e 11
rn

internal bottlenecks, etc. The use of attributes control charts on key

be otiv /20
wa Ind
nte

overall quality measures often points the way to the specific process
1
aI

3/3

areas that would need more detailed examination including the possible
gn

use of control charts for variables.

an e A n:
Ma

as m
o

If available, variables data are always preferred as they contain more

o
i
d h ut
t
to:

ted y t ira

useful information than attributes data for the same amount of effort. For
example you need a larger sample size for attributes than for variables
xp
op en ite 426 sed

data to have the same amount of confidence in the results. If the use of
ro ed E
n

variables measurement systems is infeasible, the application of attributes

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

analysis should not be overlooked.

en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

43
CHAPTER II
Control Charts

CONTROL CHARTS TO ANALYZE THE PROCESS

Process
Process evaluation requires Measurements
People Equipment Environment

1 2 3 4 5 6 7 8

The outcome is a Decision

Materials Methods Measurement
based on the Measurements

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on
Outcome Example Control Chart Examples

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

• Shaft O.D. (inches) XX for the Average of

1
aI

the Measurement
3/3

• Hole distance from reference surface (mm)

gn

• Circuit resistance (ohms)

an e A n:
Ma

as m

• Railcar transit time (hours)

o
o

R Chart for the Ranges

i
d h ut
t
to:

• Engineering change processing time (hours)

ted y t ira

of the Measurement
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

The measurement method must produce accurate and precise results over time
an is d ume be t is
5

tec
Th Doc Num men

Not Precise Precise

o
s c um S
E- der ocu

gh
Or is d

Not Accurate
Th

Accurate*

*Note: Some current metrology literature defines accuracy as the lack of bias.

Figure II.1: Variables Data

44
 CHAPTER II
Control Charts

Variables Control Charts

Variables control charts represent the typical application of statistical
process control where the processes and their outputs can be
characterized by variable measurements (see Figure II.1).
Variables control charts are particularly useful for several reasons:
• A quantitative value (e.g., “the diameter is 16.45 mm”) contains
more information than a simple yes-no statement (e.g., “the diameter
is within specification”);
• Although collecting variables data is usually more costly than
collecting attributes data (e.g., go/no-go), a decision can be reached
more quickly with a smaller sample size. This can lead to lower total

din p
.
or ou
gly
measurement costs due to increased efficiency;

cc Gr
• Because fewer parts need to be checked before making reliable

Inc

d a ion
decisions, the time delay between an “out-of-control” signal and

rke Act
al
corrective action is usually shorter; and
on

ma ry
• With variables data, performance of a process can be analyzed, and
ati

ter ust
en e 11
improvement can be quantified, even if all individual values are
rn

be otiv /20
wa Ind
within the specification limits. This is important in seeking continual
nte

improvement.
1
aI

3/3
gn

A variables chart can explain process data in terms of its process

variation, piece-to-piece variation, and its process average. Because of
an e A n:
Ma

as m
o
o

this, control charts for variables are usually prepared and analyzed in
i
d h ut
t
to:

ted y t ira

pairs, one chart for process average and another for the process variation.
xp
op en ite 426 sed

The most commonly used pair are the X and R charts. X is the
h
ro ed E

arithmetic average of the values in small subgroups – a measure of

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

process average; R is the range of values within each subgroup (highest

b
en

minus lowest) – a measure of process variation. However, there are a

an is d ume be t is
5

number of other control charts that may be more useful under certain
tec
Th Doc Num men

circumstances.
o
s c um S

The X and R charts may be the most common charts, but they may not
E- der ocu

gh

be the most appropriate for all situations.

Or is d
Th

45
CHAPTER II
Control Charts

CONTROL CHARTS TO CLASSIFY THE PROCESS

Process

Decision is based on the

People Equipment Environment Classification of the Outcome

Materials Methods Measurement

din p
.
or ou
gly
cc Gr
Outcome Example Control Chart Examples

Inc

d a ion
Vehicle does not leak p Chart for Proportion of

rke Act
al
on
Units Nonconforming
Lamp lights does not light

ma ry
ati

ter ust
en e 11
rn

Hole diameter undersized or oversized np Chart for Number of

be otiv /20
wa Ind
nte

(evaluated using a go/nogo gage) Units Nonconforming

1
aI

3/3

Shipment to dealer correct or incorrect

gn

an e A n:
Ma

as m

Bubbles in a windshield c Chart for Number of

o
o

Nonconformances per Unit

i
d h ut
t
to:

ted y t ira

Paint imperfections on door u Chart for Number of

xp
op en ite 426 sed

Errors on an invoice Nonconformities per Unit

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

The conformance criteria must be clearly defined and the procedures for deciding if
en

those criteria are met must produce consistent results over time.
an is d ume be t is
5

tec
Th Doc Num men

Acceptance Criteria Examples Comment

s c um S
E- der ocu

gh

Surface should be free from flaws What is a flaw?

Or is d

Surface should conform to master standard Conform to what degree?

Th

in color, texture, brightness Do inspectors agree?

and have not imperfections How is it measured?

Any material applied to mirror back to shall · Visible to whom?

not cause visible staining of the backing Under what conditions

Figure II.2: Attributes Data

46
 CHAPTER II
Control Charts

Attributes Control Charts

Although control charts are most often thought of in terms of variables,
control charts have also been developed for attributes; see Figure II.2.
Attributes data have discrete values and they can be counted for
recording and analysis. With attribute analysis the data are separated into
distinct categories (conforming/nonconforming, pass/fail, go/no-go,
present/absent, low/medium/high). Examples include the presence of a
required label, the continuity of an electrical circuit, visual analysis of a
painted surface, or errors in a typed document.
Other examples are of characteristics that are measurable, but where the
results are recorded in a simple yes/no fashion, such as the conformance
of a shaft diameter when measured on a go/no-go gage, the acceptability

din p
.
or ou
gly
of door margins to a visual or gage check, or on-time delivery

cc Gr
performance. Control charts for attributes are important for several
reasons:

Inc

d a ion
• Attributes data situations exist in any technical or administrative

rke Act
al
process, so attributes analysis techniques are useful in many
on

ma ry
applications. The most significant difficulty is to develop precise
ati

ter ust
en e 11
operational definitions of what is conforming.
rn

be otiv /20
wa Ind
nte

• Attributes data are already available in many situations – wherever

there are existing inspections, repair logs, sorts of rejected material,
1
aI

3/3

etc. In these cases, no additional effort is required for data collection.

gn

The only expense involved is for the effort of converting the data to
an e A n:
Ma

as m

control chart form.

o
o
i
d h ut
t
to:

ted y t ira

• Where new data must be collected, attributes information is

xp

generally quick and inexpensive to obtain. With simple gaging (e.g.,

op en ite 426 sed

h
ro ed E

a go/no-go gage or visual standards), specialized measurement skills

n

t p wn se

are often are not required. There are many occasions where
yri t is Lic 2
d i oc nt r: 4 lice

b
en

specialized measurement skills are required especially when the part

an is d ume be t is

measured falls in the “gray” area.16

5

tec
Th Doc Num men

• Much data gathered for management summary reporting are often in

o

attributes form and can benefit from control chart analysis. Examples
s c um S
E- der ocu

include scrap rates, quality audits and material rejections. Because of

gh

the ability to distinguish between special and common cause

Or is d

variation, control chart analysis can be valuable in interpreting these

Th

management reports.
This manual will use conforming/nonconforming throughout attributes
discussions simply because
• These categories are “traditionally” used
• Organizations just starting on the path to continual improvement
usually begin with these categories
• Many of the examples available in literature use these categories.
It should not be inferred that these are the only “acceptable” categories or
that attributes charts cannot be used with Case 1 processes; see Chapter I,
Section E.17

16
See the Attribute Measurement System Study chapter in the MSA Reference Manual.
17
See also: Montgomery (1997), Wheeler (1991, 1995), Wise and Fair (1998).

47
 CHAPTER II
Control Charts

Elements of Control Charts

ANALYZE MAINTAIN
There is no single “approved” manner of displaying control charts.
PLAN DO PLAN DO
However the reasons for the use of control charts (see Chapter I, Section
ACT STUDY ACT STUDY
E) must be kept in mind. Any format is acceptable as long as it contains
the following (see Figure II.3):
PLAN DO

ACT STUDY
• (A) Appropriate scale
IMPROVE
The scale should be such that the natural variation of the process can
be easily viewed. A scale which yields a “narrow” control chart does
not enable analysis and control of the process.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o

•
o

(B) UCL, LCL

i
d h ut
t
to:

ted y t ira

The ability to determine outliers which signal special causes the

xp
op en ite 426 sed

control chart requires control limits based on the sampling

h
ro ed E

distribution. Specifications limits should not be used in place of valid

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

control limits for process analysis and control.

en
an is d ume be t is

• (B) Centerline
5

tec
Th Doc Num men

The control chart requires a centerline based on the sampling

s c um S

distribution in order to allow the determination of non-random

E- der ocu

patterns which signal special causes.

gh
Or is d

• (C) Subgroup sequence / timeline

Th

Maintaining the sequence in which the data are collected provides

indications of “when” a special cause occurs and whether that special
cause is time-oriented.
• (D) Identification of out-of-control plotted values
Plotted points which are out of statistical control should be identified
on the control chart. For process control the analysis for special
causes and their identification should occur as each sample is plotted
as well as periodic reviews of the control chart as a whole for non-
random patterns.
• (E) Event Log
Besides the collection, charting, and analysis of data, additional
supporting information should be collected. This information should

48
 CHAPTER II
Control Charts

include any potential sources of variation as well as any actions

taken to resolve out-of-control signals (OCS). This information can
be recorded on the control chart or on a separate Event Log.

If there has not been any change in the process between subgroups, it is
not necessary to include an entry on the process event log.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d

Figure II.3: Elements of Control Charts

Th

During the initial analysis of the process, knowledge of what would

constitute a potential special cause for this specific process may be
incomplete. Consequently, the initial information collection activities
may include events which will prove out not to be special causes. Such
events need not be identified in subsequent information collection
activities. If initial information collection activities are not sufficiently
comprehensive, then time may be wasted in identifying specific events
which cause out-of-control signals.

49
CHAPTER II
Control Charts

For control charts which are included as a part of a report and for those
which are maintained manually the following “header” information
should be included:
• What: part/product/service name and number/identification
• Where: operation/process step information, name/identification
• Who: operator and appraiser
• How: measurement system used, name/number, units (scale)
• How many: subgroup size, uniform or by sample
•

din p
When: sampling scheme (frequency and time)

.
or ou
gly
cc Gr
Inc

d a ion
Figure II.4 shows a completed manually maintained control chart which
includes all these elements

rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

50
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

51
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou

Figure II.4a: Sample Control Chart (Front side)

din p
gly
.
CHAPTER II
Control Charts
 CHAPTER II
Control Charts

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

52
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p

Note: Sx indicates the shift; n/c indicates no changes in the process.

gly
.

Figure II.4b: Sample Control Chart (back side) – Event Log

 CHAPTER II - Section A
Control Charting Process

CHAPTER II - Section A
Control Chart Process
Preparatory Steps
Before control charts can be used, several preparatory steps should be
taken:

Establish an environment suitable for action.

Define the process.

din p
.
or ou
gly
Determine the features or characteristics to be charted based on:

cc Gr
Inc

d a ion
The customer’s needs.

rke Act
al
Current and potential problem areas.
on

ma ry
Correlation between characteristics.
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Correlation between variables does not imply a causal relationship. In

Caution
1
aI

the absence of process knowledge, a designed experiment may be

3/3

needed to verify such relationships and their significance.

gn

an e A n:
Ma

as m
o
o

Define the characteristic.

i
d h ut
t
to:

ted y t ira

The characteristic must be operationally defined so that results

xp
op en ite 426 sed

can be communicated to all concerned in ways that have the

h
ro ed E
n

same meaning today as yesterday. This involves specifying what

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

information is to be gathered, where, how, and under what

en

conditions.
an is d ume be t is
5

tec
Th Doc Num men

An operational definition describes the characteristic that is to be

o

evaluated and whether the characteristic is qualitative (discrete)

s c um S
E- der ocu

or quantitative (continuous). Attributes control charts would be

gh

used to monitor and evaluate discrete variables whereas variables

Or is d

control charts would be used to monitor and evaluate continuous

Th

variables.

Define the measurement system.

Total process variability consists of part-to-part variability and
measurement system variability. It is very important to evaluate
the effect of the measurement system's variability on the overall
process variability and determine whether it is acceptable. The
measurement performance must be predictable in terms of
accuracy, precision and stability.
Periodic calibration is not enough to validate the measurement
system's capability for its intended use. In addition to being
calibrated, the measurement system must be evaluated in terms
of its suitability for the intended use.

53
CHAPTER II - Section A
Control Charting Process

For more detail on this subject see the Measurement Systems

Analysis (MSA) Reference Manual. The definition of the
measurement system will determine what type of chart, variables
or attributes, is appropriate.

Minimize unnecessary variation.

Unnecessary external causes of variation should be reduced
before the study begins. This could simply mean watching that
the process is being operated as intended. The purpose is to
avoid obvious problems that could and should be corrected
without use of control charts. This includes process adjustment
or over control. In all cases, a process event log may be kept
noting all relevant events such as tool changes, new raw material
lots, measurement system changes, etc. This will aid in

din p
.
or ou
gly
subsequent process analysis.

cc Gr
Inc
Assure selection scheme is appropriate for detecting expected

d a ion
special causes.

rke Act
al
on

ma ry
WARNING: Even though convenience sampling and/or haphazard
ati

ter ust
sampling is often thought of as being random sampling, it is not. If one

en e 11
rn

assumes that it is, and in reality it is not, one carries an unnecessary risk
be otiv /20
wa Ind
nte

that may lead to erroneous and or biased conclusions.

1
aI

3/3
gn

For more details see Chapter I, Section H.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

54
 CHAPTER II - Section A
Control Charting Process

Control Chart Mechanics

The steps to using control charts are:

1. Data Collection
2. Establish Control Limits
3. Interpret for Statistical Control
4. Extend Control Limits for ongoing control (see Figure II.5)

Data Collection

din p
.
or ou
gly
Control charts are developed from measurements of a particular

cc Gr
characteristic or feature of the process. These measurements are
combined into a (control) statistic (e.g., average, median, range, standard

Inc

d a ion
deviation, individual) which describes an attribute of the process

rke Act
al
distributional form. The measurement data are collected from individual
on
samples from a process stream. The samples are collected in subgroups

ma ry
ati
and may consist of one or more pieces. In general, a larger subgroup size

ter ust
en e 11
rn

makes it easier to detect small process shifts.

be otiv /20
wa Ind
nte

1
aI

3/3

Create a Sampling Plan

gn

an e A n:
Ma

as m
o

For control charts to be effective the sampling plan should define

o
i
d h ut
t
to:

rational subgroups. A rational subgroup is one in which the samples are

ted y t ira

selected so that the chance for variation due to special causes occurring
xp
op en ite 426 sed

within a subgroup is minimized, while the chance for special cause

ro ed E
n

variation between subgroups is maximized. The key item to remember

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

when developing a sampling plan is that the variation between subgroups

en

is going to be compared to the variation within subgroups. Taking

an is d ume be t is
5

tec

consecutive samples for the subgroups minimizes the opportunity for the
Th Doc Num men

process to change and should minimize the within-subgroup variation.

s c um S

The sampling frequency will determine the opportunity the process has
E- der ocu

to change between subgroups.

gh
Or is d

The variation within a subgroup represents the piece-to-piece variation

Th

over a short period of time.18 Any significant variation between

subgroups would reflect changes in the process that should be
investigated for appropriate action.
Subgroup Size – The type of process under investigation dictates how the
subgroup size is defined. As stated earlier, a larger subgroup size makes
it easier to detect small process shifts. The team responsible has to
determine the appropriate subgroup size. If the expected shift is
relatively small, then a larger subgroup size would be needed compared
to that required if the anticipated shift is large.

18
See also Appendix A.

55
 Control Charting Process
CHAPTER II - Section A

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

56
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act

Figure II.5: Extending Control Limits

d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER II - Section A
Control Charting Process

The subgroup size should remain constant but there may be situations
where the subgroup size varies within a single control chart. The
calculation of the control limits depends on the subgroup size and if one
varies the subgroup size the control limits will change for that subgroup.
There are other techniques that deal with variable subgroup sizes; for
example, see Montgomery (1997) and Grant and Leavenworth (1996).
Subgroup Frequency – The subgroups are taken sequentially in time,
e.g., once every 15 minutes or twice per shift. The goal is to detect
changes in the process over time. Subgroups should be collected often
enough, and at appropriate times so that they can reflect the potential
opportunities for change. The potential causes of change could be due to
work-shift differences, relief operators, warm-up trends, material lots,
etc.

din p
.
Number of Subgroups – The number of subgroups needed to establish

or ou
gly
control limits should satisfy the following criterion: enough subgroups

cc Gr
should be gathered to assure that the major sources of variation which

Inc

d a ion
can affect the process have had an opportunity to appear. Generally, 25

rke Act
al
or more subgroups containing about 100 or more individual readings

on
give a good test for stability and, if stable, good estimates of the process

ma ry
ati
location and spread. This number of subgroups ensures that the effect of

ter ust
en e 11
rn

any extreme values in the range or standard deviation will be minimized.

be otiv /20
wa Ind
nte

In some cases, existing data may be available which could accelerate this
1
aI

first stage of the study. However, they should be used only if they are
3/3
gn

recent and if the basis for establishing subgroups is clearly understood.

an e A n:
Ma

as m

Before continuing, a rational sampling plan must be developed and

o
o
i

documented.
d h ut
t
to:

ted y t ira

Sampling Scheme – If the special causes affecting the process can occur
xp
op en ite 426 sed

unpredictably, the appropriate sampling scheme is a random (or

ro ed E
n

probability) sample. A random sample is one in which every sample

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

point (rational subgroup) has the same chance (probability) of being

en

selected. A random sample is systematic and planned; that is, all sample
an is d ume be t is
5

tec

points are determined before any data are collected. For special causes
Th Doc Num men

that are known to occur at specific times or events, the sampling scheme
s c um S

should utilize this knowledge. Haphazard sampling or convenience

E- der ocu

sampling not based on the expected occurrence of a specific special

gh
Or is d

cause should be avoided since this type of sampling provides a false

sense of security; it can lead to a biased result and consequently a
Th

possible erroneous decision.

Whichever sampling scheme is used all sample points should be
determined before any data are collected (see Deming (1950) and Gruska
(2004)).

NOTE: For a discussion about rational subgrouping and the effect of

subgrouping on control chart interpretation see Appendix A.

57
CHAPTER II - Section A
Control Charting Process

Control Chart Setup

A control chart will have sections for:

Header information including the description of the process and

sampling plan.

Recording/displaying the actual data values collected.

This should also include the date & time or other subgroup
identification.

For interim data calculations (optional for automated charts).

din p
.
or ou
This should also include a space for the calculations based

gly
on the readings and the calculated control statistic(s).

cc Gr
Inc

d a ion
For plotting each of the control statistics being analyzed.

rke Act
al
on
The value for the control statistic is usually plotted on the

ma ry
ati
vertical scale and the horizontal scale is the sequence in

ter ust
en e 11
rn

time. The data values and the plot points for the control
be otiv /20
wa Ind
nte

statistic should be aligned vertically. The scale should be

1
aI

broad enough to contain all the variation in the control

3/3

statistic. A guideline is that the initial scale could be set to

gn

twice the difference between the (expected) maximum and

an e A n:
Ma

as m
o

minimum values.
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

To log observations.
h
ro ed E
n

t p wn se

This section should include details such as process

yri t is Lic 2
d i oc nt r: 4 lice

b
en

adjustments, tooling changes, material changes, or other

an is d ume be t is

events which may affect the variability of the process.

5

tec
Th Doc Num men

o
s c um S
E- der ocu

Record Raw Data

gh
Or is d

Enter the individual values and the identification for each

subgroup.
Th

Log any pertinent observation(s).

Calculate Sample Control Statistic(s) for Each Subgroup

The control statistics to be plotted are calculated from the subgroup

measurement data. These statistics may be the sample mean, median,
range, standard deviation, etc. Calculate the statistics according to the
formulae for the type of chart that is being used.

58
 CHAPTER II - Section A
Control Charting Process

Plot the Control Statistic(s) on the Control Charts

Plot the control statistic on the chart. Make sure that the plot points for
the corresponding control statistics are aligned vertically. Connect the
points with lines to help visualize patterns and trends.
The data should be reviewed while they are being collected in order to
identify potential problems. If any points are substantially higher or
lower than the others, confirm that the calculations and plots are correct
and log any pertinent observations.

Establish Control Limits

din p
Control limits are defined by the natural variation of the control statistic.

.
or ou
gly
They define a range of values that the control statistic could randomly

cc Gr
fall within, given there is only common cause to the variation. If the

Inc

d a ion
average of two different subgroups from the same process is calculated,
it is reasonable to expect that they will be about the same. But since they

rke Act
al
on
were calculated using different parts, the two averages are not expected

ma ry
to be identical. Even though the two averages are different, there is a
ati

ter ust
en e 11
limit to how different they are expected to be, due to random chance.
rn

be otiv /20
wa Ind
nte

This defines the location of the control limits.

1
aI

This is the basis for all control chart techniques. If the process is stable
3/3
gn

(i.e., having only common cause variation), then there is a high

an e A n:
Ma

probability that for any subgroup sample the calculated control statistic
as m
o
o

will fall within the control limits. If the control statistic exceeds the
i
d h ut
t
to:

ted y t ira

control limits then this indicates that a special cause variation may be
xp
op en ite 426 sed

present.
h
ro ed E

There are two phases in statistical process control studies.

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

1. The first is identifying and eliminating the special causes of

an is d ume be t is

variation in the process. The objective is to stabilize the process.

5

tec

A stable, predictable process is said to be in statistical control.

Th Doc Num men

o
s c um S

2. The second phase is concerned with predicting future

E- der ocu

measurements thus verifying ongoing process stability. During

gh

this phase, data analysis and reaction to special causes is done in

Or is d

real time. Once stable, the process can be analyzed to determine

Th

if it is capable of producing what the customer desires.

Identify the centerline and control limits of the control chart

To assist in the graphical analysis of the plotted control statistics, draw

lines to indicate the location estimate (centerline) and control limits of
the control statistic on the chart.
In general, to set up a control chart calculate:
Centerline,
Upper Control Limit (UCL),
Lower Control Limit (LCL).
See Chapter II, Section C, for the formulas.

59
CHAPTER II - Section A
Control Charting Process

Interpret for Statistical Control

If the process has no special causes affecting its variability, then the
control statistics will fall between the control limits in a random fashion
(i.e., no patterns will be evident).
Special causes can affect either the process location (e.g., average,
median) or the variation (e.g., range, standard deviation) or both. The
objective of control chart analysis is to identify any evidence that the
process variability or the process location is not operating at a constant
level – that one or both are out of statistical control – and to take
appropriate action.

din p
In the subsequent discussion, the Average will be used for the location

.
or ou
gly
control statistic and the Range for the variation control statistic. The

cc Gr
conclusions stated for these control statistics also apply equally to the

Inc
other possible control statistics.

d a ion
Since the control limits of the location statistic are dependent on the

rke Act
al
on
variation statistic, the variation control statistic should be analyzed first

ma ry
ati
for stability. The variation and location statistics are analyzed separately,

ter ust
en e 11
rn
but comparison of patterns between the two charts may sometimes give
be otiv /20
wa Ind
nte

added insight into special causes affecting the process

1
aI

A process cannot be said to be stable (in statistical control) unless both

3/3
gn

charts have no out-of-control conditions (indications of special causes).

an e A n:
Ma

as m
o
o
i
d h ut
t

Analyze the Data Plots on the Range Chart

to:

ted y t ira
xp
op en ite 426 sed

Since the ability to interpret either the subgroup ranges or subgroup

h
ro ed E
n

averages depends on the estimate of piece-to-piece variability, the R

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

chart is analyzed first. The data points are compared with the control
en

limits, for points out of control or for unusual patterns or trends (see
an is d ume be t is
5

tec

Chapter II, Section D)

Th Doc Num men

o
s c um S
E- der ocu

Find and Address Special Causes (Range Chart)

gh
Or is d

For each indication of a special cause in the range chart data, conduct an
Th

analysis of the process operation to determine the cause and improve

process understanding; correct that condition, and prevent it from
recurring. The control chart itself should be a useful guide in problem
analysis, suggesting when the condition may have began and how long it
continued. However, recognize that not all special causes are negative;
some special causes can result in positive process improvement in terms
of decreased variation in the range – those special causes should be
assessed for possible institutionalization within the process, where
appropriate.
Timeliness is important in problem analysis, both in terms of minimizing
the production of inconsistent output, and in terms of having fresh
evidence for diagnosis. For instance, the appearance of a single point
beyond the control limits is reason to begin an immediate analysis of the

60
 CHAPTER II - Section A
Control Charting Process

process. A process event log may also be a helpful source of information

in terms of identifying special causes of variation.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Figure II.6: Control Limits Recalculation

It should be emphasized that problem solving is often the most difficult

and time-consuming step. Statistical input from the control chart can be
an appropriate starting point, but other methods such as Pareto charts,
cause and effect diagrams, or other graphical analysis can be helpful (see
Ishikawa (1976)). Ultimately, however, the explanations for behavior lie
within the process and the people who are involved with it.
Thoroughness, patience, insight and understanding will be required to
develop actions that will measurably improve performance.

61
CHAPTER II - Section A
Control Charting Process

Recalculate Control Limits (Range Chart)

When conducting an initial process study or a reassessment of process

capability, the control limits should be recalculated to exclude the effects
of out-of-control periods for which process causes have been clearly
identified and removed or institutionalized. Exclude all subgroups
affected by the special causes that have been identified and removed or
institutionalized, then recalculate and plot the new average range ( R )
and control limits. Confirm that all range points show control when
compared to the new limits; if not, repeat the identification, correction,
recalculation sequence.
If any subgroups were dropped from the R chart because of identified
special causes, they should also be excluded from the X chart. The

din p
.
or ou
gly
revised R and X should be used to recalculate the trial control limits

cc Gr
for averages, X ± A2 R (see Figure II.6).

Inc

d a ion
rke Act
al
NOTE: The exclusion of subgroups representing unstable conditions is
on
not just “throwing away bad data.” Rather, by excluding the points

ma ry
ati
affected by known special causes, there is a better estimate of the

ter ust
en e 11
rn

background level of variation due to common causes. This, in turn, gives

be otiv /20
wa Ind
nte

the most appropriate basis for the control limits to detect future
occurrences of special causes of variation. Be reminded, however, that
1
aI

3/3

the process must be changed so the special cause will not recur (if
gn

undesirable) as part of the process.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Find and Address Special Causes (Average Chart)

xp
op en ite 426 sed

h
ro ed E

Once the special cause which affect the variation (Range Chart) have
n

t p wn se
yri t is Lic 2

been identified and their effect have been removed, the Average Chart
d i oc nt r: 4 lice

b
en

can be evaluated for special causes. In Figure II.6 the new control limits
an is d ume be t is

for the averages indicate that two samples are out of control.
5

tec
Th Doc Num men

For each indication of an out-of-control condition in the average chart

data, conduct an analysis of the process operation to determine the reason
s c um S
E- der ocu

for the special cause; correct that condition, and prevent it from
gh

recurring. Use the chart data as a guide to when such conditions began
Or is d

and how long they continued. Timeliness in analysis is important, both

Th

for diagnosis and to minimize inconsistent output. Again, be aware that

not all special causes need be undesirable (see Chapter I, Section E and
Chapter II, Section B).
Problem solving techniques such as Pareto analysis and cause-and-effect
analysis can help. (Ishikawa (1976)).

Recalculate Control Limits (Average Chart)

When conducting an initial process study or a reassessment of process

capability, exclude any out-of-control points for which special causes
have been found and removed; recalculate and plot the process average
and control limits. Confirm that all data points show control when

62
 CHAPTER II - Section A
Control Charting Process

compared to the new limits; if not, repeat the identification, correction,

recalculation sequence.

Final Comments
The preceding discussions were intended to give a functional
introduction to control chart analysis. Even though these discussions
used the Average and Range Charts, the concepts apply to all control
chart approaches.
Furthermore, there are other considerations that can be useful to the
analyst. One of the most important is the reminder that, even with
processes that are in statistical control, the probability of getting a false

din p
signal of a special cause on any individual subgroup increases as more

.
or ou
gly
data are reviewed.

cc Gr
While it is wise to investigate all signals as possible evidence of special

Inc

d a ion
causes, it should be recognized that they may have been caused by the

rke Act
al
system and that there may be no underlying local process problem. If no
on
clear evidence of a special cause is found, any “corrective” action will

ma ry
ati
probably serve to increase, rather than decrease, the total variability in

ter ust
en e 11
rn

the process output. be otiv /20

wa Ind
nte

For further discussion of interpretation, tests for randomness in data, and

1
aI

3/3

problem-solving, see AT&T (1984), Duncan (1986), Grant and

gn

Leavenworth (1996), Juran and Godfrey (1999), Charbonneau and

an e A n:
Ma

as m

Gordon (1978), Ishikawa (1976), Wheeler (1991, 1995), and Ott (2000).
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

63
 Control Charting Process
CHAPTER II - Section A

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

64
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly

Figure II.7: Extend Control Limits for Ongoing Control

.
 CHAPTER II - Section A
Control Charting Process

Extend Control Limits for Ongoing Control

When the initial (or historical) data are consistently contained within the
trial control limits, extend the limits to cover future periods. It might be
desirable here to adjust the process to the target if the process center is
off target. These limits would be used for ongoing monitoring of the
process, with the operator and local supervision responding to signs of
out-of-control conditions on either the location and variation X or R
chart with prompt action (see Figure II.7).
A change in the subgroup sample size would affect the expected average
range and the control limits for both ranges and averages. This situation

din p
.
or ou
gly
could occur, for instance, if it were decided to take smaller samples more

cc Gr
frequently, so as to detect large process shifts more quickly without

Inc
increasing the total number of pieces sampled per day. To adjust central

d a ion
lines and control limits for a new subgroup sample size, the following

rke Act
al
steps should be taken:
on

ma ry
ati
• Estimate the process standard deviation (the estimate is shown as

ter ust
en e 11
rn

σˆ –“sigma hat”).19 Using the existing subgroup size calculate:

be otiv /20
wa Ind
nte

σˆ C = R d where R is the average of the subgroup ranges (for

1
aI

3/3

2
gn

periods with the ranges in control) and d 2 is a constant varying by

an e A n:
Ma

as m

sample size n, the number of samples in a subgroup, as shown in the

o
o
i
d h ut
t
to:

partial table below, taken from Appendix E:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

n 2 3 4 5 6 7 8 9 10
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

d2 1.13 1.69 2.06 2.33 2.53 2.70 2.85 2.97 3.08

an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

• Using the table factors based on the new subgroup size, calculate the
gh

new range and control limits:

Or is d
Th

Rnew = σˆ C ⋅ d 2

Plot these new control limits on the chart as the basis for ongoing process
control. As long as the process remains in control for both averages and
ranges, the ongoing limits can be extended for additional periods. If,
however, there is evidence that the process average or range has changed
(in either direction), the cause should be determined and, if the change is
justifiable, control limits should be recalculated based on current
performance.

19
This manual will distinguish between the estimated standard deviation due to the within-subgroup
variation and the total variation by using the subscripts “C” and “P”, respectively.

65
CHAPTER II - Section A
Control Charting Process

Final Concepts on “Control” – For Further

Consideration
“A perfect state of control is never attainable in a production process.
The goal of the process control charts is not perfection, but a reasonable
and economical state of control. For practical purposes, therefore, a
controlled process is not one where the chart never goes out of control. If
a chart never went out of control we would seriously question whether
that operation should be charted. For shop purposes a controlled process
is considered to be one where only a small percentage of the points go
out of control and where out-of-control points are followed by proper
action20.” See also Figure II.8.

din p
.
or ou
gly
Obviously, there are different levels or degrees of statistical control. The

cc Gr
definition of control used can range from mere outliers (beyond the

Inc

d a ion
control limits), through runs, trends and stratification, to full zone

rke Act
al
analysis. As the definition of control used advances to full zone analysis,
on
the likelihood of finding lack of control increases (for example, a process

ma ry
ati
with no outliers may demonstrate lack of control through an obvious run

ter ust
en e 11
rn

still within the control limits). For this reason, the definition of control
be otiv /20
wa Ind
nte

used should be consistent with your ability to detect this at the point of
1
aI

control and should remain the same within one time period, within one
3/3

process. Some suppliers may not be able to apply the fuller definitions of
gn

control on the floor on a real-time basis due to immature stages of

an e A n:
Ma

as m
o

operator training or lack of sophistication in the operator’s ability. The

o
i
d h ut
t
to:

ted y t ira

ability to detect lack of control at the point of control on a real-time basis

is an advantage of the control chart. Over-interpretation of the data can
xp
op en ite 426 sed

be a danger in maintaining a true state of economical control.

ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

20
AT&T (1984)

66
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

67
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.

Figure II.8: Process Variation Relative to Specification Limits

Control Charting Process
CHAPTER II - Section A
 Control Charting Process
CHAPTER II - Section A

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

68
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER II - Section B
Defining “Out of Control” Signals

CHAPTER II - Section B
Defining “Out-of-Control” Signals
Point Beyond a Control
Limit.
.
The presence of one or more points beyond either control limit is primary
evidence of special cause variation at that point. This special cause could
have occurred prior to this point.

din p
.
or ou
gly
Since points beyond the control limits would be rare if only variation

cc Gr
from common causes were present, the presumption is that a special

Inc

d a ion
cause has accounted for the extreme value. Therefore, any point beyond
a control limit is a signal for analysis of the operation for the special

rke Act
al
cause. Mark any data points that are beyond the control limits for
on

ma ry
investigation and corrective action based on when that special cause
ati

ter ust
en e 11
actually started.
rn

be otiv /20
wa Ind
nte

A point outside a control limit is generally a sign of one or more of the

1
aI

following:
3/3
gn

• The control limit or plot point has been miscalculated or misplotted.

an e A n:
Ma

as m
o
o

• The piece-to-piece variability or the spread of the distribution has

i
d h ut
t
to:

ted y t ira

increased (i.e., worsened), either at that one point in time or as part

xp
op en ite 426 sed

of a trend.
h
ro ed E
n

• The measurement system has changed (e.g., a different appraiser or

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

instrument).
en
an is d ume be t is
5

• The measurement system lacks appropriate discrimination.

tec
Th Doc Num men

For charts dealing with the spread, a point below the lower control limit
s c um S

is generally a sign of one or more of the following:

E- der ocu

gh

• The control limit or plot point is in error.

Or is d

• The spread of the distribution has decreased (i.e., becomes better).

Th

• The measurement system has changed (including possible editing or

alteration of the data).
A point beyond either control limit is generally a sign that the process
has shifted either at that one point or as part of a trend (see Figure II.9).

When the ranges are in statistical control, the process spread – the
within-subgroup variation – is considered to be stable. The averages can
then be analyzed to see if the process location is changing over time.
Since control limits for X are based upon the amount of variation in the
ranges, then if the averages are in statistical control, their variation is
related to the amount of variation seen in the ranges – the common-cause

69
CHAPTER II - Section B
Defining “Out of Control” Signals

variation of the system. If the averages are not in control, some special
causes of variation are making the process location unstable.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Figure II.9: Points Beyond Control Limits

an is d ume be t is
5

tec
Th Doc Num men

Patterns or Trends Within

s c um S
E- der ocu

the Control Limits

gh
Or is d

The presence of unusual patterns or trends, even when all ranges are
Th

within the control limits, can be evidence of the influence of a special

cause during the period of the pattern or trend. This could give the first
warning of an unfavorable condition which should be corrected.
Conversely, certain patterns or trends could be favorable and should be
studied for possible permanent improvement of the process. Comparison
of patterns between the range and average charts may give added insight.
There are situations where an “out-of-control pattern” may be a bad
event for one process and a good event for another process. An example
of this is that in an X and R chart a series of 7 or more points on one
side of the centerline may indicate an out-of-control situation. If this
happened in a p chart, the process may actually be improving if the series
is below the average line (less nonconformances are being produced). So
in this case the series is a good thing – if we identify and retain the cause.

70
 CHAPTER II - Section B
Defining “Out of Control” Signals

Runs
Runs – Each of the following are signs that a process shift or trend has
begun:

• 7 points in a row on one side of the X or R .

• 7 points in a row that are consistently increasing (equal to or greater
than the preceding points), or consistently decreasing.
Mark the point that prompts the decision; it may be helpful to extend a
reference line back to the beginning of the run. Analysis should consider
the approximate time at which it appears that the trend or shift first
began.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Figure II.10: Runs in an Average Control Chart

A run above the average range, or a run up, signifies one or both of the
following:
Greater spread in the output values, which could be from an
irregular cause (such as equipment malfunction or loose
fixturing) or from a shift in one of the process elements (e.g., a
new, less uniform raw material lot).
A change in the measurement system (e.g., new inspector or
gage).

71
CHAPTER II - Section B
Defining “Out of Control” Signals

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Figure II.11: Runs in a Range Control Chart

an is d ume be t is
5

tec
Th Doc Num men

A run below the average range, or a run down, signifies one or both of
s c um S
E- der ocu

the following:
gh

Smaller spread in output values, which is usually a good

Or is d

condition that should be studied for wider application and

Th

process improvement.
A change in the measurement system, which could mask real
performance changes.

NOTE: As the subgroup size (n) becomes smaller (5 or less), the

likelihood of runs below R increases, so a run length of 8 or more could
be necessary to signal a decrease in process variability.

A run relative to the process average is generally a sign of one or both of

the following:
The process average has changed – and may still be changing.
The measurement system has changed (drift, bias, sensitivity,
etc.).

72
 CHAPTER II - Section B
Defining “Out of Control” Signals

Obvious Nonrandom Patterns

In addition to the presence of points beyond control limits or long runs,
other distinct patterns may appear in the data that give clues to special
causes. Care should be taken not to over-interpret the data, since even
random (i.e., common cause) data can sometimes give the illusion of
nonrandomness (i.e., special causes). Examples of nonrandom patterns
could be obvious trends (even though they did not satisfy the runs tests),
cycles, the overall spread of data points within the control limits, or even
relationships among values within subgroups (e.g., the first reading
might always be the highest). One test for the overall spread of subgroup
data points is described below.

din p
.
or ou
gly
Distance of points from R or X : Generally, about 2/3 of the plotted

cc Gr
points should lie within the middle third of the region between the

Inc
control limits; about 1/3 of the points should be in the outer two-thirds of

d a ion
the region. If substantially more than 2/3 of the plotted points lie close to

rke Act
al
on
R or X investigate one or more of the following:

ma ry
ati

ter ust
• The control limits or plot points have been miscalculated or
en e 11
rn

misplotted. be otiv /20

wa Ind
nte

•
1
aI

The process or the sampling method is stratified; each subgroup

3/3

systematically contains measurements from two or more process

gn

streams that have very different process averages (e.g., one piece
an e A n:
Ma

as m
o

from each of several spindles).

o
i
d h ut
t
to:

ted y t ira

• The data have been edited (subgroups with ranges that deviated
xp
op en ite 426 sed

much from the average have been altered or removed).

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

If substantially fewer than 2/3 of the plotted points lie close to R (for 25
an is d ume be t is
5

tec

subgroups if 40% or fewer are in the middle third), investigate one or

Th Doc Num men

both of the following:

s c um S
E- der ocu

• The control limits or plot points have been miscalculated or

gh

misplotted.
Or is d

• The process or the sampling method causes successive subgroups to

Th

contain measurements from two or more process streams that have

dramatically different variability (e.g., mixed lots of input materials).
If several process streams are present, they should be identified and
tracked separately (see also Appendix A). Figure II.12 shows a
nonrandom pattern for the R chart.

73
 CHAPTER II - Section B

Th
Or is d
E- der ocu
Defining “Out of Control” Signals

Th Doc Num men

an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

74
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p

Figure II.12: Nonrandom Patterns in a Control Chart

gly
.
 CHAPTER II - Section B
Defining “Out of Control” Signals

Special Cause Criteria

There are several criteria for identifying special causes (see table below
and AT&T (1984)). The most commonly used are discussed above. The
decision as to which criteria to use depends on the process being
studied/controlled.

Summary of Typical Special Cause Criteria

1 1 point more than 3 standard deviations21 from centerline
2 7 points in a row on same side of centerline

din p
3 6 points in a row, all increasing or all decreasing

.
or ou
gly
4 14 points in a row, alternating up and down

cc Gr
Inc
5 2 out of 3 points > 2 standard deviations from centerline (same side)

d a ion
6 4 out of 5 points > 1 standard deviation from centerline (same side)

rke Act
al
on
7 15 points in a row within 1 standard deviation of centerline (either side)

ma ry
ati

ter ust
8 8 points in a row > 1 standard deviation from centerline (either side)
en e 11
rn

be otiv /20
wa Ind
nte

Table II.1
1
aI

3/3
gn

Note 1: Except for the first criterion, the numbers associated with the
an e A n:
Ma

as m

criteria do not establish an order or priority of use. Determination of

o
o
i

which of the additional criteria to use depends on the specific process

d h ut
t
to:

ted y t ira

characteristics and special causes which are dominant within the

xp
op en ite 426 sed

process.
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Note 2: Care should be given not to apply multiple criteria except in

en

those cases where it makes sense. The application of each additional

an is d ume be t is
5

criterion increases the sensitivity of finding a special cause but also

tec
Th Doc Num men

increases the chance of a Type I error.

o
s c um S
E- der ocu

In reviewing the above, it should be noted that not all these

gh
Or is d

considerations for interpretation of control can be applied on the

production floor. There is simply too much for the appraiser to
Th

remember and utilizing the advantages of a computer is often not feasible

on the production floor. So, much of this more detailed analysis may
need to be done offline rather than in real time. This supports the need
for the process event log and for appropriate thoughtful analysis to be
done after the fact.
Another consideration is in the training of operators. Application of the
additional control criteria should be used on the production floor when
applicable, but not until the operator is ready for it; both with the
appropriate training and tools. With time and experience the operator
will recognize these patterns in real time.

21
In this table, “standard deviation” refers to the standard deviation used in the calculations of the
control limits.

75
CHAPTER II - Section B
Defining “Out of Control” Signals

Average Run Length (ARL)

Chapter I stated that decisions made based on charts should balance the
risks of Type I errors (over-control, false alarms) to Type II errors
(under-control). A measure of this balance is the Average Run Length
( ARL ).
The Average Run Length is the number of sample subgroups expected
between out-of-control signals. The in-control Average Run Length
( ARL0 ) is the expected number of subgroup samples between false
alarms.

din p
.
or ou
gly
1

cc Gr
ARL0 =
Pr {Type I Error }

Inc

d a ion
rke Act
al
The ARL is dependent on how out-of-control signals are defined, the
on
true target value’s deviation from the estimate, and the true variation

ma ry
ati

ter ust
relative to the estimate.
en e 11
rn

be otiv /20
wa Ind
nte

Below is a table of approximate ARL’s for the standard Shewhart X

1
aI

control chart with exceeding the ± 3σ X control limits as the only out-of-
3/3
gn

control signal.
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Shift in Target
xp
op en ite 426 sed

h
ro ed E

σX ARL
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

0 370.4
an is d ume be t is
5

0.1 352.9
tec
Th Doc Num men

0.2 308.4
s c um S
E- der ocu

0.3 253.1
gh
Or is d

0.5 155.2
Th

1.0 43.9
1.5 15.0
2.0 6.3
3.0 2.0
4.0 1.2

This table indicates that a mean shift of 1.5 standard deviations (of the
mean) would be signaled (on average) by the 15th subgroup after the
shift. A shift of 4 standard deviations would be identified within 2
subgroups.

76
 CHAPTER II - Section B
Defining “Out of Control” Signals

This table also shows that a false signal may be indicated for a process
without a shift (i.e., the process remains in statistical control) every 370
subgroups (on average).
σ
Since σ X = , the practical magnitude of the shifts can be reduced by
n
increasing the number of items in each subgroup. Larger subgroups
reduce the size of σ X and tighten the control limits around X .

Alternatively, the ARL ' s can be reduced by adding more out-of-control

criteria. Other signals such as runs tests and patterns analysis along with
the control limits will reduce the size of the ARL ' s .
The following table is approximate ARL ' s for the same chart adding

din p
.
or ou
gly
the runs test of 7-points in a row on one side of X .

cc Gr
Inc

d a ion
Shift in Target

rke Act
al
σ X ’s on

ma ry
ati
ARL

ter ust
en e 11
rn

0
be otiv /20 59.8
wa Ind
nte

1
aI

0.1 53.9
3/3
gn

0.2 41.8
an e A n:
Ma

as m
o

0.3 30.8
o
i
d h ut
t
to:

ted y t ira

0.5 17.9
xp
op en ite 426 sed

1.0 8.7
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

1.5 6.9
en
an is d ume be t is

2.0 6.1
5

tec
Th Doc Num men

3.0 2.0
o
s c um S

4.0 1.2
E- der ocu

gh
Or is d

As can be seen, adding the one extra out-of-control criterion significantly

Th

reduces the ARLs for small shifts in the mean, a decrease in the risk of a
Type II error. Note that the zero-shift (the in-control) ARL is also
reduced significantly. This is an increase in the risk of a Type I error or
false alarm.
This balance between wanting a long ARL when the process is in
control versus a short ARL when there is a process change has led to the
development of other charting methods. Some of those methods are
briefly described in Chapter III.

77
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

78
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion

Figure II.13: Average and Range Charts

cc Gr
or ou
din p
gly
.
 CHAPTER II - Section C
Control Chart Formulas:

CHAPTER II - Section C
Control Chart Formulas
Control chart constants for all control charts discussed in this section are
listed in Appendix E.

Variables Control Charts

Average and Range Charts ( X , R )

din p
.
or ou
gly
Subgroup Average:

cc Gr
Inc

d a ion
x1 + x2 + ... + xn
X = ;

rke Act
al
n
on

ma ry
n = number of samples in a subgroup
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Subgroup Range:
1
aI

3/3
gn

R = xMax − xMin (within each subgroup)

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

Grand Average:
ted y t ira
xp
op en ite 426 sed

X 1 + X 2 + ... + X k
h
ro ed E

X= ;
n

k
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

k = number of subgroups used to determine the

an is d ume be t is
5

Grand Average and Average Range

tec
Th Doc Num men

Average Range:
s c um S
E- der ocu

R1 + R2 + ....Rk
gh

R= ;
Or is d

k
Th

Estimate of the Standard Deviation of X :

σˆ C = R d
2

Estimate of the Standard Deviation of X :

σˆ X = σ C
ˆ
n

79
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
CLR = R
CLX = X

yri t is Lic 2 to:

gh o en
Centerline

t p wn se Ma
ro ed E gn
tec b xp aI
Chart Features:

ted y t ira nte

h t i

80
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
UCLR = D4 R

ma ry
Control Limits
UCLX = X + A2 R

rke Act
d a ion
cc Gr
or ou
din p
gly
.
LCLR = D3 R
LCLX = X − A2 R
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

81
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Control Chart Formulas:
CHAPTER II - Section C
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

82
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p

Figure II.14: Average and Standard Deviation Charts

gly
.
 CHAPTER II - Section C
Control Chart Formulas:

Average and Standard Deviation Charts ( X , s )

Subgroup Average:
x1 + x2 + ... + xn
X = ;
n
n = number of samples in a subgroup

Subgroup Standard Deviation (Within-subgroup Variation):

sk =
∑(X i ,k −X k )2
n −1

din p
.
or ou
gly
Grand Average:

cc Gr
X 1 + X 2 + ... + X k

Inc

d a ion
X=
k

rke Act
al
on
k = number of subgroups used to determine

ma ry
ati
the Grand Average and Average

ter ust
en e 11
rn

be otiv /20 Standard Deviation

wa Ind
nte

1
aI

3/3
gn

Average Standard Deviation:22

an e A n:
Ma

as m

s1 + s2 + ....sk
o
o
i

s=
d h ut
t
to:

ted y t ira

k
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Estimate of the Standard Deviation of X :

en
an is d ume be t is

σˆ C = s c
5

tec
Th Doc Num men

4
s c um S

Estimate of the Standard Deviation of X :

E- der ocu

gh

σˆ X = σ C
Or is d

ˆ
n
Th

Chart Features:
Centerline Control Limits
CLX = X UCLX = X + A3 s LCLX = X − A3 s

CLs = s UCLs = B4 s LCLs = B3 s

22
Also known as the pooled standard deviation.

83
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

84
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act

Figure II.15: Median and Range Charts

d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER II - Section C
Control Chart Formulas:

Median and Range Charts X , R ( )

Sample Value: xi , i = 1...n (sample size)
Subgroup Median:
X ( O ) is the value of the Oth element in the sample
when the data are arranged in ascending order

  n2+1 

din p
X   if n is odd

.
or ou
gly

X k =   n 

cc Gr
 n+2 
 
X + X 2 
 2

Inc

d a ion
 if n is even
2

rke Act
al
on
n = number of elements in a subgroup

ma ry
ati

ter ust
en e 11
rn

k = number of subgroups used to determine the

be otiv /20
wa Ind
nte

Average Median and Average Range

1
aI

3/3
gn

Subgroup Range:
an e A n:
Ma

as m
o
o

R = xMax − xMin (within each subgroup)

i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

Average Median:
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

X 1 + X 2 + ... + X k
b

X =
en

;
k
an is d ume be t is
5

tec
Th Doc Num men

Average Range:
o
s c um S

R1 + R2 + ....Rk
E- der ocu

R= ;
gh

k
Or is d

Estimate of the Standard Deviation of X :

Th

σˆ C = R d
2

Chart Features:23
Centerline Control Limits

CLX = X UCLX = X + A2 R LCLX = X − A2 R

CLR = R UCLR = D4 R LCLR = D3 R

23
This approach to the Median Chart uses averages in the calculation of the centerline and control
limits. There are other approaches in the literature which do not use averages.

85
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

86
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p

Figure II.16: Individual and Moving Range Charts

gly
.
 CHAPTER II - Section C
Control Chart Formulas:

Individuals and Moving Range Charts ( X , MR )

Individual Value: xi , i = 1,..., k individual values :
Average of Individual Values:
x1 + x2 + ...xk
X =
k
Moving Range:
MRi = xi − xi −1 , i = 2...k
(Range between current value and previous value.)

din p
.
or ou
gly
cc Gr
Average Moving Range:

Inc

d a ion
MR2 + MR3 + ....MRk
MR =

rke Act
al
k −1
on

ma ry
ati
Estimate of the Standard Deviation of X :

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

σˆ C = R d
1
aI

2
3/3
gn

an e A n:
Ma

as m
o
o

Chart Features:
i
d h ut
t
to:

ted y t ira

Centerline Control Limits

xp
op en ite 426 sed

h
ro ed E

CLX = X UCLX = X + E2 R LCLX = X − E2 R

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

CLR = R UCLR = D4 R LCLR = D3 R

en
an is d ume be t is
5

tec
Th Doc Num men

Because moving ranges are involved, the points being plotted on the
s c um S
E- der ocu

range chart are correlated. Therefore, valid signals occur only in the
gh

form of points beyond the control limits. Other rules used to evaluate the
Or is d

data for non-random patterns (see Chapter II, Section B) are not reliable
indicators of out-of-control conditions.
Th

87
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

88
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou

Figure II.17: Proportion Nonconforming Chart

din p
gly
.
 CHAPTER II - Section C
Control Chart Formulas:

Attributes Control Charts

Control Charts for Nonconforming Items24

Attributes charts are part of probability based charts discussed in Chapter
III. These control charts use categorical data and the probabilities related
to the categories to identify the presences of special causes. The analysis
of categorical data by these charts generally utilizes the binomial, or
poisson distribution approximated by the normal form.
Traditionally attributes charts are used to track unacceptable parts by
identifying nonconforming items and nonconformities within an item.
There is nothing intrinsic in attributes charts that restricts them to be
solely used in charting nonconforming items. They can also be used for

din p
.
or ou
gly
tracking positive events. However, we will follow tradition and refer to

cc Gr
these as nonconformances and nonconformities.

Inc

d a ion
rke Act
al
Proportion Nonconforming (p Chart)
on

ma ry
ati

ter ust
en e 11
Guideline:
rn

be otiv /20
wa Ind
nte

Since the control limits are based on a normal approximation, the

1
aI

sample size used should be such that np ≥ 5 .

3/3
gn

an e A n:
Ma

as m
o
o
i

Individual Value
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

npi
pi = ni = number of parts inspected;
h
ro ed E

ni
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

npi = number of nonconforming items found

an is d ume be t is
5

tec
Th Doc Num men

Average of Individual Values

s c um S
E- der ocu

np1 + np2 + .... + npk

p=
gh

where k = number of subgroups

Or is d

n1 + n2 + .... + nk
Th

p1 + p2 + .... + pk
p= if all the ni’s are equal
k

24
An alternative to these charts is the Individuals and Moving Range Chart (see Wheeler (1995)).

89
CHAPTER II - Section C
Control Chart Formulas:

Chart Features:
Centerline Control Limits
p (1 − p )
CLp = p UCLPi = p + 3
ni

p (1 − p )
LCLPi = p − 3
ni

If the sample size is constant (n)

din p
.
or ou
Control Limits

gly
cc Gr
p (1 − p )

Inc

d a ion
UCLP = p + 3
n

rke Act
al
p (1 − p ) on

ma ry
ati
LCLP = p − 3

ter ust
en e 11
rn

be otiv /20n

wa Ind
nte

1
aI

3/3

Constant control limits when the sample size varies

gn

an e A n:
Ma

as m

min ni
≥ 0.75 )
o

(for situations where

o
i
d h ut
t

max ni
to:

ted y t ira
xp
op en ite 426 sed

Control Limits
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

p (1 − p )
en

UCLP = p + 3 ( n = average sample size)

an is d ume be t is
5

n
tec
Th Doc Num men

p (1 − p )
s c um S

LCLP = p − 3 ( n = average sample size)

E- der ocu

n
gh
Or is d

Example Uses:
Th

• Accept/Reject Decisions with constant or variable subgroup size

First Time Quality (FTQ) results25
Proportion nonconforming
Proportion conforming26
Proportion of items above (or below) a threshold value
• Judgment Decisions
Proportion of items within a specified category
Proportion of items above (or below) a threshold value
Proportion Uptime (equipment)

25
This is alternatively known as FTC (First Time Capability) and RTY (Rolled Throughput Yield).
26
This chart is sometimes called a q-chart; this is based on the practice of calculating the parameter
q = 1 – p.

90
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

91
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Control Chart Formulas:
CHAPTER II - Section C
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

92
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr

Figure II.18: Number of Nonconforming Chart

or ou
din p
gly
.
 CHAPTER II - Section C
Control Chart Formulas:

Number of Nonconforming Chart (np Chart)

Restriction:
Requires a constant subgroup size = n
Guideline:
Since the control limits are based on a normal approximation,
the sample size used should be such that np ≥ 5

Individual Value:
npi n = number of parts inspected;

din p
.
or ou
gly
np = number of nonconforming items found

cc Gr
Inc
Average of Individual Values:

d a ion
rke Act
al
np1 + np2 + .... + npk
np =
on

ma ry
k
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Chart Features:
1
aI

3/3
gn

Centerline Control Limits:

an e A n:
Ma

as m
o
o

np
i
d h ut
t
to:

CLnp = np UCLnp = np + 3 np(1 − ) = np + 3 np(1 − p )

ted y t ira

n
xp
op en ite 426 sed

h
ro ed E
n

np
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

LCLnp = np − 3 np(1 − ) = np − 3 np(1 − p )

b
en

n
an is d ume be t is
5

tec
Th Doc Num men

Example Uses:
s c um S
E- der ocu

•
gh

Accept/Reject Decisions with constant subgroup size

Or is d

First Time Quality (FTQ) results

Number nonconforming
Th

Number conforming
Number of items above (or below) a threshold value
• Judgment Decisions
Number of items within a specified category
Number of items above (or below) a threshold value
Number of times a condition occurs

93
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

94
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly

Figure II.19: Number of Nonconforming per Unit Chart

.
 CHAPTER II - Section C
Control Chart Formulas:

Number of Nonconformities per Unit Chart (u Chart)

Guideline:
Since the control limits are based on a normal
approximation, the sample size used must be large enough
so that the number of subgroups with c = 0 is small.
Individual Value:
ci
ui = ci = number of nonconformities found in sample i;
ni
ni = is the sample size
Average of Individual Values:

din p
.
or ou
gly
u1 + u2 + .... + uk

cc Gr
u=

Inc
k

d a ion
rke Act
al
Chart Features:
on

ma ry
Centerline Control Limits
ati

ter ust
en e 11
rn

be otiv /20 3 u u
wa Ind
nte

CLu = u UCLu = u + =u +3
ni ni
1
aI

3/3
gn

an e A n:

3 u u
Ma

as m

LCLu = u −
=u −3
o
o
i

ni ni
d h ut
t
to:

ted y t ira

For constant control limits when the sample size varies

xp
op en ite 426 sed

h
ro ed E

min ni
n

t p wn se
yri t is Lic 2

≥ 0.75 )
d i oc nt r: 4 lice

(for situations where

b
en

max ni
an is d ume be t is
5

tec
Th Doc Num men

Control Limits:
o
s c um S
E- der ocu

3 u u
UCLu = u + = u +3 ( n = average sample size)
gh

n
Or is d

n
Th

3 u u
LCLu = u − = u −3 ( n = average sample size)
n n
Example Uses:
• Accept/Reject Decisions with variable number items per unit
Quality rates for specified unit designation
Average number (rate) of nonconformities per unit
Average number (rate) of items within one or more categories
• Judgment Decisions
Average number (rate) of items within one or more categories
Average number (rate) of items above (or below) a threshold
value per unit

95
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

96
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou

Figure II.20: Number of Nonconformities Chart

din p
gly
.
 CHAPTER II - Section C
Control Chart Formulas:

Number of Nonconformities Chart (c Chart)

Restriction:
Requires a constant subgroup size = n
Guideline:
Since the control limits are based on a normal approximation, the sample
used must be large enough so that the number of subgroups with c = 0 is
small

Individual Value:
ci = number of nonconformities found in sample; i = 1,… , k

din p
.
or ou
gly
cc Gr
Inc

d a ion
Average of Individual Values:

rke Act
al
c1 + c2 + .... + ck
on
c= k = number of samples

ma ry
k
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Chart Features:
1
aI

3/3
gn

Centerline Control Limits

an e A n:
Ma

as m
o

CLC = c UCLC = c + 3 c
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

LCLC = c − 3 c
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Example Uses:
an is d ume be t is
5

tec
Th Doc Num men

• Accept/Reject Decisions with a constant number items per unit

o

Quality level for specified unit designation

s c um S
E- der ocu

Total number of nonconformities per unit

gh

Total number of items within one or more categories

Or is d

• Judgment Decisions
Th

Total number of items within one or more categories per unit

Total number of items above (or below) a threshold value per
unit
Total number of times a condition occurs within a unit

97
 Control Chart Formulas:
CHAPTER II - Section C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

98
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i

99
an e A n: o rn
ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
CHAPTER III

wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
Other Types of Control Charts

or ou
din p
gly
.
CHAPTER III
Other Types of Control Charts

CONTROL CHARTS

Upper Control Limit

din p
.
or ou
gly
Center Line

cc Gr
Inc

d a ion
rke Act
al
on
Lower Control Limit

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i

1. Collection
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

• Gather Data and plot on a chart.

h
ro ed E
n

t p wn se

2. Control
yri t is Lic 2
d i oc nt r: 4 lice

b
en

• Calculate trial control limits from process data.

an is d ume be t is
5

tec
Th Doc Num men

• Identify special causes of variation and act upon them.

o
s c um S

3. Analysis and Improvement

E- der ocu

gh
Or is d

• Quantify common cause variation; take action to reduce it.

Th

These three phases are repeated for continual process improvement

Figure III.1: Control Charts

100
 CHAPTER III
Other Types of Control Charts

Introduction
There are several types of control charts other than those discussed in the
previous chapters. Most of these charts were developed to address
specific process situations or conditions which can affect the optimal use
of the standard control charts. A brief description of the more common
charts will follow below. This description will define the charts, discuss
when they should be used and list the formulas associated with the chart,
as appropriate. If more information is desired regarding these charts or
others, please consult a reference text that deals specifically with these
types of control charts.

Probability Based Charts

din p
.
or ou
gly
cc Gr
Probability based charts belong to a class of control charts that uses

Inc

d a ion
categorical data and the probabilities related to the categories. The
analysis of categorical data generally uses the binomial, multinomial or

rke Act
al
on
poisson distribution. Examples of these charts are the attributes charts

ma ry
ati
discussed in Chapter II Section C. The attributes charts use the categories

ter ust
en e 11
rn
of “good” and “bad” (e.g., conforming and nonconforming). However,
be otiv /20
wa Ind
nte

there is nothing inherent in any of these forms (or any other forms) that
requires one or more categories to be “bad.”
1
aI

3/3
gn

The problem is that users tend to apply by example, rather than by

an e A n:
Ma

as m

knowledge. This is as much the fault of professionals and teachers, as it

o
o

is the student’s. There is a tendency to take the easy way out, using
i
d h ut
t
to:

ted y t ira

traditional (and stereotypical) examples. This leads to a failure to realize

xp
op en ite 426 sed

that quality practitioners once had (or were constrained to) the tolerance
h
ro ed E

philosophy; i.e., make it “to print” (or “close enough”).

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

Stoplight Control
o
s c um S

With stoplight control charts, the process location and variation are
E- der ocu

controlled using one chart. The chart tracks the number of data points in
gh

Red
Or is d

Yellow the sample in each of the designated categories. The decision criteria are
based on the expected probabilities for these categories.
Th

Green
A typical scenario will divide the process variation into three parts:
warning low, target, warning high. The areas outside the expected
process variation ( 6σ̂ ) are the stop zones. One simple but effective
control procedure of this type is stoplight control which is a semi-
variables (more than two categories) technique using double sampling. In
this approach the target area is designated green, the warning areas as
yellow, and the stop zones as red. The use of these colors gives rise to
the “stoplight” designation.

101
CHAPTER III
Other Types of Control Charts

LSL LSL
Stop Stop
Warning Warning

Target
Target
Warning
Stop
Warning

din p
.
or ou
gly
cc Gr
Stop

Inc
USL USL

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Figure III.2: Stoplight Control

1
aI

3/3
gn

an e A n:
Ma

as m

With this categorization the process can be controlled by identifying and

o
o
i
d h ut
t
to:

charting the proportion of data points designated as “warning” within a

ted y t ira

sample. The apportionment (% warning) controls the sample size and

xp
op en ite 426 sed

frequency required. Of course, this allows process control only if the

h
ro ed E
n

process distribution is known. The quantification and analysis of the

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

process requires variables data.

en
an is d ume be t is
5

tec
Th Doc Num men

The focus of this tool is to detect changes (special causes of variation) in

s c um S

the process. That is, this is an appropriate tool for stage 2 activities27
E- der ocu

only. At its basic implementation, stoplight control requires no

gh
Or is d

computations and no plotting, thereby making it easier to implement than

control charts. Since it splits the total sample (e.g., 5) into a two-stage
Th

sampling (e.g., 2, 3), this approach can signal out-of-control conditions

with the same or better efficiency than a control chart with the same total
sample size (see Heaphy and Gruska (1982)).
Although, the development of this technique is thoroughly founded in
statistical theory, it can be implemented and taught at the operator level
without involving mathematics.

27
See Chapter I, Section F.

102
 CHAPTER III
Other Types of Control Charts

The assumptions in stoplight control are:

• The process is in statistical control.
• Process performance (including measurement variability) is
acceptable.
• The process is on target.

Once the assumptions have been verified by a process performance study

using variables data techniques, the process distribution can be divided
such that the average ± 1.5 standard deviations is labeled as the green
area and the rest of the area within the process distribution is yellow.
Any area outside the process distribution (the 99.73% range) is labeled

din p
.
or ou
gly
red.

cc Gr
If the process distribution follows the normal form, approximately

Inc

d a ion
86.6% of the distribution is in the green area, 13.2% is in the yellow area

rke Act
al
and 0.3% is in the red area. Similar conditions can be established if the

on
distribution is found to be non-normal.

ma ry
ati
Setup

ter ust
en e 11
For control equivalent to an X and R chart with a sample size of 5, the
rn
Process
be otiv /20
wa Ind
nte

steps for stoplight control can be outlined as follows:

1
aI

1. Check 2 pieces; if both pieces are in the green area, continue to run.
3/3

Normal
gn

Production
2. If one or both are in the red zone, stop the process, notify the
an e A n:
Ma

as m

designated person for corrective action and sort material. When setup
o
o
i
d h ut
t
to:

or other corrections are made, repeat step # 1.

ted y t ira

Select 2
Yes
Samples
xp
op en ite 426 sed

3. If one or both are in a yellow zone, check three more pieces. If any
h
ro ed E

pieces fall in a red zone, stop the process, notify the designated
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

All person for corrective action and sort material. When setup or other
b
en

Green?
corrections are made, repeat step # l.
an is d ume be t is
5

tec

No
If no pieces fall in a red zone, but three or more are in a yellow
Th Doc Num men

Implement zone (out of 5 pieces) stop the process, notify the designated
s c um S

No Any Red? Yes Reaction person for corrective action. When setup or other corrections are
E- der ocu

Plan
made, repeat step #1.
gh
Or is d

No If three pieces fall in the green zone and the rest are yellow,
continue to run.
Th

Select 3
Additional
Samples
Measurements can be made with variables as well as attributes gaging.
Certain variables gaging such as dial indicators or air-electronic columns
Yes are better suited for this type of program since the indicator background
Any Red? can be color coded. Although no charts or graphs are required, charting is
Yes recommended, especially if subtle trends (shifts over a relatively long
period of time) are possible in the process.
3 -5
Yellow?
In any decision-making situation there is a risk of making a wrong
decision. With sampling, the two types of errors are:
• Probability of calling the process bad when it is actually good
(false alarm rate).
• Probability of calling the process good when it is actually bad
(miss rate).

103
CHAPTER III
Other Types of Control Charts

In addition to these two measures, the sensitivity of the sampling plan

can be quantified. Sensitivity refers to the ability of the sampling plan to
detect out-of-control conditions due to increased variation or shifts from
the process average.
The disadvantage of stoplight control is that it has a higher false alarm
rate than an X and R chart of the same total sample size.
The advantage of stoplight control is that it is as sensitive as an
X and R chart of the same total sample size.
Users tend to accept control mechanisms based on these types of data
due to the ease of data collection and analysis. Focus is on the target not
specification limits – thus it is compatible with the target philosophy and
continuous improvement.

din p
.
or ou
gly
cc Gr
Pre-Control

Inc

d a ion
An application of the stoplight control approach for the purpose of

rke Act
al
on
nonconformance control instead of process control is called Pre-

ma ry
control. It is based on the specifications not the process variation. Its
ati

ter ust
en e 11
origins can be traced to work by Frank Satterthwaite from Rath & Strong
rn

at the Jones & Lamson Machine Company in 1954.28

be otiv /20
wa Ind
nte

1
aI

3/3
gn

The assumptions in pre-control are:

an e A n:
Ma

as m

• the process has a flat loss function (see section on Loss Function, in
o
o
i
d h ut
t

Chapter IV.)
to:

ted y t ira
xp
op en ite 426 sed

• process performance (including measurement system variability) is

h
ro ed E

less than or equal to the tolerance.

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

The first assumption means that all special sources of variation in the
tec
Th Doc Num men

process are being controlled. The second assumption states that 99.73%
o

of the pieces being produced are within specification without sorting.

s c um S
E- der ocu

If the foregoing assumptions are satisfied, the tolerance can be divided so

gh
Or is d

that Nominal ± ¼ Tolerance is labeled as the green area and the rest of
the area within the specification is yellow. The area outside the
Th

specifications is labeled red. For a process that is normal with Cp, Cpk
equal to 1.00, approximately 86.6% of the pieces are in the green area,
13.2% are in the yellow area and 0.3% are in the red area. Similar
calculations could be done if the distribution was found to be non-normal
or highly capable.

The pre-control sampling uses a sample size of two. However, before the
sampling can start, the process must produce 5 consecutive parts in the
green zone. Each of the two data points are plotted on the chart and
reviewed against a set of rules.

28
See Bhote (1991) and ASQ Statistics Newsletter Vol 05 No 2 Feb. 1984.

104
 CHAPTER III
Other Types of Control Charts

Nom - ½ Tol LSL Nom - ½ Tol LSL

Nom - ¼ Tol Nom - ¼ Tol

Nominal Nominal

din p
.
or ou
gly
cc Gr
Nom + ¼ Tol Nom + ¼ Tol

Inc

d a ion
rke Act
al
Nom + ½ Tol USL Nom + ½ Tol
on
USL

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

Figure III.3: Pre-Control

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

When using a pre-control the following rules should be used.

xp
op en ite 426 sed

•
ro ed E

Two data points in the green zone – continue to run the process.
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

• One data point in the green zone and one data point in the yellow
en

zone – continue to run the process.

an is d ume be t is
5

tec
Th Doc Num men

• Two yellow points in a row (same zone) – adjust the process

o
s c um S

• Two yellow points in a row (opposite zone) – stop the process

E- der ocu

and investigate
gh
Or is d

• One red data point – stop the process and investigate.

Th

Every time the process is adjusted, before the sampling can start the
process must produce 5 consecutive parts in the green zone.

Pre-control is not a process control chart but a nonconformance control

chart so great care must be taken as to how this chart is used and
interpreted. Pre-control charts should be not used when you have a Cp,
Cpk greater than one or a loss function that is not flat within the
specifications (see Chapter IV).

105
CHAPTER III
Other Types of Control Charts

Loss Functions

LSL USL LSL USL

Target Target
“Flat” Loss Function Sensitive Loss Function

din p
.
or ou
gly
cc Gr
The benefit of pre-control is its simplicity. The disadvantage of pre-

Inc

d a ion
control is that potential diagnostics that are available with normal process

rke Act
al
control methods are not available. Further, pre-control does not evaluate

on
nor monitor process stability. Pre-control is a compliance based tool not

ma ry
ati
a process control tool.

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

106
 CHAPTER III
Other Types of Control Charts

Short-Run Control Charts

Standard control chart approaches are well suited for long production
runs. However there are processes that only produce a small number of
products during a single run (e.g., job shops). Further, the increasing
focus on just-in-time (JIT) inventory and lean manufacturing methods is
driving production runs to become shorter. From a business perspective,
producing large batches of product several times per month and holding
it in inventory for later distribution, can lead to avoidable, unnecessary
costs. Manufacturers now are moving toward JIT – producing much
smaller quantities on a more frequent basis to avoid the costs of holding
“work in process” and inventory. For example, in the past, it may have

din p
been satisfactory to make 10,000 parts per month in batches of 2,500 per

.
or ou
gly
week. Now, customer demand, flexible manufacturing methods and JIT

cc Gr
requirements might lead to making and shipping only 500 parts per day.

Inc

d a ion
To realize the efficiencies of short-run processes it is essential that SPC

rke Act
al
methods be able to verify that the process is truly in statistical control,
on
(i.e., predictable), and be able to detect special-cause variation during

ma ry
ati
these “short runs.”

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Wheeler (1991) describes four requirements for an “Ideal State” of

process operation essential for competing in this arena:
1
aI

3/3
gn

a. "The process must be inherently stable over time.

an e A n:
Ma

as m

b. The process must be operated in a stable and consistent manner.

o
o
i
d h ut
t
to:

ted y t ira

c. The process aim must be set and maintained at the proper level.
xp
op en ite 426 sed

d. The Natural Process Limits must fall within the specification limits."
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Effective control charts can be constructed even with small amounts of

an is d ume be t is
5

data. Short-run oriented charts allow a single chart to be used for the
tec
Th Doc Num men

control of multiple products. There are a number of variations on this

o

theme. Among the more widely described short-run charts are: 29

s c um S
E- der ocu

gh

a. Difference or Deviation from Nominal (DNOM) X & R chart.

Or is d

Production processes for short runs of different products can be

Th

characterized easily on a single chart by plotting the differences

between the product measurement and its target value. These charts
can be applied both to individual measurements and to grouped data.

29
Caution should be used when subgroups are formed from small populations or when the subgroups
use measurements taken over extended periods of time (see Appendix A). Wheeler (1991) discusses
evaluating the data with an Individuals and Moving Range (I & MR) chart to ensure that important
process behavior information is not being masked by the subgrouping.

107
CHAPTER III
Other Types of Control Charts

DNOM Chart for Part Types 760, 822 and 937

Type 822 Type 937 Type 760
UCL=2.748

Sample Mean
_
_
X=0.167
0

-2
LCL=-2.414

1 2 3 4 5 6 7 8 9 10 11 12 13 14
Sample

UCL=6.495
6
Sample Range

din p
.
or ou
gly
_
R=2.523
2

cc Gr
Inc

d a ion
0 LCL=0

1 2 3 4 5 6 7 8 9 10 11 12 13 14

rke Act
Sample

al
on

ma ry
Figure III.4: DNOM Control Chart
ati

ter ust
en e 11
rn

b. Standardized X & R chart.

be otiv /20
wa Ind
nte

1
aI

The DNOM approach assumes a common, constant variance among the

3/3

products being tracked on a single chart. When there are substantial

gn

differences in the variances of these products, using the deviation from

an e A n:
Ma

as m
o

the process target becomes problematic. In such cases the data may be
o
i
d h ut
t
to:

ted y t ira

standardized to compensate for the different product means and

xp

variability using a transformation of the form:

op en ite 426 sed

h
ro ed E

X−µ
n

t p wn se

Z=
yri t is Lic 2
d i oc nt r: 4 lice

σ
en
an is d ume be t is
5

This class of charts sometimes is referred to as Z or Zed charts.

tec
Th Doc Num men

In some short-run processes, the total production volume may be too

s c um S

small to utilize subgrouping effectively. In these cases subgrouping

E- der ocu

measurements may work counter to the concept of controlling the

gh
Or is d

process and reduce the control chart to a report card function. But when
subgrouping is possible, the measurements can be standardized to
Th

accommodate this case.

c. Standardized Attributes Control Charts.
Attributes data samples, including those of variable size, can be
standardized so that multiple part types can be plotted on a single chart.
The standardized statistic has the form:
Difference from Mean
Zi =
Standard Deviation
For example, a u statistic for defect rate would be standardized as:
u −u
Zi = i
u /n
This method also applies to np, p, c and u charts.

108
 CHAPTER III
Other Types of Control Charts

See Farnum (1992), Juran and Godfrey (1999), Montgomery (1997),

Wheeler (1991) and Wise and Fair (1998) for detailed discussions and
examples of short-run applications.

Charts for Detecting Small

Changes
There are situations where small changes in the process mean can cause
problems. Shewhart control charts may not be sensitive enough to
efficiently detect these changes, e.g., less than 1.5 σ . The two
alternative charts discussed here were developed to improve sensitivity
for detecting small excursions in the process mean. While the typical

din p
Shewhart chart uses only the information supplied by the most recent

.
or ou
gly
datum point, the Cumulative Sum (CUSUM) and the Exponentially

cc Gr
Weighted Moving-Average (EWMA) charts exploit the information

Inc

d a ion
available in accumulated, historical data. See Montgomery (1997),
Wheeler (1995) and Grant and Leavenworth (1996) for in-depth

rke Act
al
on
discussions of these methods and comparisons with the supplemental

ma ry
detection rules for enhancing the sensitivity of the Shewhart chart to
ati

ter ust
en e 11
small process shifts
rn

be otiv /20
wa Ind
nte

1
aI

CUSUM (Cumulative Sum) Chart

3/3
gn

A CUSUM chart plots the cumulative sum of deviations of successive

an e A n:
Ma

as m
o

sample means from a target specification so that even minor permanent

o
i
d h ut
t
to:

shifts (0.5 sigma or below) in the process mean will eventually signal
ted y t ira

that a shift has occurred. For larger shifts, Shewhart control charts are
xp
op en ite 426 sed

just as effective and take less effort.

ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

These charts are most often used to monitor continuous processes, such
b
en

as in the chemical industry, where small shifts can have significant

an is d ume be t is
5

effects.
tec
Th Doc Num men

Vmask Chart for Coating Thickness

s c um S
E- Sumder ocu

0.0 Target=0
gh
Or is d

-0.2
Th

-0.4
Cumulative

-0.6

-0.8

-1.0

-1.2

-1.4
1 5 10 15 20 23 25 30 35
Sample

Figure III.5: CUSUM Chart with V-Mask

109
 CHAPTER III
Other Types of Control Charts

The CUSUM chart evaluates the slope of the plotted line. A graphical
tool (V-mask) is laid over the chart with a vertical reference line offset
from origin of the V passing through the last plotted point (see Figure
III.5). The offset and angle of the arms are functions of the desired level
of sensitivity to process shifts. An out-of-control
Vmask Chart for Coating Thickness
condition (e.g., a significant process shift) is
indicated when previously plotted points fall
0.00 Target=0 outside of the V-mask arms. These arms take the
place of the upper and lower control limits.
-0.25
Cumulative Sum

-0.50

The chart in Figure III.5 indicates that a process

-0.75
shift occurred around the time of sample 14 or 15.
-1.00 Due to the nature of this chart, the shift was not

din p
.
detected until sample 23 was plotted. When the

or ou
gly
-1.25
V-mask was positioned on prior data points, all

cc Gr
1 5 10 15 20 21 25 30 35
samples fell within the control limits, so there was

Inc

d a ion
Sample
no indication of an out-of-control situation.

rke Act
al
In comparison, an Individual and Moving Range (X, MR) plot of the
on
same data (Figure III.6) does not detect the process shift until sample 27.

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

I/mR Chart for Coating Thickness

1
aI

3/3
gn

U C L=1.0536
5
an e A n:
Ma

as m

6
1.02
o
Individual V alue

o
i
d h ut
t

_
to:

ted y t ira

X=0.9702
0.96
xp
op en ite 426 sed

h
ro ed E
n

0.90
t p wn se
yri t is Lic 2

LC L=0.8869
d i oc nt r: 4 lice

b
en

1
0.84 1
an is d ume be t is

1
5

1 5 9 13 17 21 25 29 33 37
tec
Th Doc Num men

O bser vation
o
s c um S
E- der ocu

0.12
gh

1
Or is d

U C L=0.1024
M oving Range

Th

0.08

0.04 __
M R=0.0313
2
0.00 LC L=0

1 5 9 13 17 21 25 29 33 37
O bser vation

Figure III.6: X, MR Chart

A tabular CUSUM is an alternative to the V-mask approach. See
Montgomery (1997) for a discussion of this procedure.

110
 CHAPTER III
Other Types of Control Charts

EWMA (Exponentially Weighted Moving Average)

Charts
An EWMA Chart plots moving averages of past and current data in
which the values being averaged are assigned weights that decrease
exponentially from the present into the past.30 Consequently, the average
values are influenced more by recent process performance.31 The
exponentially weighted moving average is defined by the equation:
zt = λ xt + (1 − λ )zt −1
where λ is the weighting constant 0 < λ < 1,
t is an index number (t = 1....),

din p
.
or ou
gly
x t is the current sample value, and

cc Gr
zt is the current weighted moving average.

Inc

d a ion
An initial value, z0 must be estimated to start the process with the first sample.

rke Act
al
on

ma ry
ati
Through recursive substitution, successive values of Z t can be

ter ust
en e 11
rn

determined from the equation:

be otiv /20
wa Ind
nte

t −1
1
aI

zt = λ ∑ (1 − λ ) xt −i + (1 − λ ) z0
i i
for 0 < λ < 1
3/3
gn

i =0
an e A n:
Ma

as m

The value of λ is determined from tables or graphs based on Average

o
o
i
d h ut
t

Run Length (ARL) performance. Some authors also consider control

to:

ted y t ira

limit widths other than three-sigma when designing an EWMA chart.

xp
op en ite 426 sed

But, current literature indicates that this approach may not necessary.
h
ro ed E
n

The EWMA chart becomes an X chart for λ = 1.0. See Montgomery

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

(1997) and Wheeler (1995) for detailed discussions.

en
an is d ume be t is
5

The advantage of this chart is its ability to efficiently detect small

tec
Th Doc Num men

process mean shifts, typically less then 1.5 sigma, and it can be used with
o

an autocorrelated process32 with a slowly drifting mean.

s c um S
E- der ocu

Its disadvantage is its inability to efficiently detect large changes in the

gh
Or is d

process mean. In situations where large process mean shifts are

expected, the Shewhart control chart is recommended.
Th

A common use of the EWMA is in the chemical industry where large

day-to-day fluctuations are common but may not be indicative of the lack
of process predictability.
Figures III.7 and III.8 are EWMA and X, MR plots of the same data. The
EWMA chart detects a mean shift at sample 29, but there is no indication
of this shift on the X, MR chart

30
In contrast, the CUSUM chart gives equal weight to the previous data.
31
Another type of time weighted control chart is the Moving Average chart (MA chart). This approach
is based on a simple, unweighted moving average. See Montgomery (1997).
32
See Appendix A.

111
CHAPTER III
Other Types of Control Charts

Valid signals occur only in the form of points beyond the control limits.33
Other rules used to evaluate the data for non-random patterns (see
Chapter II, Section B) are not reliable indicators of out-of-control
conditions.

EWMA Chart of Viscosity

80.8
UCL=80.688
80.6

80.4

80.2
_
_
EWMA

80.0 X=80

din p
.
or ou
gly
79.8

cc Gr
79.6

Inc

d a ion
79.4

rke Act
al
LCL=79.312

on
79.2

ma ry
1 4
ati 7 10 13 16 19 22 25 28 30

ter ust
Sample

en e 11
rn

be otiv /20
wa Ind
nte

Figure III.7: EWMA Chart of Viscosity

1
aI

3/3
gn

an e A n:
Ma

as m

X, MR Chart of Viscosity
o
o
i
d h ut
t
to:

UCL=83
ted y t ira

82.5
xp
Individual Value

op en ite 426 sed

81.0
h

_
ro ed E

X=80
n

79.5
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

78.0
en

LCL=77
an is d ume be t is
5

1 4 7 10 13 16 19 22 25 28
tec

Observation
Th Doc Num men

4
s c um S

UCL=3.686
E- der ocu

3
Moving Range

gh
Or is d

2
__
1 MR=1.128
Th

0 LCL=0

1 4 7 10 13 16 19 22 25 28 30
Observation

Figure III.8: X, MR Chart of Viscosity

EWMA and CUSUM essentially are equivalent in their ability to detect
the presence of assignable causes that result in a small shift in the mean.
However, the EWMA also can be used to forecast a “new” process mean
for the next time period. These charts can be useful to signal a need to

33
Because moving averages are involved, the points being plotted are correlated (dependent) and
therefore detection of special causes using pattern analysis is not appropriate since they assume
independence among the points.

112
 CHAPTER III
Other Types of Control Charts

adjust (maintain) a process. But they are not appropriate as tools for
process improvement (see Wheeler (1995)).
Multivariate forms of these charts, MCUSUM and MEWMA, have been
developed. See Lowery et al. (1992) and Lowry and Montgomery
(1995).

Non-Normal Charts
If the underlying distribution of a process is known to be non-normal,
there are several approaches that can be used:

din p
• Use the standard Shewhart control charts with appropriate sample

.
or ou
gly
size.

cc Gr
• Use adjustment factors to modify the control limits to reflect the non-

Inc

d a ion
normal form.

rke Act
al
on
• Use a transformation to convert the data into a (near) normal form

ma ry
ati
and use the standard charts.

ter ust
en e 11
rn

• Use control limits based on the native non-normal form.

be otiv /20
wa Ind
nte

1
aI

The approach which is used depends on the amount the process

3/3

distribution deviates from normality and specific conditions related to the

gn

process.
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Shewhart Control Charts

xp
op en ite 426 sed

h
ro ed E

Although the sensitivity and risks associated with the standard control
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

charts have been analyzed by assuming the process distribution was

b
en

normal, Shewhart’s development was not based on an assumption of

an is d ume be t is

normality. His goal was to develop a tool useful for the economic control
5

tec
Th Doc Num men

of quality. Shewhart control charts can be used for all processes.

o

However, as the process distribution deviates from normality, the

s c um S
E- der ocu

sensitivity to change decreases, and the risk associated with the Type I
Sample Mean

gh

error increases.
Or is d
Th

Subg
roup 0 10 20

For many non-normal process distributions, the Central Limit Theorem

can be used to mitigate the effect of non-normality. That is, if a
Sample Range

sufficiently large subgroup size is used,34 the Shewhart control chart can
be used with near normal sensitivity and degree of risk.

34
For example, see Wheeler (1995).

113
CHAPTER III
Other Types of Control Charts

Central Limit Theorem

Let X 1 ,… , X n be a set of n independent random variates from the same
arbitrary probability distribution P ( x1 ,… , xn ) with mean µ X and
variance σ X .
n

∑X i
Consider the average Xn = i =1
.
n
 σX 
The distribution of X n approaches the normal distribution N  µ X , 

din p
 n 

.
or ou
gly
as n → ∞

cc Gr
Inc

d a ion
rke Act
al
on
The “rule of thumb” is that the range chart should be used with

ma ry
subgroups of size fifteen or less. The standard deviation chart can be
ati

ter ust
en e 11
used for all subgroup sizes.
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

Adjustment Factors
an e A n:
Ma

as m
o
o

When a large subgroup size is not possible, the control limits of the
i
d h ut
t
to:

ted y t ira

Shewhart control charts can be modified using adjustment factors to

xp
op en ite 426 sed

compensate for the effect of the non-normality. Since non-normal

h
ro ed E

distributions are either asymmetric, have heavier tails than the normal
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

distribution, or both, use of the standard ± 3 sigma control limits can

b
en

increase the risk of false alarms, especially if pattern analysis for special
an is d ume be t is
5

causes is used.
tec
Th Doc Num men

In this approach the non-normal distributional form is characterized by

s c um S

its skewness or kurtosis or both. Tabled or algorithmic correction factors

E- der ocu

are then applied to the normal control limits.35

gh
Or is d
Th

UCL

LCL

S 0 10 20 10 20

This approach requires an initial capability study with a sample size

sufficiently large to effectively capture the non-normal form.

35
For example see: Burr, I. W., (1967), Chan, L.K., and Cui, Heng, J, and (2003) Pham, H., (2001).

114
 CHAPTER III
Other Types of Control Charts

For this and the following approaches, the process should be studied
periodically to verify that the distributional form has not changed. Any
significant change in the distribution is an indicator that the process is
being affected by special causes.

Transformations
An alternative to the adjustment factors is to convert the data instead of
the control limits. In this approach, a transformation is determined which
transforms the non-normal process distribution into a (near) normal
distribution. Examples of transformations36 used in these situations are
the Johnson family of transformations and the Box-Cox transformations.

din p
.
The selected transformation is then used to transform each datum point

or ou
gly
and the standard Shewhart control chart methodologies are used on the

cc Gr
converted data.

Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

For this approach to be effective, the transformation must be valid. This

h
ro ed E

typically requires a capability study with a sample size sufficiently large

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

to effectively capture the non-normal form. Also, because the

en

transformations tend to be mathematically complex, this approach is only

an is d ume be t is
5

effective and efficient when implemented using a computer program.

tec
Th Doc Num men

o
s c um S

Non-Normal Form
E- der ocu

gh
Or is d

There are situations when the above approaches are not easily handled.
Examples of these situations occur when the process distribution is
Th

highly non-normal and the sample size cannot be large, e.g., when
tracking equipment reliability. In these situations a control chart can be
developed using the non-normal form directly to calculate the chart
control limits.

36
For example, see Johnson (1949) and Box and Cox (1964).

115
CHAPTER III
Other Types of Control Charts

UCL

Mean

LCL

0 10 20

In the example of tracking equipment reliability, a Time-to-Failure Chart

with a subgroup size of one can be used. The control limits are based on

din p
.
the exponential distribution with parameter θ equal to the mean time

or ou
gly
between failures (MTBF). In general, control limits for this approach are

cc Gr
selected to be the 0.135 and 99.865 percentile points of the underlying

Inc

d a ion
distribution.

rke Act
al
Like the other approaches above, for this approach to be effective, it
on

ma ry
typically requires a capability study with a sample size sufficiently large
ati

ter ust
en e 11
to capture the non-normal form. Advantages of this approach are that the
rn

be otiv /20
wa Ind
data can be plotted without complex calculations and it provides more
nte

exact control limits than adjustment factors.

1
aI

3/3
gn

an e A n:

Multivariate
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Multivariate charts are appropriate when it is desired to simultaneously

xp
op en ite 426 sed

control two or more related characteristics that influence the performance

h
ro ed E
n

of a process or product. Their advantage is that the combined effect of

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

all variables can be monitored using a single statistic. For

en

instance, the combined effects of pH and temperature of a

an is d ume be t is
5

part washing fluid may be linked to part cleanliness measured

tec
Th Doc Num men

by particle count. A multivariate chart provides a means to

s c um S

detect shifts in the mean and changes in the parameter

E- der ocu

relationships.
gh
Or is d

A correlation matrix of variables can be used to test whether

Th

a multivariate control chart could be useful. For the

multivariate approach to be viable the matrix entries should
indicate that the variables are sufficiently correlated.
Three of the most popular multivariate control chart statistics
are Hotelling’s T2, the Multivariate Exponentially-Weighted
Moving Average (MEWMA) and the Multivariate
Cumulative Sum (MCUSUM).
A multivariate chart reduces Type I error, i.e., false out-of-
control signals are less likely to occur compared to using
univariate charts to make decisions separately for each
variable.
The simplicity of this approach is also its disadvantage. An
out-of-control condition can be detected using a single

116
 CHAPTER III
Other Types of Control Charts

statistic but the analysis of the charted results may not tell which variable
caused it. Additional analysis using other statistical tools may be
required to isolate the special cause(s). See Kourti and MacGregor
(1996).
Multivariate charts are mathematically complex, and computerized
implementation of these methods is essential for practical application. It
is important, however, that the use of appropriate techniques for
estimating dispersion statistics be verified. See Wheeler (1995),
Montgomery (1997) and current literature such as Mason and Young
(2001), for detailed discussions of multivariate control charts.

Other Charts

din p
.
or ou
gly
cc Gr
Inc

d a ion
In Chapter I, Section E, a Case 3 process was defined as one not in

rke Act
al
statistical control but acceptable to tolerance. Special causes of variation
on
are present, the source of variation is known and predictable but may not

ma ry
ati
be eliminated for economic reasons. However, this predictability of the

ter ust
en e 11
rn

special cause may require monitoring and control. One method to

be otiv /20
wa Ind
nte

determine deviations in the predictability of special cause variation is the

1
aI

Regression chart.
3/3
gn

Regression Control Charts

an e A n:
Ma

as m
o
o
i
d h ut
t

Regression charts are used to monitor the relationship between two

to:

ted y t ira

correlated variables in order to determine if and when deviation from the

xp
op en ite 426 sed

known predictable relationship occurs. These charts originally were

h
ro ed E
n

applied to administrative processes but they have also been used to

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

analyze the correlation between many types of variables.

en
an is d ume be t is

Regression charts track the linear correlation between two variables, for
5

tec
Th Doc Num men

example:
o
s c um S

• Product cost versus weight.

E- der ocu

gh

• Throughput versus machine cycle time (line speed).

Or is d

• Temperature versus pressure.

Th

• Dimensional change relative to tooling cycles.

For example, if a tool has constant wear relative to each cycle of the
process, a dimensional feature such as diameter (Y) could be predicted
based on the cycles (X) performed. Using data collected over time this
linear relationship can be modeled as

Y = b0 + b1 X

When X equals zero cycles, the predicted Y is equal to b0. So b0 is the

predicted dimension from a tool never used.

117
CHAPTER III
Other Types of Control Charts

b0 and b1 are estimated using the equations for simple linear regression.
The chart is constructed by drawing the line Yˆ = βˆ0 + βˆ1 X which is the
estimate for Y = b0 + b1 X and computing the 95% or 99% predictive
interval. The predictive limits computed are curved lines with the
tightest point at X . Often they are replaced with the Yˆ ± 3s in order to
tighten the control limits at each extreme for X .

Points that exceed the control limits indicate tooling which has a tool life
which is significantly different from the base tool life. This can be
advantageous or detrimental depending on the specific situation.

din p
.
or ou
gly
cc Gr
A line is only one type of correlation between variables. Regression

Inc
charts can be applied to any relationship for which the mathematical

d a ion
model can be determined.

rke Act
al
on

ma ry
ati
Care should be taken in making predictions (extrapolating) outside of the

ter ust
en e 11
range of the original observations. The accuracy of the regression model
rn

be otiv /20
wa Ind
for use outside of this range should be viewed as highly suspect. Both
nte

the prediction interval for future values and the confidence interval for the
1
aI

regression equation become increasingly wide. Additional data may be

3/3
gn

needed for model validation.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Discussion on confidence intervals can be found in Hines and

xp
op en ite 426 sed

Montgomery (1980).
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Residual Charts
an is d ume be t is
5

tec
Th Doc Num men

An alternative approach to the Regression Chart is to chart the residual

o

values. From the regression equation, the residual value (ε ) is Y − Yˆ .

s c um S
E- der ocu

gh

A chart of the residual values could be treated in the same manner as an

Or is d

Individuals chart with X equal to zero.

Th

The Residuals Chart and the Regression Chart are technically equivalent
and differ only in their presentation.

This approach would be more useful and intuitive when the variable
relationships are more complex.

Autoregressive Charts
Control chart methods generally assume that the data output from a
process are independent and identically distributed. For many processes
this assumption is not correct. Data from a time series, data taken

118
 CHAPTER III
Other Types of Control Charts

sequentially in time, are often serially dependent. These types of

processes have output that are autocorrelated and analysis with standard
charting methods may result in erroneous conclusions.
One common approach to contend with serial dependency is to take
samples far enough apart in time that the dependency is not apparent.
This often works in practice but its effectiveness relies on postponing
sample collection and may extend the sampling interval longer than is
appropriate. Also, this approach ignores information which may be
useful or even necessary for accurate prediction in order to utilize
techniques which were not designed for that type of data.
Autoregressive Moving Average (ARMA) is a method of analysis for
data from a time series and it results in the prediction of future
observations based on their dependency on past observations. Processes

din p
.
which drift, walk or cycle through time are good candidates for time

or ou
gly
series analysis and an ARMA method may be appropriate.

cc Gr
Inc
The autoregressive (AR) model is defined by

d a ion
rke Act
al
X i = ξ + φ1 X i −1 + φ2 X i −2 … + ε i (AR)
on

ma ry
ati
The current value observed is equal to a constant, a weighted

ter ust
en e 11
rn

combination of prior observations and a random component.

be otiv /20
wa Ind
nte

The moving average (MA) model is defined by

1
aI

3/3

X i = µ − θ1ε i −1 −θ 2ε i −2 … + ε i (MA)
gn

an e A n:
Ma

as m

The current value observed is equal to a constant, plus a weighted

o
o
i
d h ut
t
to:

combination of prior random adjustments and a random component.

ted y t ira
xp
op en ite 426 sed

The ARMA model is a combination of the AR and MA models.

h
ro ed E
n

X i = ξ + φ1 X i −1 − θ1ε i −1 + φ2 X i −2 − θ 2ε i −2 … + ε i (ARMA)
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Box and Jenkins defined the terminology ARMA(p,d,q) to define subsets

an is d ume be t is
5

tec

of the full ARMA model, where p is the number of autoregressive

Th Doc Num men

parameters, d the number of times the data is differenced (defined

s c um S

below), and q is the number of moving average parameters.

E- der ocu

gh

So, ARMA(1,0,0) is the first order AR model without differencing

Or is d

X i = ξ + φ1 X i −1 + ε i .
Th

ARMA(0,0,1) is the first order MA model without differencing

X i = µ − θ1ε i −1 + ε i .
And ARMA(1,1,1) is the first order ARMA model differenced once
Yi = ξ + φ1Yi −1 − θ1ε i −1 + ε i .
For the first order AR and ARMA models the parameter φ must be in
the interval −1 < φ < 1 for the model to be stationary, i.e., does not
diverge to infinity. (There are similar restrictions for the φ ’s in the
higher order models.) For processes that are not stationary, the data will

119
CHAPTER III
Other Types of Control Charts

need to be differenced. Differencing removes the serial dependence

between an observation and another lagged observation.
Yi = X i − X i − k
The differenced observation is equal to the current observation minus the
observation made k samples prior. The data should only be differenced
if the model is not stationary. Most data from manufacturing processes
will not need differencing. The processes do not diverge to infinity.
The next step is to determine the number of autoregressive and moving
average parameters to include in the model. Typically the number of
φ ’s or θ ’s needed will not be more than two. ARMA(1,d,0),
ARMA(2,d,0), ARMA(1,d,1), ARMA(2,d,1), ARMA(1,d,2),
ARMA(2,d,2), ARMA(0,d,1), ARMA(0,d,2) are the common

din p
.
or ou
gly
combinations and it is feasible to estimate them all before selecting the

cc Gr
best.

Inc

d a ion
To estimate the parameters use Non-Linear Estimation.

rke Act
al
Once the model is determined and stationary, and the parameters are
on
estimated then the next observation can be predicted from past

ma ry
ati
observations. For example (ARMA(1,0,1)):

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Xˆ t = ξ + φ1 X t −1 − θ1ε t −1
1
aI

3/3

The residual is calculated by

gn

an e A n:
Ma

as m

ε t = X t − Xˆ t
o
o
i
d h ut
t
to:

ted y t ira

and the values for ε will be independent normally distributed random

xp
op en ite 426 sed

variables and may be analyzed using an Individuals Chart or Residuals

h
ro ed E
n

Chart. For a more complete discussion see Box, Jenkins and Reinsel
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

(1994).
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

120
 CHAPTER III
Other Types of Control Charts

Zone Charts

Chapter II, Section B, Table II.1 provides various rules for detecting out-
of-control signals. The first four rules can be easily implemented with
manual control charts, but the latter rules do not lend themselves to rapid
visual identification since they require the determination of the number
of standard deviations a plotted point is from the centerline. This can be
aided by dividing the control chart into “zones” at 1, 2, and 3 standard
deviations from the centerline.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

These zones are sometimes referred to as “sigma” zones (sigma here is

an e A n:
Ma

as m
o

the standard deviation of the distribution of the sample averages, not the
o
i
d h ut
t
to:

ted y t ira

individual values). The zones assist in the visual determination of

xp

whether a special cause exists using one or more of the tabled criteria.
op en ite 426 sed

See Montgomery (1997) and Wheeler (1995).

ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

This division of the control chart can be coupled with run sums analysis
of control chart to produce the Zone Control Chart. The run sums control
chart analysis was introduced by Roberts (1966) and studied further by
Reynolds (1971). This approach assigns a score to each zone. The score
α i , assigned to the region R+i is nonnegative; and the score, β i assigned
to the region R-i is nonpositive. A typical set of scores are:

121
CHAPTER III
Other Types of Control Charts

Zone Score

 µ X , µ X + σ X ) 0 or +1

 µ X + σ X , µ X + 2 σ X ) +2

 µ X + 2σ X , µ X + 3σ X ) +4

 µ X + 3σ X , ∞ ) +8

The four regions placed symmetrically below the centerline are assigned

din p
.
the corresponding negative scores.

or ou
gly
cc Gr
A zone control chart is a hybrid between an X (or Individuals) chart

Inc

d a ion
and a CUSUM chart. It analyzes a cumulative score, based on the zones.
The cumulative score is the absolute value of the sum of the scores of the

rke Act
al
on
zones in which the points are plotted. Every time the centerline is crossed

ma ry
the cumulative score is reset to zero.
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d

A point is out of control if its cumulative score is greater than or equal to

Th

8. Thus, the analyst does not need to recognize the patterns associated
with non-random behavior as on a Shewhart chart. With the scoring of 0,
2, 4, 8 this method is equivalent to the standard criteria 1, 5, and 6 for
special causes in an X (or Individuals) chart and is more stringent than
criterion 8. With the scoring of 1, 2, 4, 8 this method is equivalent to the
standard criteria 1, 2, 5, and 6 for special causes in an X (or Individuals)
chart and is more stringent than criteria 7 and 8. As shown in the figure
above, trends (criterion 3) can also be detected depending on the start
and stop of the trend.
Zone control charts can be modified to eliminate the point-plotting
process; the points are plotted in the zone not to a scale. Thus, one
standard zone control chart can fit most needs; when to act on a process
is determined by the charting procedure.

122
 CHAPTER III
Other Types of Control Charts

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

The zone chart can be used with a weighting scheme to provide the
xp
op en ite 426 sed

sensitivity needed for a specific process. For example, one set of weights
h
ro ed E

(scores) can be used during the initial phase for detecting special causes.
n

t p wn se
yri t is Lic 2

Then the weights could be changed when the process is in control and it
d i oc nt r: 4 lice

b
en

is more important to detect drift.

an is d ume be t is
5

The efficiency of the zone control chart is demonstrated by comparing its

tec
Th Doc Num men

average run lengths with those of standard control tests. For the chart
s c um S

divided into scores of 0, 2, 4, and 8, the zone control chart performs as

E- der ocu

well as or better than Shewhart charts (see Davis et al. (1990)).

gh
Or is d
Th

123
 CHAPTER III

Th
Or is d
Other Types of Control Charts

E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

124
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

125
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
CHAPTER IV

wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
Understanding Process Capability

gly
.
and Process Performance for Variables Data
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

126
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER IV
Understanding Process Capability and Process Performance for Variables Data

Introduction
The output of a stable process can be described by its statistical
distribution. The process must be stable (in statistical control) in order
for the distribution to be useful for predicting future results.37 A
distribution is described in terms of characteristics (statistics) that are
calculated from measurements of samples taken from the process.
The statistics of most frequent interest are estimates of distribution
location (or center) and spread relative to the customer requirements.
Typically, the location is estimated by the sample mean or sample
median. Spread usually is estimated using the sample range or sample
standard deviation.

din p
Process centering and spread interact with respect to producing an

.
or ou
gly
acceptable product. As the distribution moves off center, the “elbow

cc Gr
room” available to accommodate process variation (spread) is reduced.

Inc

d a ion
A shift in process location, an increase in process spread or a
combination of these factors may produce parts outside the specification

rke Act
al
limits. A process with such a distribution would not be qualified to meet
on

ma ry
the customer’s needs.
ati

ter ust
en e 11
rn

This section addresses some of the techniques for evaluating process

be otiv /20
wa Ind
nte

capability and performance with respect to product specifications. In

1
aI

general, it is necessary that the process being evaluated be stable (in

3/3
gn

statistical control). A discussion of process variation and the associated

an e A n:

capability indices has little value for unstable processes. However,

Ma

as m
o
o

reasonable approaches have been developed to assess the capability of

i
d h ut
t
to:

ted y t ira

processes exhibiting systematic special causes of process variation, such

xp

as tool wear (see Spiring, F. A. (1991)).

op en ite 426 sed

h
ro ed E

In addition, it is generally assumed that the individual readings from the

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

subject processes have a distribution that is approximately normal.38

en

This section will discuss only the more popular indices and ratios:
an is d ume be t is
5

tec

• Indices of process variation-only, relative to specifications: Cp, and

Th Doc Num men

Pp .
s c um S
E- der ocu

• Indices of process variation and centering combined, relative to

gh
Or is d

specifications: Cpk, and Ppk.

Th

• Ratios of process variation-only, relative to specifications: CR and

PR.

NOTE: Although other indices are not discussed in this manual, see
Appendix D and References for information on other indices.

Finally, this section describes the conditions and assumptions associated

with these process measures and concludes with a suggestion as to how
these measures might be applied toward enhancing process
understanding within the framework of continual process improvement.

37
See Chapter I, Sections C-F.
38
For non-normal distributions and autocorrelated processes see Chapter IV, Section B.

127
CHAPTER IV
Understanding Process Capability and Process Performance for Variables Data

This manual recognizes both the misunderstanding and controversy

surrounding the fundamental concepts and definitions regarding process
“Control”, “Capability”, and “Performance”. It is not the purpose of this
manual to fully resolve these issues, but to expose and discuss them to an
extent that allows each reader the opportunity to develop a better
understanding of them in order to provide value and knowledge for
continual process improvement.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

128
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

129
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Understanding Process Capability and Process Performance for Variables Data
CHAPTER IV
 Th
CHAPTER IV - Section A
Definition of Process Terms

Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

130
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly

Figure IV.1: Within- and Between-Subgroup Variation

.
 CHAPTER IV - Section A
Definition of Process Terms

CHAPTER IV - Section A
Definitions of Process Terms

Process variation has various aspects:

• Inherent Process Variation – That portion of process variation due to
common (systematic) causes only.
• Within-subgroup Variation ( σ C ) – This is the variation due only to
the variation within the subgroups. If the process is in statistical
control this variation is a good estimate of the inherent process

din p
.
or ou
gly
R s
variation. It can be estimated from control charts by d 2 or c4 .

cc Gr
Inc

d a ion
• Between-subgroup Variation – This is the variation due to the

rke Act
al
variation between subgroups. If the process is in statistical control

on
this variation should be zero.

ma ry
ati

ter ust
Total Process Variation ( σ P ) – This is the variation due to both
en e 11
•
rn

be otiv /20
wa Ind
nte

within-subgroup and between-subgroup variation. If the process is

not in statistical control the total process variation will include the
1
aI

3/3

effect of the special cause(s) as well as the common causes. This

gn

variation may be estimated by s, the sample standard deviation, using

an e A n:
Ma

as m
o

all of the individual readings obtained from either a detailed control

o
i
d h ut
t
to:

ted y t ira

n
( xi − x ) 2
chart or a process study: σ P = s = ∑i n − 1 where xi is an
xp
op en ite 426 sed

h
ro ed E
n

t p wn se

individual reading, x is the average of the individual readings, and

yri t is Lic 2
d i oc nt r: 4 lice

b
en

n is the total number of individual readings.

an is d ume be t is
5

• Process Capability — The 6 σˆ range of inherent process variation,

tec
Th Doc Num men

for statistically stable processes only, where σˆ is usually estimated

s c um S
E- der ocu

R s
by d 2 or c4 .
gh
Or is d

• Process Performance — The 6 σˆ range of total process variation,

Th

where σˆ is usually estimated by s , the total process standard

deviation.

If the process is in statistical control the process capability will be very

close to the process performance. A large difference between the
capability and performance σˆ indicates the presence of a special
cause(s).

131
CHAPTER IV - Section A
Definition of Process Terms

Process Measures for Predictable Processes

Indices –
Bilateral Tolerances
This section discusses commonly used indices where the specification
has both an upper and lower limit.39

CAUTION: The indices discussed below are valid only when the process
is stable (in statistical control). If the process is not in statistical control
then these indices can be very misleading, as can be seen by Figure

din p
IV.4.

.
or ou
gly
cc Gr
Cp: This is a capability index. It compares the process capability to the

Inc
Cp

d a ion
maximum allowable variation as indicated by the tolerance. This index

rke Act
al
provides a measure of how well the process will satisfy the variability
on
USL − LSL USL − LSL

ma ry
requirements. Cp is calculated by C p = =
( )
ati

ter ust
6σ C
en e 11
6 R
rn

be otiv /20 d2
wa Ind
nte

1
aI

Cp is not impacted by the process location. This index can be calculated

3/3
gn

only for two-sided (bilateral) tolerances.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Cpk: This is a capability index. It takes the process location as well as the
Cpk
xp
op en ite 426 sed

capability into account. For bilateral tolerances Cpk will always be less
h
ro ed E

than or equal to Cp.

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

C pk ≤ C p
en
an is d ume be t is
5

tec

Cpk will be equal to Cp only if the process is centered.

Th Doc Num men

Cpk is calculated as the as the minimum of CPU or CPL where:

s c um S
E- der ocu

USL − X USL − X
gh

CPU = =
Or is d

( )
and
3σ C 3 R
Th

d2

X − LSL X − LSL
CPL = =
3σ C 3 R
d2( )
Cpk and Cp should always be evaluated and analyzed together. A Cp value
significantly greater than the corresponding Cpk indicates an opportunity
for improvement by centering the process.

39
As discussed in Chapter II, Section A, process analysis requires that the data have been collected
using measurement system(s) that are consistent with the process and have acceptable measurement
system characteristics.

132
 CHAPTER IV - Section A
Definition of Process Terms

Pp: This is a performance index. It compares the process performance to

Pp the maximum allowable variation as indicated by the tolerance. This
index provides a measure of how well the process will satisfy the
variability requirements. Pp is calculated by

USL − LSL USL − LSL

Pp = =
6σ P 6s

Pp is not impacted by the process location.

din p
.
or ou
gly
Ppk

cc Gr
Ppk: This is a performance index. It takes the process location as well as
the performance into account. For bilateral tolerances Ppk will always be

Inc

d a ion
less than or equal to Pp. Ppk will be equal to Pp only if the process is

rke Act
al
centered.
on

ma ry
ati
Ppk ≤ Pp

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Ppk is calculated as the as the minimum of PPU or PPL where:

1
aI

3/3

USL − X USL − X
gn

PPU = = and
3σ P 3s
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

X − LSL X − LSL
PPL = =
xp
op en ite 426 sed

3σ P 3s
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Figure IV.2: Cpk and Ppk Comparison

133
CHAPTER IV - Section A
Definition of Process Terms

Ppk and Pp should always be evaluated and analyzed together. A Pp value

significantly greater than the corresponding Ppk indicates an opportunity
for improvement by centering the process.

If the process is in statistical control the process capability will be very

close to the process performance. A large difference between the C and P
indices indicate the presence of a special cause(s). See Figure IV.3 and
IV.4.

CR CR: This is the capability ratio and is simply the reciprocal of Cp;
1

din p
.
or ou
gly
CR = .
Cp

cc Gr
Inc

d a ion
rke Act
al
PR PR: This is the performance ratio and is simply the reciprocal of Pp;
1 on

ma ry
ati
PR = .

ter ust
en e 11
rn

Pp be otiv /20
wa Ind
nte

1
aI

3/3

NOTE: Example calculations for all of these measures are shown in

gn

Appendix F.
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

PPM A parts-per-million (ppm) nonconformance rate is sometimes used as a

xp
op en ite 426 sed

supplemental measure of process capability. To estimate the

h
ro ed E

nonconformance rate using capability index information, a probability

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

distribution of the data must be defined. While the normal distribution

en

often is used for this purpose, this is an assumption that should be

an is d ume be t is
5

validated using a goodness-of-fit test before proceeding further. The

tec
Th Doc Num men

nonlinear relationship between the capability index and the proportion

s c um S

nonconforming should be understood in order to make correct inferences

E- der ocu

(see Wheeler (1999) for a detailed discussion of this subject).

gh
Or is d
Th

134
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

135
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.

Figure IV.3: Comparison between a Predictable and Immature Process

CHAPTER IV - Section A
Definition of Process Terms
 Th
CHAPTER IV - Section A
Definition of Process Terms

Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

136
an e A n: ati

Process
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.

Figure IV.4: Cpk and Ppk Values Produced by a Predictable and Immature
 CHAPTER IV - Section A
Definition of Process Terms

Indices –
Unilateral Tolerances
This section discusses commonly used indices where the specification
has either an upper or lower limit but not both.

Cp: This is a capability index. It compares the process capability to the

Cp maximum allowable variation as indicated by the tolerance. This index
has no meaning for unilateral tolerances.

din p
If the product characteristic has a physical limit (e.g., flatness cannot be

.
or ou
gly
less than zero), a Cp could be calculated using the physical limit (0.0) as

cc Gr
a surrogate lower limit. But this number will not have the same

Inc

d a ion
relationship to Cpk as it does in the bilateral case.

rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

Cpk: This is a capability index. It takes the process location as well as the
be otiv /20
wa Ind
nte

capability into account. With unilateral tolerances with a physical limit,

Cpk
1
aI

Cpk can be less than, equal to or greater than Cp.

3/3
gn

Cpk is directly related to the proportion nonconforming produced by the

an e A n:
Ma

as m

process. It is equal to CPU or CPL depending whether the tolerance is an

o
o
i
d h ut
t

USL or a LSL where:

to:

ted y t ira
xp
op en ite 426 sed

USL − X
h
ro ed E

CPU =
( )
n

t p wn se

3 Rd
yri t is Lic 2
d i oc nt r: 4 lice

b
en

2
an is d ume be t is
5

tec

X − LSL
Th Doc Num men

CPL =
( )
s c um S

3 Rd
E- der ocu

gh

2
Or is d
Th

Pp: This is a performance index. It compares the process performance to

Pp the maximum allowable variation as indicated by the tolerance. This
index has no meaning for unilateral tolerances.
If the product characteristic has a physical limit (e.g., flatness cannot be
less than zero), a Pp could be calculated using the physical limit (0.0) as a
surrogate lower limit. But this number will not have the same
relationship to Ppk as it does in the bilateral case.

137
CHAPTER IV - Section A
Definition of Process Terms

Ppk is directly related to the proportion nonconforming produced by the

Ppk process. It is equal to PPU or PPL depending whether the tolerance is an
USL or a LSL where:

USL − X
PPU =
3s

X − LSL
PPL =
3s

An alternate notation for Ppk in the case of unilateral tolerances is Ppku or

Ppkl depending on whether the limit is an USL or LSL.

din p
.
or ou
gly
cc Gr
CR: This is the capability ratio and is simply the reciprocal of Cp. As
CR

Inc

d a ion
such, this index has no meaning for unilateral tolerances.

rke Act
al
on

ma ry
PR: This is the performance ratio and is simply the reciprocal of Pp. As
ati

ter ust
PR en e 11
such, this index has no meaning for unilateral tolerances.
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

NOTE: Example calculations for all of these measures are shown in

an e A n:
Ma

as m

Appendix F.
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

138
 CHAPTER IV - Section B
Description of Conditions

CHAPTER IV - Section B
Description of Conditions
It is appropriate to point out that process variation and process centering
are two separate process characteristics. Each needs to be understood
separately from the other. To assist in this analysis it has become
convenient to combine the two characteristics into indices, such as Cp,
Cpk or Pp, Ppk. These indices can be useful for:
• Measuring continual improvement using trends over time.
• Prioritizing the order in which processes will be improved.

din p
.
or ou
The capability index, Cpk , is additionally useful for determining whether

gly
cc Gr
or not a process is capable of meeting customer requirements. This was
the original intent of the capability index. The performance index, Ppk,

Inc

d a ion
shows whether the process performance is actually meeting the customer

rke Act
al
requirements. For these indices (as well as all of the other process
on
measures described in Chapter IV, Section A) to be effectively used, the

ma ry
ati
CONDITIONS which surround them must be understood. If these

ter ust
en e 11
rn

conditions are not met, the measures will have little or no meaning and
be otiv /20
wa Ind
nte

can be misleading in understanding the processes from which they were

1
aI

generated. The following three conditions are the minimum that must be
3/3
gn

satisfied for all of the capability measures described in Section A:

an e A n:
Ma

as m

• The process from which the data come is statistically stable, that is,
o
o
i
d h ut
t

the normally accepted SPC rules must not be violated.

to:

ted y t ira
xp
op en ite 426 sed

• The individual measurements from the process data form an

h
ro ed E

approximately normal distribution.40

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

• The specifications are based on customer requirements.

en
an is d ume be t is
5

tec
Th Doc Num men

Commonly, the computed index (or ratio) value is accepted as the “true”
s c um S

index (or ratio) value; i.e., the influence of sampling variation on the
E- der ocu

computed number is discounted. For example, computed indices Cpk of

gh
Or is d

1.30 and 1.39 can be from the same stable process simply due to
sampling variation.
Th

See Bissell, B.A.F. (1990), Boyles, R. A. (1991) and Dovich, R. A.

(1991) for more on this subject.

40
For non-normal distributions, see the pages that follow.

139
CHAPTER IV - Section B
Description of Conditions

41
Handling Non-Normal and Multivariate Distributions
Although the normal distribution is useful in describing and analyzing a
wide variety of processes, it cannot be used for all processes. Some
processes are inherently non-normal, and their deviations from normality
are such that using the normal distribution as an approximation can lead
to erroneous decisions. Other processes have multiple characteristics that
are interrelated and should be modeled as a multivariate distribution.
Of the indices described above, Cp, Pp, CR, and PR are robust with
respect to non-normality. This is not true for Cpk, and Ppk.

din p
.
or ou
Relationship of Indices and

gly
cc Gr
Proportion Nonconforming

Inc

d a ion
rke Act
al
Although many individuals use the Cpk, and Ppk indices as scalar-less
on
(unit-less) metrics, there is a direct relationship between each index and

ma ry
ati

ter ust
the related process parameter of proportion nonconforming (or ppm).

en e 11
rn

Assuming that Cp > 1, the capability index relationship is given by:

be otiv /20
wa Ind
nte

1
2
aI

 x
− 
3/3

zc
proportion nonconforming = 1 − ∫ e  2
dx
gn

−∞
an e A n:
Ma

as m
o

zc = 3 C pk and
o

where
i
d h ut
t
to:

ted y t ira

C pk = min {CPU , CPL}

xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Similarly, Ppk is related to the performance proportion nonconformance

b
en

through:
an is d ume be t is
5

z p = 3 Ppk
tec
Th Doc Num men

o
s c um S

With this understanding of Cpk, and Ppk, indices for non-normal

E- der ocu

distributions can be developed with the same relationships between the

gh
Or is d

index and the process proportion nonconforming.

Th

The determination of these indices for non-normal distributions requires

extensive tables or the use of iterative approximation techniques. They
are rarely calculated without the assistance of a computer program.

Non-Normal Distributions
Using Transformations
One approach is to transform the non-normal form to one that is (near)
normal. The specifications are also transformed using the same

41
As discussed in Chapter II, Section A, process analysis requires that the data have been collected
using measurement system(s) that are consistent with the process and have acceptable measurement
system characteristics.

140
 CHAPTER IV - Section B
Description of Conditions

parameters. The Cpk, and Ppk indices are then determined in the
transformed space using standard calculations based on the normal
distribution.
Two general transformation approaches which have gained support are:

• Box-Cox Transformations
The methods of analysis of designed experiments are “appropriate
and efficient when the models are (a) structurally adequate, and the
(supposedly independent) errors (b) have constant variance and (c)
are normally distributed.42” Box and Cox (1964) discussed a
transformation which reasonably satisfies all three of these
requirements. This transformation is given by:

din p
.
or ou
gly
w = xλ

cc Gr
Inc
−5 ≤ λ ≤ 5

d a ion
where

rke Act
al
and λ = 0 for the natural log transformation
on

ma ry
ati
λ = 0.5 for the square root transformation

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Although this transformation was developed with the focus of the

analysis of designed experiments, it has found an application in the
1
aI

3/3

transformation of process data to normality.

gn

an e A n:
Ma

as m
o
o
i
d h ut
t

•
to:

Johnson Transformations
ted y t ira
xp
op en ite 426 sed

In 1949, Norman L. Johnson developed a system of transformations

h
ro ed E

which yields approximate normality.43 This system is given by:

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is

 x 
5

SB
w = log 
tec

 Bounded
Th Doc Num men

 (1- x) 
o
s c um S

w = log ( x )
E- der ocu

SL Log Normal
gh
Or is d

SU (
w = sinh -1 ( x ) = log x + 1 + y2 + 1 ) Unbounded
Th

As in the case of the Pearson Family of distributions (see below), this

system of curves encompasses all the possible unimodal distributional
forms; i.e., it covers the entire feasible skewness-kurtosis plane. It also
contains as a boundary form the familiar lognormal distribution.
However, in the general case, the Johnson curves are four parameter
functions.

42
Box, G. E. P., Hunter, W. G., and Hunter, J. S., Statistics for Experimenters, John Wiley and Sons,
New York, 1978, pg.239.
43
See Johnson (1949).

141
CHAPTER IV - Section B
Description of Conditions

Non-Normal Distributions
Using Non-Normal Forms
Non-normal forms model the process distribution and then determine the
proportion nonconforming, i.e., the area of the non-normal distribution
outside the specifications.
A common approach to the modeling of the non-normal distribution is to
use the Pearson Family of Curves. The most appropriate member of this
family is determined by the method of matching moments; i.e., the curve
with skewness (SK) and kurtosis (KU) that match that of the sampled
distribution is used as a model for the underlying form. As in the case of
the Johnson Transformation System (see above), this family of curves

din p
encompasses all the possible unimodal distributional forms; i.e., it covers

.
or ou
gly
the entire feasible SK-KU plane.

cc Gr
To calculate the non-normal equivalent to the Ppk index, the non-normal

Inc

d a ion
form ( f ( x ) ) is used to determine the proportion nonconforming, i.e.,

rke Act
al
the area of the non-normal distribution outside the upper and lower
on

ma ry
specifications:
ati

ter ust
en e 11
rn

LSL
pL = ∫be otiv /20
wa Ind
f ( x ) dx and
nte

−∞
1
aI

3/3

∞
∫
gn

pU = f ( x ) dx .
an e A n:
Ma

USL
as m
o
o

These values are converted to a z value using the inverse standard normal
i
d h ut
t
to:

ted y t ira

distribution. That is, the zL and zU values in the following equations are
xp
op en ite 426 sed

determined such that:

h
ro ed E
n

t p wn se

2
yri t is Lic 2

 x
d i oc nt r: 4 lice

zL − 
pL = ∫  2
en

e dx and
−∞
an is d ume be t is
5

tec
Th Doc Num men

2
 x
o

∞ − 
pU = ∫ e 2
dx
s c um S
E- der ocu

zU
gh

min {z L , zU }
Or is d

Then Ppk =
3
Th

Although the standard calculation of Pp is a robust estimate, a more exact

estimate can be found using the convention that the process spread is defined as
the range that includes 99.73% of the distribution (representing the equivalence
of a ±3σ normal distribution spread). The limits of this range are called the
“0.135% quantile” ( Q0.00135 ) and the “99.865% quantile” ( Q0.99865 ). That is,
0.135% of the values of the population are to be found both below Q0.00135 and
above Q0.99865 .44

44
For the normal form: Q0.99865 = − Q0.00135 = z0.99865 s

142
 CHAPTER IV - Section B
Description of Conditions

Q0.00135
0.00135 = ∫−∞
f ( x) dx and
∞
0.99865 = ∫Q0.99865
f ( x) dx .

The calculation for Pp then is:

specification range specification range
Pp = =
Est 99.73% Range Q0.99865 − Q0.00135
where the non-normal form is used to calculate the quantiles.

din p
( d)

.
or ou
gly
The capability index Cp is calculated as above replacing s with R .

cc Gr
2

Inc

d a ion
Because this approach uses the total variation to calculate the proportion
nonconforming, there is no analogue of a non-normal Cpk available.

rke Act
al
on

ma ry
ati
An alternate approach to calculating Ppk using quantiles is given in some

ter ust
en e 11
rn

documents by: be otiv /20

wa Ind
nte

 USL − X X − LSL 
Ppk = min  ,
1
aI


3/3

 Q0.00865 − X X − Q0.00135 
gn

an e A n:
Ma

as m

This approach does not tie the Ppk index to the proportion
o
o
i

nonconforming. That is, different non-normal forms will have the same
d h ut
t
to:

ted y t ira

index for different proportion nonconforming. To properly interpret and

xp
op en ite 426 sed

compare these indices, the non-normal form as well as the index value
h
ro ed E

should be considered.
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

143
CHAPTER IV - Section B
Description of Conditions

Multivariate Distributions
When multiple characteristics are interrelated, the process distribution
should be modeled using a multivariate form. The process performance
index Ppk can be evaluated by first determining the proportion
nonconforming, i.e., the area of the multivariate distribution outside the
specifications.
For many geometrically dimensioned (GD&T) characteristics, the
bivariate normal form is useful in describing the process.
A pair of random variables X and Y have a bivariate normal distribution
if and only if their joint probability density is given by
 

din p
− z 

.
or ou
( )

gly
1  2 1− ρ 2 
f ( x, y ) = e  

cc Gr
2πσ xσ y 1 − ρ 2

Inc

d a ion
2ρ ( x − µx ) ( y − µ y )  y − µ y 

rke Act
al
2 2
 x − µx 
on
where z =   − +
 σ 

ma ry
 σx  σ xσ y
ati
 

ter ust
y

en e 11
rn

be otiv /20
wa Ind
nte

σ xy
ρ = cov ( x , y ) =
1
aI

σ xσ y
3/3
gn

an e A n:
Ma

−∞ < x < ∞; − ∞ < y < ∞; where σ x > 0; σ y > 0

as m

for
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

To calculate the multivariate equivalent to the Ppk index, the multivariate

h
ro ed E

form ( e. g ., f ( x , y ) ) is used to determine the proportion nonconforming,

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

i.e., the volume of the multivariate distribution outside the specification

an is d ume be t is

(tolerance) zone. In the bivariate case this would be:

5

tec
Th Doc Num men

pz = ∫∫ f ( x, y ) dx dy and
s c um S
E- der ocu

tolerance zone
gh
Or is d

Tolerance Zone
Th

pz

Bivariate Distribution

This value is converted to a z value using the inverse standard normal

distribution. That is, the z value such that:

144
 CHAPTER IV - Section B
Description of Conditions

2
x
z − 
pz = ∫−∞
e  2
dx

z
Then Ppk =
3
An estimate Pp can be found using:
specification area
Pp =
Est 99.73% area
where the multivariate form is used to calculate the estimated 99.73%

din p
area.

.
or ou
gly
Because this approach uses the total variation to calculate the proportion

cc Gr
nonconforming, there is no analogue of a multivariate Cpk available.45

Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

45
See also Bothe (2001) and Wheeler (1995).

145
 Th
CHAPTER IV - Section C

Or is d
E- der ocu
Suggested Use of Process Measures

Th Doc Num men

an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

146
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 CHAPTER IV - Section C
Suggested Use of Process Measures

CHAPTER IV - Section C
Suggested Use of Process Measures
The key to effective use of any process measure continues to be the level
of understanding of what the measure truly represents. Those in the
statistical community who generally oppose how Cpk indices are being
used, are quick to point out that few “real world” processes completely
satisfy all of the conditions, assumptions, and parameters within which
Cpk has been developed (see Gunter, B. (1989) and Herman, J. T. (1989)).
It is the position of this manual that, even when all conditions are met, it
is difficult to assess or truly understand a process on the basis of a single
index or ratio number, for reasons discussed below.

din p
.
or ou
gly
No single index or ratio should be used to describe a process. It is

cc Gr
strongly recommended that all four indices (Cp, Cpk and Pp, Ppk) be

Inc

d a ion
calculated on the same data set. The comparison of the indices among

rke Act
al
themselves can provide insight to potential process issues and aid in

on
measuring and prioritizing improvement over time. For example, low Cp,

ma ry
ati
Cpk values may indicate within-subgroup variability issues, whereas low

ter ust
en e 11
rn

Pp, Ppk may imply overall variability issues.

be otiv /20
wa Ind
nte

Graphical analyses should be used in conjunction with the process

1
aI

measures. Examples of such analyses include control charts, plots of

3/3
gn

process distributions, and loss function graphs.

an e A n:
Ma

as m
o

Additionally, it is helpful to graph the inherent process variation,

o
i
d h ut
t
to:

ted y t ira

6σC 6σ C = 6 R versus total process variation, 6σ P = 6 s , to compare the

d2
xp
op en ite 426 sed

h
ro ed E

process “capability” and “performance” and to track improvement.

6σP
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Generally, the size of this gap is an indication of the effect that special
b
en

causes have on the process. These types of graphical analyses can be

an is d ume be t is

done for better process understanding even if process indices are not
5

tec
Th Doc Num men

used.
o
s c um S

Process measures should be used with the objective of aligning the

E- der ocu

“Voice of the Process” to the “Voice of the Customer”.46

gh
Or is d

All capability and performance assessments should be confined to single

Th

process characteristics. It is never appropriate to combine or average the

capability or performance results for several processes into one index.47

46
See Figure I.1.
47
Methods for addressing multivariate processes are addressed in Chapter IV, Section B.

147
CHAPTER IV - Section C
Suggested Use of Process Measures

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Figure IV.5: “Goal Post” vs. Loss Function

The Loss Function Concept

The driving force behind the use of capability indices (and other process
measures) has been the desire to produce all parts within customer
specifications. The underlying concept motivating this desire is that all
parts within specification, regardless of where they are located within the
specification range, are equally “good” (acceptable), and all parts beyond
specifications, regardless of how far beyond specifications they may be,
are equally “bad” (unacceptable). Quality professionals sometimes refer
to this concept as “Goal Post” mentality (see Figure IV.5(A)).

148
 CHAPTER IV - Section C
Suggested Use of Process Measures

Although this mental model (good/bad) has been extensively used in the
past, it is suggested that a more useful model (i.e., one that is a lot closer
to the behavior of the real world), is illustrated in Figure IV.5(B). In
general, this model is a quadratic form and uses the principle that an
increasing loss is incurred by the customer or society the further a
particular characteristic gets from the specification target. Implicit in
this concept, referred to as the loss function, is the presumption that the
design intent (specification target) is aligned with the customer’s
requirement.

The first step in managing variation is to understand how much variation

is acceptable; i.e., how much deviation from a target or nominal value is
allowable. Traditionally the value judgment of “acceptable” and

din p
.
“allowable” is based on the design engineer’s understanding of the

or ou
gly
functional requirements and the physics of the design and usage

cc Gr
environment (engineering subject matter knowledge), tempered by the

Inc

d a ion
economic constraints of the production process. The results of this part

rke Act
al
of the design process are reflected in the engineering specifications

on
(tolerances).

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

But what do the specifications mean? Ideally, all characteristics of a

tec
Th Doc Num men

design should be equal to the design intent – the target value that would
o

yield perfect results. But variation exists. So what is the difference to the
s c um S
E- der ocu

customer between two different parts, one with a characteristic on target

gh

and one having the same characteristic off target but within
Or is d

specification?
Th

A common approach can be described using the “Goal Post” analogy. In

many sports (e.g., football, soccer, hockey, basketball) a field goal is
awarded if the ball passes through the goal posts (or hoop in basketball).
It doesn’t make a difference if the ball or puck enters dead center or just
slips in. The score awarded is the same.
In manufacturing processes this means that everything within the
specification limits is considered equally good, and everything outside is
equally bad.

149
CHAPTER IV - Section C
Suggested Use of Process Measures

This approach may be valid for discrete characteristics (e.g., the part has
a clearance hole or not), but when dealing with characteristics with a
continuous response, this approach does not reflect how the customer

din p
.
reacts to different levels of the output.

or ou
gly
cc Gr
Without considering the specifications, it is possible to determine the

Inc

d a ion
customer’s sensitivity to deviations from the target (design intent). See
Goble, et al (1981). As a characteristic deviates farther from the target,

rke Act
al
more customers will be able to “sense” that it is different than the design
on

ma ry
intent - primarily because it takes more “effort” to use. In many cases a
ati

ter ust
en e 11
loss (in time, cost, efficiency, etc.) can be associated with each deviation
rn

increment. This loss can apply to the individual customer, but it also may
be otiv /20
wa Ind
nte

extend to the organization, or even to society.

1
aI

3/3

A typical sensitivity curve (loss function) has a quadratic form.

gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

Increasing sensitivity or
h
ro ed E
n

Sensitivity Curve
t p wn se

loss due to deviation

yri t is Lic 2
d i oc nt r: 4 lice

b
en

(Loss Function) from Target

an is d ume be t is
5

tec
Th Doc Num men

Target
s c um S
E- der ocu

There are two ways to analyze a loss function. It can be compared either
gh

to the design intent or functional specifications.

Or is d
Th

Figure IV.6: Comparison of Loss Function and Specifications

150
 CHAPTER IV - Section C
Suggested Use of Process Measures

From the customer’s perspective, Figure IV.4 shows that there is

functionally little difference between a characteristic that is a “nudge” on
one side of the specification limit or the other.

A comparison of the loss function to the specifications provides a way to

classify characteristics. Figure IV.5 shows that the loss function for
Characteristic A is relatively flat within the specification limits. This
means that the customer will be insensitive to variation within
specification for Characteristic A. Since all characteristics are expected
to be within specification, this characteristic satisfies the operational
definition for robust.48

A characteristic is called Robust if the customer is insensitive to the

din p
characteristic’s expected variation.

.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

Figure IV.7: Comparison of Loss Functions

o
s c um S
E- der ocu

gh
Or is d
Th

48
Alternative definition: A design is robust if it is tolerant (insensitive) to variation that is expected
from the manufacturing, processes, materials and environment.

151
CHAPTER IV - Section C
Suggested Use of Process Measures

PROCESS CONTROL SYSTEM MODEL

WITH FEEDBACK

VOICE
OF THE

din p
PROCESS

.
or ou
gly
cc Gr
Inc

d a ion
STATISTICAL

rke Act
al
METHODS
on

ma ry
ati

ter ust
en e 11
rn

PEOPLE be otiv /20

wa Ind
nte

EQUIPMENT THE WAY

1
aI

PRODUCTS
3/3

MATERIAL WE WORK/
gn

OR CUSTOMERS
METHODS BLENDING OF
an e A n:
Ma

as m

SERVICES
o

MEASUREMENT RESOURCES
o
i
d h ut
t
to:

ted y t ira

ENVIRONMENT
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2

IDENTIFYING
d i oc nt r: 4 lice

b
en

CHANGING NEEDS
an is d ume be t is
5

INPUTS PROCESS/SYSTEM OUTPUTS AND EXPECTATIONS

tec
Th Doc Num men

o
s c um S
E- der ocu

VOICE
gh
Or is d

OF THE
Th

CUSTOMER

Figure IV.8: A Process Control System

152
 CHAPTER IV - Section C
Suggested Use of Process Measures

Alignment of Process to
Customer Requirements
In Chapter I, Section B (see Figure IV.8), a process control system is
described as a feedback system. An output characteristic of such a
process can also be expressed graphically in terms of a probability
distribution. This distribution might be referred to as the process
distribution (see Figure IV.9(a)).

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

The sensitivity curve also provides direction in the control of the

be otiv /20
wa Ind
nte

production process. The comparison of the process to the loss function

1
aI

and the specification together shows that the total loss to the customer
3/3
gn

increases as the process center (average) deviates from the target.

an e A n:
Ma

as m

To assess the impact of the process distribution to the customer, a loss

o
o
i
d h ut
t

function (see Figure IV.9(b)) can be established for the process

to:

ted y t ira

characteristic. Superimposing the process distribution on the customer

xp
op en ite 426 sed

requirement loss function curve (see Figure IV.9(c)) shows:

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

• How well the process center is aligned with the customer target
b
en

requirement.
an is d ume be t is
5

tec

• The loss to the customer being generated by this process.

Th Doc Num men

Based upon these observations the following can be concluded:

s c um S
E- der ocu

• In order to minimize customer losses, the process (process center)

gh
Or is d

should be aligned with the customer requirement (specification

target).
Th

• It is beneficial to the customer if variation around the target value is

continually reduced (see Figure IV.9(e)).

This analysis is sometimes called aligning the “Voice of the Process”

with the “Voice of the Customer” (see Scherkenbach, W. W. (1991) for
more details).

In the example in Figure IV.9(d), the parts beyond specification account

for only 45% of the total loss to the customer. The remaining loss is
coming from parts within specification but not at the target.

153
 Th
CHAPTER IV - Section C

Or is d
E- der ocu
Suggested Use of Process Measures

Th Doc Num men

an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

154
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou

Figure IV.9: Process Alignment to Requirements

din p
gly
.
 CHAPTER IV - Section C
Suggested Use of Process Measures

This was determined by estimating the total loss: combining the loss
generated by the actual distribution of the parts (nonconforming parts)
and the loss due to the customer’s sensitivity to variation within the
specifications. This strongly suggests that the “Goal Post” model, or
computing percentage of “bad” parts (parts beyond specifications), in
and of itself does not provide a proper appreciation for understanding the
effect the process is actually having on the customer.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

155
 Th
CHAPTER IV - Section C

Or is d
E- der ocu
Suggested Use of Process Measures

Th Doc Num men

an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

156
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 APPENDIX A
Some Comments on Subgrouping

APPENDIX A
Some Comments on Sampling
Effects of Subgrouping
Control charts are used to answer questions about a process. In order to
have a control chart be useful, it is important that the charts answer the
right questions. An X chart asks the question, “Is the variation present
between subgroup averages more than is expected based on the variation
within subgroups?”. Therefore, understanding sources of variation within
and between subgroups is of paramount importance in understanding the

din p
control chart and the process variation. Most variables control charts

.
or ou
gly
compare within-subgroup variation to between-subgroup variation, so it

cc Gr
is important in interpreting the control charts to form subgroups with an

Inc

d a ion
understanding of the possible sources of variation affecting the process

rke Act
results.

al
on

ma ry
ati

ter ust
Autocorrelated Data
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

There are generally 3 features to any sampling that is performed when

3/3

doing SPC:
gn

an e A n:
Ma

as m

1. Size: How many parts are selected in the sample?

o
o
i
d h ut
t
to:

ted y t ira

2. Frequency: How often do we take a sample?

xp
op en ite 426 sed

3. Type: Will the sample consist of consecutively selected pieces,

h
ro ed E

randomly selected49 pieces, or some other structured plan?

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Of the 3 features mentioned above, most people are experienced with 1

en
an is d ume be t is

and 2, but 3 is seldom considered. In fact, sample type is not even

5

tec

covered in most control plan templates. The type of sample can have a
Th Doc Num men

large impact on the results of SPC charting and should be understood.

s c um S
E- der ocu

Some factors that influence the impact of the sample type have to do
gh

with the process itself – they are dependent on the nature of the
Or is d

manufacturing process. One particular phenomenon common with many

Th

modern day, high speed, automated processes is known as

autocorrelation.
The concept of correlation may be familiar to many people. There are
many examples of correlation which are part of experience in everyday
life (e.g., height/weight) where two features are compared in order to
determine if there appears to be a significant relationship between them.
As the value of one feature rises, the value of the other feature may rise
with it (indicating a positive correlation), or it may fall with it (indicating
49
It is important to understand the real meaning of “random”. In practice, many people think that by
blindly selecting pieces that what they are doing is “random” selection. In reality, this may be
haphazard sampling or convenience sampling. Selection of a random sample requires specific
techniques to ensure that the sample is random. Using haphazard or convenience sampling when
random sampling is required can lead to erroneous and biased conclusions.

157
APPENDIX A
Some Comments on Subgrouping

a negative correlation), or it may act independently of it (zero

correlation). The mathematical formula for correlation leads to a value
between -1 (negative correlation), through zero (no correlation), to +1
(positive correlation). In order to achieve these results, several samples
are taken from the population and the two features of interest are
compared within each other. In the production world, different
characteristics of the same product/process may be compared.
In autocorrelation, instead of comparing two features within a part, one
feature is compared to that same feature on a part produced before it. It
may be compared to the part produced immediately prior to it (called a
lag of 1), or two parts prior (a lag of 2), etc.

High-speed, automated processes are often found to exhibit

din p
.
or ou
gly
autocorrelation on some characteristics. This is often because there is an

cc Gr
underlying predictable special cause variation which is large when

Inc
compared to the common cause variation. That is, the important process

d a ion
input variables have not had time to vary much in the period of time the

rke Act
al
sample was taken when compared to the between sample variation. This
may be illustrated with examples. on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Temperature example:
1
aI

3/3

If one were asked to do an X and R control chart on the temperature of

gn

a room (or outside patio), it does not make sense to have a sampling plan
an e A n:
Ma

as m

that calls for taking 5 consecutive temperature readings – each of the 5

o
o
i
d h ut
t
to:

values would be essentially the same as each other. However, an hour

ted y t ira

later when the next sample is taken, the temperature would likely be
xp
op en ite 426 sed

different than it was an hour before, yet those 5 readings would again be
h
ro ed E
n

the same as each other. And so on.

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

When such a chart is completed, there would likely be some apparent

an is d ume be t is
5

random variation in the X chart, but the Range chart would be primarily
tec
Th Doc Num men

a stream of zeroes. The average range would then be approximately

o
s c um S

zero. R is used to calculate the control limits on the average, in the

E- der ocu

formula X ± A2 × R , so the control limits would be extremely tight on

gh
Or is d

the grand average of the data, and most points would show as being out
Th

of control. This is an extreme example, but it serves to point out what

happens when autocorrelation is present and is ignored.

Stamping example:
Data from a coil-fed, progressive die process is typically autocorrelated.
If this data were randomized (i.e., coil-fed steel were to be cut into
blanks, randomized and then measured), the data would then not be
autocorrelated. Yet the final shipped outcome (the total process
distribution as indicated by a histogram) would be identical. The
underlying cause for the autocorrelation has been broken. Is this
practical or feasible to do forever? No, not in this case – but this
example does serve to illustrate the possible nature of autocorrelation in
a process.

158
 APPENDIX A
Some Comments on Subgrouping

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o

Autocorrelation can lead to misleading conclusions if Cpk is calculated

i
d h ut
t
to:

ted y t ira

while ignoring its effect on the process variation. Since Cpk is based on
xp
op en ite 426 sed

R (a within-subgroup estimate of the standard deviation), it is evident in

h
ro ed E

the above example that the Cpk will be extremely high, yet it is obvious
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

that there is more variation in the process which has not been captured by
en

Cpk.
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

Identifying Autocorrelation
gh
Or is d

To discover if a process is autocorrelated, firstly, consider the process

Th

inputs in terms of the 6M’s.50 If a process is highly dependent on the

operator, it is not likely that the process would be autocorrelated. On the
other hand, if the process is highly dependent on raw material and that
raw material is a continuous variable (such as a coil of steel used to feed
a metal stamping process), autocorrelation within each coil is highly
likely. Similarly, for a process which is highly dependent on specific
machine related characteristics (such as a stamping press and die
combination as affected by lubrication, die temperature, tool condition,
etc.). When a process is both material and machine dependent,
autocorrelation may be significant.

50
Man, Material, Method, Machine, Mother Nature (Environment), Measurement System.

159
APPENDIX A
Some Comments on Subgrouping

Secondly, there are statistical analyses51 that can be used to determine the
actual autocorrelation coefficient and pattern. The methodology of
paired sample correlation analysis can be used to compare the current
sample to the prior sample, then the next sample to the current sample,
etc. When samples from a process are stable and independent, the plotted
point will be positioned “randomly” (random from a normal distribution)
between the control limits. The plotted points from an autocorrelated
process will not vary far from their neighboring sampling points, forming
a lazy, wandering pattern.

Ways to Address Autocorrelation

Often, nothing can be done to change an autocorrelated process.
Different sampling methods may be called for.

din p
.
or ou
gly
I and MR:

cc Gr
Inc
If the within-subgroup variation is less than or equal to the

d a ion
discrimination of the measurement system which is appropriate for the

rke Act
al
process, an I and MR chart may be a suitable method to control the
on
process variation. However, very strong autocorrelation may still display

ma ry
ati

ter ust
itself in a non-random pattern.
en e 11
rn

be otiv /20
wa Ind
nte

Structured Samples:
1
aI

The selection of the sampling quantity and frequency should reflect the
3/3
gn

dominant sources of variation. For example, if the process is material

an e A n:
Ma

as m

dominant, then the sampling should occur whenever the material changes
o
o

(e.g., with the change of coils).

i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E

Autoregressive Charts:
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

In cases where the assumption that the sample data are independent is
en

violated, an autoregressive model would be appropriate. See Chapter III.

an is d ume be t is
5

tec
Th Doc Num men

Structured Charting:
o
s c um S

If the source (special cause) of the autocorrelation is predictable, it is

E- der ocu

possible to control the process by segregating the within-subgroup

gh
Or is d

variation from the between-subgroup variation on separate charts. The

Between/Within chart utilizes an I and MR chart approach as well as the
Th

typical Range chart:

• The Individuals chart plots the subgroup averages treated as
individuals against the control limits based on the Moving Ranges.
• The MR chart plots the between-subgroup variation using the moving
ranges based on the subgroup averages.
• The Range (or Standard Deviation) chart plots the within-subgroup
variation.

51
The Durbin-Watson test statistic is one method to determine the degree of autocorrelation and is
included in many statistical software packages. See Biometrika, 38, pp. 159-178, 1951.
52
For example, see Appendix A for discussion on autocorrelation.

160
 APPENDIX A
Some Comments on Subgrouping

These would be analyzed using the standard control charting methods to

assure that both the common cause (within-subgroup) variation and the
cause of the autocorrelation (between-subgroup) remain consistent (see
Wheeler (1995)).

Summary
What is important here is considering the concept of autocorrelation and
the ability to recognize it in a process, then understanding its possible
impact on statistical results.
This discussion of autocorrelation is intended only to raise awareness
that such a phenomenon exists, how to recognize it, and that its effects, if

din p
.
or ou
gly
not recognized or understood, can be quite harmful to otherwise good

cc Gr
SPC practices. If the reader should suspect autocorrelation in a process,
then a statistician should be consulted.

Inc

d a ion
It is important to understand the real meaning of “random”. In practice,

rke Act
al
on
many people think that by blindly selecting pieces here and there that

ma ry
what they are doing is “random” selection. In reality, this may be
ati

ter ust
en e 11
haphazard sampling or convenience sampling (see Glossary). Selection
rn

be otiv /20
wa Ind
nte

of a random sample requires specific techniques (see a statistical

reference book). Using haphazard or convenience sampling when
1
aI

3/3

random sampling is required can lead to biased and therefore erroneous

gn

conclusions.
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

161
APPENDIX A
Some Comments on Subgrouping

Multiple Stream Process

Example
Consider the following example: A production process consists of four
parallel operations. It is suggested that variation in process output should
be studied with control charts, so a decision needs to be made on how to
collect the data for the charts. There is a variety of possible sampling
schemes that could be considered. Parts could be taken from each stream
to form a subgroup, or parts from only one stream could be included in
the same subgroup, or subgroups could be formed by taking parts from
the combined stream of output without regard to their source. The

din p
numerical example below provides an example of possible results

.
or ou
gly
obtained using these three methods.

cc Gr
Inc

d a ion
rke Act
al
Methods to collect data from the output of a multiple stream (spindle) production
on

ma ry
process
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Every hour a 16 part sample is collected by taking the parts from four
consecutive cycles from each stream.

162
 APPENDIX A
Some Comments on Subgrouping

The following is an example of the data.

CYCLE OF THE MACHINE

SAMPLE # A B C D
Stream S1 17 18 18 20
Stream S2 12 15 12 12
Stream S3 9 10 9 12
Stream S4 10 11 12 12

There are three sources of variation captured in the data. Cycle-to-Cycle

variation is captured by different columns in the array, stream-to-stream
variation is captured by the rows of the array, and hour-to-hour variation
is captured by different samples of 16 parts.

din p
.
or ou
gly
cc Gr
One subgrouping scheme would be to plot the average and range of each

Inc

d a ion
column of each array of data. Using this subgrouping scheme, stream-to-

rke Act
al
stream variation would be contained within each subgroup. Hour-to-hour
on
variation and cycle-to-cycle variation would contribute to differences

ma ry
ati
between subgroups. Another possible subgrouping scheme would be to

ter ust
en e 11
rn

plot the average and range of each row of each array of data. With this
be otiv /20
wa Ind
nte

subgrouping scheme, cycle-to-cycle variation would be contained within

1
aI

each subgroup and hour-to-hour and stream-to-stream variation would

3/3
gn

contribute to differences between subgroups.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Data from 20 consecutive hours are used to construct control charts with
each subgrouping method.

163
APPENDIX A
Some Comments on Subgrouping

Method 1: Subgrouping by column (Cycle)

This subgrouping scheme yields 80 subgroups of size n = 4. The grand
average is 11.76 units. The average range is 7.85. The control limits for
the X chart are 17.48 and 6.04 units, and the upper control limit for the
range chart is 17.91 units.
A review of the Range chart indicates that the within-subgroup variation
appears to be stable using this method.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

164
 APPENDIX A
Some Comments on Subgrouping

Method 2: Subgrouping by Row

The second subgrouping scheme yields 80 subgroups of size n = 4. The

average range is 2.84 units. The control limits for the X chart are 13.83
and 9.70 units, and the upper control limit for the range chart is 6.46
units. The control charts for this subgrouping scheme are shown below.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d

The control charts for the different subgrouping schemes are very
different even though they are derived from the same data. The X chart
Th

for data subgrouped by row shows a pattern: All of the points

corresponding to spindle 3 are noticeably higher than those from the
other streams. The first X chart does not reveal the stream-to-stream
differences because readings from each stream are averaged to obtain
each X value.
By grouping the data differently, the charts address different questions.
For the first set of charts, stream-to-stream variation is used as a basis of
comparison. The R chart checks to see that stream-to-stream variation is
stable over time and the X chart compares cycle-to-cycle and hour-to-
hour with stream-to-stream variation.

165
APPENDIX A
Some Comments on Subgrouping

The second set of charts use cycle to cycle variation as a basis for
comparison. The R chart checks to see that the cycle-to-cycle variation is
stable over time and the X chart compares stream-to-stream variation
and hour-to-hour variation with the base level of variation established by
the ranges; i.e., cycle-to-cycle variation. The second set of charts identify
that a special cause is affecting the process; i.e., the third stream is
different from the other streams. Since the stream to stream differences
are so large, the control limits in the first set of charts are much wider
than the second set.

With the second subgrouping method, the data could be used to create
four separate sets of control charts from the data, one for each stream.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh

This comparison of the charts shows that the average of the third stream is
Or is d

higher than the others and the individual processes are out of control. The
Th

base level of variation used for study of the results from each stream is cycle-
to-cycle variation as reflected in the range. For each stream the effects of
hour-to-hour variation are shown on the X charts. By plotting the charts
using the same scale, the level and variation for each stream can be
compared.

Method 3:
The third method of sampling would be to sample the parts from the
combined output from all four streams. This method gives some insight
into the variation that is sent to the next process but the parts can no
longer be differentiated by production stream. Provided the parts in the

166
 APPENDIX A
Some Comments on Subgrouping

combined stream are mixed, the ranges reflect a mixture of stream-to-

stream and cycle-to-cycle variation. The X values contain, in addition,
hour-to-hour variation. If the hour to hour contribution to variation is
large enough, that contribution will be seen as out-of-control points on
the X chart.

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E

The R chart checks to see if stream-to-stream and cycle-to-cycle

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

variation is consistent over time. The X chart answers the question, “Is
b
en

the variation in X values what would be expected if cycle-to-cycle and

an is d ume be t is
5

stream-to-stream variation were the only kinds of variation present in the

tec
Th Doc Num men

process, or, is there additional change hour to hour?”

s c um S
E- der ocu

As a general rule, the variation that is represented within subgroups

gh

should be the kind of variation that is believed to be the least significant

Or is d

or least interesting as a subject for current study. In all cases, a method of

Th

subgrouping should be used that will allow questions about the effects of
potential sources of variation to be answered.

167
APPENDIX A
Some Comments on Subgrouping

Effects of Sample Size on

Indices
A common scenario when evaluating a process using indices is that the
results of one sample may seemingly contradict the results of a second
sample. This is especially prevalent with new processes where the initial
sample taken to qualify the process for production has index values that
meet or exceed the customer requirements but a subsequent sample taken
during normal production has indices that fall short of the requirements.
The reasons for this are varied:
• The process has changed from the initial sampling to the full
production sampling – e.g., the initial sampling may have been using

din p
.
or ou
gly
different material, setups, procedures, etc.

cc Gr
• The initial sampling did not include all the possible sources of

Inc

d a ion
variation which are affecting the production process. This is a real

rke Act
possibility if the initial sample size is small.

al
• on
The actual process index is close to the target index and sampling

ma ry
ati

ter ust
variation is causing the difference in conclusions.
en e 11
rn

be otiv /20
wa Ind
nte

The first two reasons relate to the understanding of the sources of

variation acting on the process and are discussed in Chapter I.
1
aI

3/3
gn

The third reason deals with the sampling variation inherent in any
an e A n:
Ma

as m

sampling scheme (see also Chapter I, Section G). Unless the sample
o
o

includes all the output of the process, there will be sampling variation53
i
d h ut
t
to:

ted y t ira

when calculating a statistic (in this case an index) of the process

xp
op en ite 426 sed

distribution.
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Actual Ppk
en

Sampling
an is d ume be t is
5

tec

Variation
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

Confidence Bounds
Using the sampling distribution (the distribution of the statistic (index)),
it is possible to calculate confidence bounds for the index. These values

53
Note: Although the actual sampling distribution of the indices will generally be non-normal, this
discussion will use a symmetric distribution as an example.

168
 APPENDIX A
Some Comments on Subgrouping

can then be used to make a decision about the process (e.g., is it

acceptable or not).
Using a common alpha risk level of .05 as an example, the 95%
confidence bounds will identify the range of possible values that will
contain the actual (and unknown) value 95% of the time. That is, if the
sampling was identically repeated 100 times, the same decision
(acceptable or unacceptable) on the process would be made 95 times.

Upper Confidence Bound

din p
.
or ou
gly
Observed Ppk

cc Gr
Inc

d a ion
rke Act
al
Actual Ppk on

ma ry
ati
Lower Confidence Bound

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

Sample Size
an e A n:
Ma

as m
o
o
i
d h ut

The width of the sampling distribution is a function of the sample size.

t
to:

ted y t ira

The larger the sample size the “tighter” the sampling distribution. It is
xp
op en ite 426 sed

this attribute of the sampling distribution that can lead to seemingly

h
ro ed E

contradictory conclusions.
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

For example, when evaluating new processes the initial sample is usually
an is d ume be t is

small due to the availability of raw materials/parts. Once the process is in

5

tec

production, this constraint is not present.

Th Doc Num men

o
s c um S
E- der ocu

gh

Initial Extended
Or is d

Study Study
Th

Target Ppk
Observed Ppk

Actual Ppk

Sample Size

169
APPENDIX A
Some Comments on Subgrouping

When the actual index is close to the target index then the differences in
sampling variation can lead to seemingly contradictory conclusions even
if there are no changes in the process and both samples encompass the
same sources of variation.

Initial Extended
Study Study

Observed Ppk
Actual Ppk

din p
.
or ou
gly
Target Ppk

cc Gr
Inc

d a tion
al

ar Ac
on

rm try
ati

ke
n w In 1
ate dus
rn

ee tive 201
Sample Size
nte

/
ha tom /31
aI
gn

nd Au n: 3
sb o

In the case where the actual index (unknown) is exactly equal to the
Ma

d a the tio

target index then, regardless of the sample size, the probability of calling
to:

cte by xpira

the process acceptable is only 50%. In other words, the calculated index
will be greater or equal to the target index only half the time.
py ent te L 262 ed
pr ne e E
i 54 ns
d i oc nt r: 4 lice

Initial Extended
ht ow ns
ote d
rig is ice

Study Study
an is d ume be t is
Th Doc Num men

s c um S
E- der ocu

Actual Ppk
Or is d

oPpk

and
Th

Target Ppk
Observed

Sample Size
The sample size used in a process study and how close the actual index
is to the target index has a significant impact on the validity of any
predictive decision made about the process.

170
 APPENDIX B
Some Comments on Special Causes

APPENDIX B
Some Comments on Special Causes
Over-Adjustment

Over-adjustment is the practice of treating each deviation from the

target as if it were the result of the action of a special cause of variation
in the process.
If a stable process is adjusted on the basis of each measurement made,
then the adjustment becomes an additional source of variation. The

din p
.
or ou
gly
following examples demonstrate this concept. The first graph shows the

cc Gr
variation in results with no adjustment. The second graph shows the

Inc
variation in results when an adjustment is made to the process to

d a ion
compensate for each deviation from the target. The third graph shows

rke Act
al
variation in results when adjustments are made to compensate only when
on
the last result was more than one unit from the target. This third case is

ma ry
ati

ter ust
an example of compensation to stay within a set of specifications. Each
en e 11
rn

method of adjustment increases the variation in the output, since the

be otiv /20
wa Ind
nte

variation without adjustment is stable (see Deming (1989), Chapter 11).

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

171
 APPENDIX B

Th
Or is d
E- der ocu
Some Comments on Special Causes

Th Doc Num men

an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

172
an e A n: ati
d h ut on
as m o 3/3
1 al

assessed and is appropriate.

be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.

NOTE: These charts assume the measurement system has been

 APPENDIX B
Some Comments on Special Causes

Time Dependent Processes

Case 3 processes (see Chapter I, Section E) are usually difficult to fit into
the classical control chart model. Few processes of these types remain
strictly stable over time. Because the within-subgroup variation is usually
small, minor fluctuations in process location or dispersion may cause a
process to be out of statistical control, when, in fact, the condition has a
minimal practical effect on product quality and the customer.
For example, consider a process that has a fairly constant dispersion, but
has small, random location changes. When the control limits were
established based on the first 25 samples, numerous points out of control
were revealed.

din p
.
or ou
gly
cc Gr
Inc
Xbar Chart of Process

d a ion
30

rke Act
al
on

ma ry
ati
25

ter ust
en e 11
rn

be otiv /20
wa Ind
nte
Sample Mean

UCL=20.92
_
_
1
aI

20 X=19.65
3/3
gn

LCL=18.37
an e A n:
Ma

as m
o
o

15
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E

10
n

1 5 11 15 21 25 31 35 41 45 50
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Sample
en
an s d me be is
5

tec
t
Th Doc Num men

Yet, because the process capability is small when compared to the

s c um S

specifications and the loss function is flat, a histogram of the data

E- der ocu

suggests that there is a minimal risk of impacting the customer.

gh
Or is d

u
Th

LSL USL
i

Target

173
APPENDIX B
Some Comments on Special Causes

LSL = 13 Histogram of Process USL = 27

35

30

25

Frequency
20

15

10

din p
.
or ou
gly
0
12 14 16 18 20 22 24 26 28

cc Gr
Inc

d a ion
rke Act
al
The process is being run on one shift per day. When the data are
on
evaluated on that basis, the process exhibited short-term periods of

ma ry
ati
statistical control.

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Xbar Chart of Combined Process by Stage

1
aI

Time Period 0 1 2 3 4
3/3

30
gn

an e A n:
Ma

as m
o
o
i

25
d h ut
t
to:

ted y t ira
Th Doc Num men Sample Mean

xp
op en ite 426 sed

h
ro ed E

20
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is

15
5

tec
o
s c um S
E- der ocu

10
1 5 11 15 21 25 31 35 41 45 50
gh

Sample
Or is d
Th

This implies that the process could be monitored using a short run chart
(see Chapter III). Other time dependent processes can be monitored by
the Individuals and Moving Range Chart, the EWMA Chart, the ARIMA
Chart and others.
The charts reveal sensitivity to some special cause. The need for further
investigation or process improvement should be considered in the
context of business priorities.
The question is whether process parameters are reliable when estimated
under such conditions. The answer is no. Deming (1986), Wheeler and
Chambers (1992) and Bothe (2002) discuss the risks involved in making
capability evaluations when the process lacks statistical control. The
consequences of making an erroneous decision based on data from an

174
 APPENDIX B
Some Comments on Special Causes

unstable process can be severe. In general, a stable process is a

prerequisite for correctly estimating process capability.
However, in certain situations, such as the time dependent process in the
example, the classical indices will provide a conservation estimate of the
process performance.

Under certain circumstances, the customer may allow a producer to run a

process even though it is a Case 3 process. These circumstances may
include:
• The customer is insensitive to variation within specifications (see
discussion on the loss function in Chapter IV).

din p
• The economics involved in acting upon the special cause exceed the

.
or ou
gly
benefit to any and all customers. Economically allowable special

cc Gr
causes may include tool wear, tool regrind, cyclical (seasonal)

Inc

d a ion
variation, etc.

rke Act
al
• The special cause has been identified and has been documented as
consistent and predictable.
on

ma ry
ati

ter ust
en e 11
In these situations, the customer may require the following:
rn

be otiv /20
wa Ind
nte

• The process is mature.

1
aI

3/3

• The special cause to be allowed has been shown to act in a consistent

gn

manner over a known period of time.

an e A n:
Ma

as m
o
o

•
i

A process control plan is in effect which will assure conformance to

d h ut
t
to:

ted y t ira

specification of all process output and protection from other special

xp
op en ite 426 sed

causes or inconsistency in the allowed special cause.

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Repeating Patterns
an is d ume be t is
5

tec
Th Doc Num men

There are times when repetitive patterns are present in control charts due
s c um S

to known assignable causes -- causes that cannot economically be

E- der ocu

eliminated.
gh
Or is d

Consider an operation where an outer diameter of a shaft is being

machined. As the machining tool wears, the outer diameter will becomes
Th

larger. In this example the average chart would have an increasing trend.
This trend would continue until the tool is replaced. Over time the
average chart will exhibit a sawtooth pattern. As this example
highlights, repetitive trends will be present when a process has
significant input variables that change consistently overtime which
cannot economically be reduced to random causes.

Another example of a process that can produce trends is a process

involving chemicals. As parts are processed, the concentration of the
chemicals becomes weaker thereby producing a trend. The trend
continues until the chemical concentration is brought back to the initial
level by process adjustment.

175
APPENDIX B
Some Comments on Special Causes

Other examples include processes influenced by ambient temperature,

humidity and human fatigue. When these types of repetitive patterns
exist, the average chart will exhibit conditions associated with an out-of-
control process since there is the (economically influenced) special cause
acting on the process. If the influence of this special cause can be shown
to be predictable over time and the additional variation is acceptable to
the customer, then the process controls can be modified to allow it.
One approach to this is replacing standard control limits with modified
control limits. See AT&T (1984), Grant and Leavenworth (1996),
Duncan (1986), Charbonneau Webster (1978) for more information on
modified control limits.

Where modified control limits are used caution should be employed since
these charts may fail to disclose the presence or absence of statistical

din p
.
or ou
gly
control in the manufacturing process.

cc Gr
An alternate approach is to use the Regression Control Chart discussed in

Inc

d a ion
Chapter III.

rke Act
al
Besides influencing trends, these types of special causes may also cause
on
a batch to batch mean shift. If this additional variation is acceptable to

ma ry
ati
the customer then the process may be controlled using Short Run Charts

ter ust
en e 11
rn

discussed in Chapter III.

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

176
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

177
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
APPENDIX C

wa Ind
ter ust
ma ry
rke Act
Described in This Manual

d a ion
cc Gr
or ou
din p
gly
.
Selection Procedure for the Use of the Control Charts
APPENDIX C
Selection Procedure for the Use of the Control Charts Described in This Manual
 APPENDIX C

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

178
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
Selection Procedure for the Use of the Control Charts Described in This Manual

wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
APPENDIX D
Relationship Between Cpm and Other Indices

APPENDIX D
Relationship Between Cpm and Other Indices

The Cpm index, often associated with Taguchi's Loss Function, was
developed as an alternate way to account for the effect of process
centering on estimates of process capability or performance. The Cpk and
Ppk indices focus on the process mean and not the specification target
value, while the Cpm index focuses on the target value. As discussed in
Chapter IV, all four of the standard indices (Cp, Cpk, Pp, and Ppk) should
be evaluated for the same data set to obtain a comprehensive assessment

din p
of process capability and performance. A large difference between Cp

.
or ou
gly
and Cpk or between Pp and Ppk is an indication of a centering problem. In

cc Gr
contrast, by including the variation between the process mean and the

Inc

d a ion
specification target value in the calculation, the Cpm index evaluates how
well the process meets the specification target whether it is centered or

rke Act
al
not. See Boyles (1991) and Chan, L. J., S.W. Cheng, and F.A. Spiring
on

ma ry
(1988) for additional information.
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

The difference between Cpm and other indices discussed in this manual
1
aI

results from the way the standard deviation is calculated. The indices
3/3

discussed in the text use the standard deviation; i.e., the variation around
gn

an e A n:
Ma

process mean, X . Cpm uses an analogue based on the target, i.e., the
as m
o
o

variation around the target, T .

i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

USL − LSL n
(x i − T ) 2
t p wn se

∑
yri t is Lic 2
d i oc nt r: 4 lice

C pm = sC pm =
b

, where
en

6 sC pm i=1 n −1
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S

The following graphs assume bilateral tolerance:54 That is,

E- der ocu

(USL – T) = (T – LSL)
gh
Or is d
Th

54
See Bothe (2001) for a discussion of the situation where T is not the middle of the specification.

179
 APPENDIX D

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
Relationship Between Cpm and Other Indices

s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

180
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 APPENDIX E
Table of Constants and Formulas for Control Charts

APPENDIX E
Table of Constants and Formulas for Control Charts*
X and R Charts X and s Charts
Chart for Chart for Ranges (R) Chart for Chart for Ranges (R)
Averages Averages
Control Divisors to Factors for Control Divisors to Factors for
Limits Estimate Control Limits Limits Estimate Control Limits
Factor σX Factor σX
Subgroup A2 d2 D3 D4 A3 c4 B3 B4
Size

din p
.
or ou
gly
2 1.880 1.128  3.267 2.659 0.7979  3.267

cc Gr
3 1.023 1.693  2.574 1.954 0.8862  2.568

Inc

d a ion
4 0.729 2.059  2.282 1.628 0.9213  2.266

rke Act
5 0.577 2.326 2.114 1.427 0.9400 2.089

al
 

on
6 0.483 2.534  2.004 1.287 0.9515 0.030 1.970

ma ry
ati
7 0.419 2.704 0.076 1.924 1.182 0.9594 0.118 1.882

ter ust
en e 11
rn

8 0.373 2.847 0.136 1.864 1.099

be otiv /20 0.9650 0.185 1.815

wa Ind
nte

9 0.337 2.970 0.184 1.816 1.032 0.9693 0.239 1.761

1
aI

10 0.308 3.078 0.223 1.777 0.975 0.9727 0.284 1.716

3/3
gn

11 0.285 3.173 0.256 1.744 0.927 0.9754 0.321 1.679

an e A n:
Ma

as m

12 0.266 3.258 0.283 1.717 0.886 0.9776 0.354 1.646

o
o

13 0.249 3.336 0.307 1.693 0.850 0.9794 0.382 1.618

i
d h ut
t
to:

ted y t ira

14 0.235 3.407 0.328 1.672 0.817 0.9810 0.406 1.594

xp
op en ite 426 sed

15 0.223 3.472 0.347 1.653 0.789 0.9823 0,428 1.572

h
ro ed E

16 0.212 3.532 0.363 1.637 0.763 0.9835 0.448 1.552

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

17 0.203 3.588 0.378 1.622 0.739 0.9845 0.466 1.534

b
en

18 0.194 3.640 0.391 1.608 0.718 0.9854 0.482 1.518

an is d ume be t is
5

19 0.187 3.689 0.403 1.597 0.698 0.9862 0.497 1.503

tec
Th Doc Num men

20 0.180 3.735 0.415 1.585 0.680 0.9869 0.510 1.490

o

21 0.173 3.778 0.425 1.575 0.663 0.9876 0.523 1.477

s c um S
E- der ocu

22 0.167 3.819 0.434 1.566 0.647 0.9882 0.534 1.466

gh

23 0.162 3.858 0.443 1.557 0.633 0.9887 0.545 1.455

Or is d

24 0.157 3.895 0.451 1.548 0.619 0.9892 0.555 1.445

Th

25 0.153 3.931 0.459 1.541 0.606 0.9896 0.565 1.435

Centerline Control Limits

X and R Charts CLX = X UCLX = X + A2 R LCLX = X − A2 R

CLR = R UCLR = D4 R LCLR = D3 R

X and s Charts CLX = X UCLX = X + A3 S LCLX = X − A3 S
CLs = s UCLs = B4 s LCLs = B3 s

*
From ASTM publication STP-15D, Manual on the Presentation of Data and Control Chart Analysis, 1976;
pp 134-136. Copyright ASTM, 1916 Race Street, Philadelphia, Pennsylvania 19103. Reprinted, with
permission.

181
APPENDIX E
Table of Constants and Formulas for Control Charts

APPENDIX E - Table of Constants and Formulas for Control Charts (Cont.)

Median Charts ** Charts for Individuals

Chart for Chart for Ranges (R) Chart for Chart for Ranges (R)
Medians Individuals
Control Divisors to Factors for Control Divisors to Factors for
Limits Estimate Control Limits Limits Estimate Control Limits
Factor σX Factor σX
Subgroup
Size A2 d2 D3 D4 E2 d2 D3 D4
2 1.880 1.128  3.267 2.660 1.128  3.267

din p
.
3 1.187 1.693  2.574 1.772 1.693  2.574

or ou
gly
cc Gr
4 0.796 2.059  2.282 1.457 2.059  2.282

Inc

d a ion
5 0.691 2.326  2.114 1.290 2.326  2.114

rke Act
al
6 0.548 2.534 2.004 1.184 2.534 2.004
on
 

ma ry
ati
7 0.508 2.704 0.076 1.924 1.109 2.704 0.076 1.924

ter ust
en e 11
rn

be otiv /20
wa Ind
8 0.433 2.847 0.136 1.864 1.054 2.847 0.136 1.864
nte

1
aI

9 0.412 2.970 0.184 1.816 1.010 2.970 0.184 1.816

3/3
gn

10 0.362 3.078 0.223 1.777 0.975 3.078 0.223 1.777

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Centerline Control Limits

xp
op en ite 426 sed

CLX = X UCLX = X + A2 R LCLX = X − A2 R

h

Median Charts
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

CLR = R UCLR = D4 R LCLR = D3 R

en
an is d ume be t is
5

tec
Th Doc Num men

Charts for Individuals CLX = X UCLX = X + E2 R LCLX = X − E2 R

s c um S
E- der ocu

CLR = R UCLR = D4 R LCLR = D3 R

gh
Or is d
Th

For extended d2 tables see the MSA Manual 3rd edition.

**
A2 factors derived from ASTM-STP-15D Data and Efficiency Tables contained in Dixon and Massey
(1969), page 488.

182
 APPENDIX E
Table of Constants and Formulas for Control Charts

APPENDIX E - Table of Constants and Formulas for Control Charts (Cont.)

Attributes Charts

Centerline Control Limits

Samples not necessarily of constant size
p (1 − p ) p (1 − p )
p chart for UCLPi = p + 3 LCLPi = p − 3
CLp = p ni ni
proportions of
units in a category If the sample size is constant (n)
p (1 − p ) p (1 − p )
UCLP = p + 3 LCLP = p − 3
n n

din p
.
or ou
gly
np np

cc Gr
np chart for UCLnp = np + 3 np (1 − ) LCLnp = np − 3 np (1 − )
number/rate of CLnp = np n n

Inc

d a ion
units in a category

rke Act
= np + 3 np(1 − p ) = np − 3 np(1 − p )

al
c chart for number on

ma ry
ati

ter ust
of incidences in
en e 11
rn

one or more CLC = c UCLC = c + 3 c

be otiv /20 LCLC = c − 3 c
wa Ind
nte

categories
1
aI

Samples not necessarily of constant size

3/3
gn

3 u 3 u
an e A n:
Ma

as m

UCLu = u + LCLu = u −
o
o

ni ni
i
d h ut
t
to:

ted y t ira
xp

u u
op en ite 426 sed

=u +3 = u −3
h
ro ed E

ni ni
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Using average sample size

en
an is d ume be t is

3 u 3 u
5

UCLu = u + LCLu = u −
tec
Th Doc Num men

u chart for
o

n n
number of CLu = u
s c um S
E- der ocu

incidences per unit u u

= u +3 = u −3
gh

in one or more
Or is d

n n
categories
If the sample size is constant (n)
Th

3 u 3 u
UCLu = u + LCLu = u −
n n
u u
= u +3 = u −3
n n

183
 APPENDIX E

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
Table of Constants and Formulas for Control Charts

yri t is Lic 2 to:

gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

184
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 APPENDIX F
Capability Index Calculations Example

APPENDIX F
Capability Index Calculations Example
For capability indices to be valid, several assumptions should be satisfied
(see Chapter IV, Section A and Section B). They are:
• The process from which the data come is statistically stable, that is,
the normally accepted SPC rules must not be violated.
• The individual measurements from the process data form an
approximately normal distribution.55
• A sufficient number of parts must be evaluated in order to capture

din p
.
or ou
gly
the variation that is inherent in the process. It is recommended that at

cc Gr
least 125 individual values be collected using a subgroup size of five.
Other subgroup sizes may be more appropriate for a particular

Inc

d a ion
application, but the total sample size should be at least 125.

rke Act
al
on
• The specifications are based on customer requirements.

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

The following data set is evaluated against these assumptions and, since
the assumptions hold, the capability indices are calculated.
1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

55
For non-normal distributions, see Chapter IV, Section B.

185
APPENDIX F
Capability Index Calculations Example

Data Set:

Data Diameter Data Diameter Data Diameter Data Diameter Data Diameter
Point Point Point Point Point
1 22.30 26 22.37 51 22.28 76 22.65 101 22.48
2 22.54 27 22.34 52 22.55 77 22.50 102 22.38
3 22.01 28 22.75 53 22.38 78 22.41 103 22.28
4 22.62 29 22.71 54 22.65 79 22.39 104 22.72
5 22.65 30 22.51 55 22.56 80 22.48 105 22.96

din p
.
or ou
gly
6 22.86 31 22.23 56 22.54 81 22.50 106 22.53

cc Gr
7 22.68 32 22.36 57 22.25 82 22.86 107 22.52

Inc

d a ion
8 22.43 33 22.90 58 22.40 83 22.60 108 22.61

rke Act
al
on
9 22.58 34 22.45 59 22.72 84 22.60 109 22.62

ma ry
ati

ter ust
en e 11
10 22.73 35 22.48 60 22.90 85 22.66 110 22.60
rn

be otiv /20
wa Ind
nte

11 22.88 36 22.60 61 22.31 86 22.79 111 22.54

1
aI

3/3

12 22.68 37 22.72 62 22.57 87 22.61 112 22.56

gn

an e A n:
Ma

as m

13 22.46 38 22.35 63 22.38 88 22.81 113 22.36

o
o
i
d h ut
t
to:

ted y t ira

14 22.30 39 22.51 64 22.58 89 22.66 114 22.46

xp
op en ite 426 sed

15 22.61 40 22.69 65 22.30 90 22.37 115 22.71

h
ro ed E
n

t p wn se
yri t is Lic 2

16 22.44 41 22.61 66 22.42 91 22.65 116 22.84

d i oc nt r: 4 lice

b
en

17 22.66 42 22.52 67 22.21 92 22.75 117 22.52

an is d ume be t is
5

tec
Th Doc Num men

18 22.48 43 22.52 68 22.45 93 21.92 118 22.88

o
s c um S

19 22.37 44 22.49 69 22.24 94 22.00 119 22.68

E- der ocu

gh

20 22.56 45 22.31 70 22.55 95 22.45 120 22.54

Or is d
Th

21 22.59 46 22.42 71 22.25 96 22.51 121 22.76

22 22.65 47 22.64 72 22.36 97 22.58 122 22.65
23 22.78 48 22.52 73 22.25 98 22.46 123 22.51
24 22.58 49 22.40 74 22.34 99 22.76 124 22.77
25 22.33 50 22.63 75 22.67 100 22.56 125 22.43

186
 APPENDIX F
Capability Index Calculations Example

Analysis
Histogram Chart, Normality Plot and the Xbar and R Chart can be used to determine the validity of the
first two assumptions.

Histogram of Diameter Data

LSL USL

Within

Overall

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

21.5 22.0 22.5 23.0 23.5

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Normal Probability Plot

xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

.999
an is d ume be t is
5

tec

.99
Th Doc Num men

.95
s c um S
E- der ocu
Probability

.80
gh
Or is d

.50
Th

.20
P-Value: 0.677
.05
.01
.001

22.0 22.5 23.0

Diameter

The above two graphs provide evidence that the data likely came from a normally distributed population.

187
APPENDIX F
Capability Index Calculations Example

Xbar/R Chart for Diameter

22.8
Sample Mean UCL=22.77
22.7
22.6
Mean=22.53
22.5
22.4
22.3 LCL=22.29

Subgroup 0 5 10 15 20 25

din p
1.0

.
or ou
gly
UCL=0.8847
Sample Range

cc Gr
Inc

d a ion
0.5
R=0.4184

rke Act
al
on

ma ry
ati
0.0 LCL=0

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

Control charts provide evidence that the process is in statistical control. Consequently it is
an e A n:
Ma

as m

appropriate to calculate the indices for this data set.

o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

Diameter Statistics:
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Sample size = n = 125

an is d ume be t is
5

tec

Subgroup Size = 5
Th Doc Num men

Number of Subgroups = 25
s c um S
E- der ocu

Upper Specification Limit = 23.5

gh
Or is d

Lower Specification Limit = 21.5

Th

The specifications are based on customer and functional requirements.

0.4184
Within-subgroup standard deviation = σ̂ C = R = = 0.179880
d2 2.326
n
(x i − x) 2
Total variation standard deviation = s = ∑i n − 1 = 0.189037

The above information is necessary for the evaluation of the indices.

188
 APPENDIX F
Capability Index Calculations Example

USL - LSL 23.5 − 21.5

Cp = = = 1.85

6 R  6 × 0.179881

 d2 

 
 X − LSL USL − X 
C pk = Minimum  CPL = , CPU = 
 3  R  3  R  
  d 2  d2  

 22.5308 − 21.5 23.5 − 22.5308 

din p
= Minimum  CPL =

.
, CPU =

or ou


gly
 3 × 0.179880 3 × 0.179880 

cc Gr
Inc

d a ion
= Minimum ( CPL = 1.91, CPU = 1.80 )

rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

= 1.80 be otiv /20

wa Ind
nte

1
aI

3/3
gn

USL - LSL 23.5 − 21.5

Pp = = = 1.76
an e A n:
Ma

as m

6s 6 × 0.189037
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

 USL − X 
h

X − LSL
ro ed E

Ppk = Minimum  PPL = , PPU = 

n

t p wn se


yri t is Lic 2
d i oc nt r: 4 lice

6s 6s
b

 
en
an is d ume be t is
5

tec
Th Doc Num men

 22.5308 − 21.5 23.5 − 22.5308 

= Minimum  PPL = , PPU = 
s c um S
E- der ocu

 3 × 0.189037 3 × 0.189037 
gh
Or is d

= Minimum ( PPL = 1.82, PPU = 1.71)

Th

= 1.71

189
APPENDIX F
Capability Index Calculations Example

Conclusion:
The following observations are made:
• Cpk is approximately equal to Cp, and Ppk is approximately equal to
Pp. Both of these conditions are indicators that the process is well
centered.
• All indices are relatively high indicating that the process is capable
of producing near-zero nonconformances if the process remains in
statistical control.
• Since the Cp and Pp are approximately equal it implies minimal
between-subgroup variation.

din p
.
or ou
gly
• A large discrepancy between Cpk and Ppk would indicate the presence

cc Gr
of excessive between-subgroup variation.

Inc

d a ion
• A large discrepancy between Cp and Cpk (or between Pp and Ppk)

rke Act
al
would indicate a process centering problem.
on

ma ry
ati

ter ust
en e 11
rn
NOTE: The process variability is an integral part of the capability index
be otiv /20
wa Ind
nte

calculations, hence it is important to be consistent in choosing the

method to calculate the within-subgroup variability. As the table below
1
aI

3/3

shows there are two ways to estimate the process variability (σˆ C ) and its
gn

effect on the Cpk calculations. Both are correct; i.e., both are valid
an e A n:
Ma

as m
o
o

estimates of the “true” variation. Use R if an X and R chart is used

i
d h ut
t
to:

ted y t ira

d2
xp
op en ite 426 sed

to collect the data and s if an X and s chart is used.

h
ro ed E

c4
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

Method for calculating σˆ C Summary of Results

o
s c um S

Within-subgroup Variation Cpk

E- der ocu

gh

R 0.1799 1.80
Or is d

d2
Th

s 0.1820 1.78
c4

NOTE: The total variation standard deviation value (σˆ P = 0.1890) is

not affected by the methodology used to estimate within-subgroup
variation.

190
 Appendix G
Glossary of Terms and Symbols

APPENDIX G
Glossary of Terms and Symbols

Terms Used in This Manual

ARMA Control Chart The Autoregressive Moving Average Control Chart is a control
chart which uses a regression model to account for
interrelationship among the data. It may be used in cases where
the assumption that the sample data are independent is violated.

din p
.
or ou
gly
Attributes Data Qualitative data that can be categorized for recording and

cc Gr
analysis. Examples include characteristics such as: the

Inc
presence of a required label, the installation of all required

d a ion
fasteners, the absence of errors on an expense report. Other

rke Act
al
on
examples are characteristics that are inherently measurable

ma ry
ati
(i.e., could be treated as variables data), but where the

ter ust
en e 11
rn

results are recorded in a simple yes/no fashion, such as

be otiv /20
wa Ind
nte

acceptability of a shaft diameter when checked on a go/no-

1
aI

go gage, or the presence of any engineering changes on a

3/3
gn

drawing. Attributes data are usually gathered in the form of

an e A n:
Ma

nonconforming units or of nonconformities; they are

as m
o
o

analyzed by p, np, c and u control charts (see also Variables

i
d h ut
t
to:

ted y t ira

Data).
xp
op en ite 426 sed

h
ro ed E
n

t p wn se

Autocorrelation The degree of relationship between elements of a stationary

yri t is Lic 2
d i oc nt r: 4 lice

b
en

time series.
an is d ume be t is
5

tec
Th Doc Num men

Average (see also Mean) The sum of values divided by the number (sample size) of
s c um S
E- der ocu

values. It is designated by a bar over the symbol for the

gh

values being averaged. For example:

Or is d
Th

o X (X-bar) is the average of the X values within a

subgroup;

o X (X double bar) is the average of subgroup averages

( X );

o X (X tilde-bar) is the average of subgroup medians;

o R (R-bar) is the average of subgroup ranges.

Average Run Length The number of sample subgroups expected between out-of-
control signals. The in-control Average Run Length

191
Appendix G
Glossary of Terms and Symbols

( ARL0 ) is the expected number of subgroup samples

between false alarms.

Between-subgroup Variation See Variation.

Binomial Distribution A discrete probability distribution for attributes data that

applies to conforming and nonconforming units and
underlies the p and np charts.

Cause and Effect Diagram A simple tool for individual or group problem solving that

din p
.
or ou
gly
uses a graphic description of the various process elements

cc Gr
to analyze potential sources of process variation. Also

Inc

d a ion
called fishbone diagram (after its appearance) or Ishikawa
diagram (after its developer).

rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

Centerline The line on a control chart that represents the average value
be otiv /20
wa Ind
nte

of the items being plotted.

1
aI

3/3
gn

Characteristic A distinguishing feature of a process or its output.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

Common Cause A source of variation that affects all the individual values
h
ro ed E

of the process output being studied; this is the source of the

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

inherent process variation.

b
en
an is d ume be t is
5

tec
Th Doc Num men

Confidence Interval An interval or range of values, calculated from sample data,

o
s c um S

that contains, with a (100 - α) degree of certainty, the

E- der ocu

population parameter of interest, e.g., the true population

gh
Or is d

average. α, called the Level of Significance, is the

probability of committing a Type I error. See Montgomery
Th

(1997) or Juran and Godfrey (1999) for calculation

methods.

Consecutive Units of output produced in succession; a basis for

selecting subgroup samples.

Continual Improvement The operational philosophy that makes best use of the
talents within the Company to produce products of
increasing quality for our customers in an increasingly
efficient way that protects the return on investment to our
stockholders. This is a dynamic strategy designed to

192
 Appendix G
Glossary of Terms and Symbols

enhance the strength of the Company in the face of present

and future market conditions. It contrasts with any static
strategy that accepts (explicitly or implicitly) some
particular level of outgoing nonconformances as inevitable.

Control See Statistical Control.

Control Chart A graphic representation of a characteristic of a process,

showing plotted values of some statistic gathered from that
characteristic, a centerline, and one or two control limits. It
minimizes the net economic loss from Type I and Type II

din p
.
errors. It has two basic uses: as a judgment to determine if a

or ou
gly
process has been operating in statistical control, and to aid

cc Gr
in maintaining statistical control.

Inc

d a ion
rke Act
al
Control Limit on
A line (or lines) on a control chart used as a basis for

ma ry
ati

ter ust
judging the stability of a process. Variation beyond a
en e 11
rn

control limit is evidence that special causes are affecting

be otiv /20
wa Ind
nte

the process. Control limits are calculated from process data

1
aI

3/3

and are not to be confused with engineering specifications.

gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

Control Statistic The statistic used in developing and using a control chart.
ted y t ira

A value calculated from or based upon sample data (e.g., a

xp
op en ite 426 sed

subgroup average or range), used to make inferences about

ro ed E
n

t p wn se

the process that produced the output from which the sample
yri t is Lic 2
d i oc nt r: 4 lice

b
en

came.
an is d ume be t is
5

Convenience Sampling A sample scheme wherein the samples are collected using
tec
Th Doc Num men

an approach which makes it “easy” to collect the samples

s c um S

but does not reflect the nature of potential special causes

E- der ocu

which could affect the process. Examples of this are

gh
Or is d

collecting samples just before a break period, or from the

Th

top of a bin, pallet or other storage container. This type of

sampling is not appropriate for process analysis or control
because it can lead to a biased result and consequently a
possible erroneous decision.

Correlation The degree of relationship between variables.

Correlation Matrix A matrix of all possible correlations of factors under

consideration.

193
Appendix G
Glossary of Terms and Symbols

CUSUM Control Chart A control chart approach that uses the current and recent
past process data to detect small to moderate shifts in the
process average or variability. CUSUM stands for
“cumulative sum” of deviations from the target and puts
equal weight on the current and recent past data.

Datum The singular of “data”. A single point in a series of data.

Not to be confused with the word as used within Geometric
Dimensioning & Tolerancing (GD&T).

Detection A reactive (past-oriented) strategy that attempts to identify

din p
.
unacceptable output after it has been produced and then

or ou
gly
separate it from acceptable output (see also Prevention).

cc Gr
Inc

d a ion
rke Act
al
Dispersion See Process Spread.
on

ma ry
ati

ter ust
en e 11
rn

Distribution A way of describing the output of a stable system of

be otiv /20
wa Ind
nte

variation, in which individual values as a group form a

1
aI

3/3

pattern that can be described in terms of its location,

gn

spread, and shape. Location is commonly expressed by the

an e A n:
Ma

as m

mean or average, or by the median; spread is expressed in

o
o
i
d h ut
t
to:

terms of the standard deviation or the range of a sample;

ted y t ira

shape involves many characteristics such as symmetry

xp
op en ite 426 sed

(skewness) and peakedness (kurtosis). These are often

ro ed E
n

t p wn se

summarized by using the name of a common distribution

yri t is Lic 2
d i oc nt r: 4 lice

b
en

such as the normal, binomial, or poisson.

an is d ume be t is
5

tec
Th Doc Num men

EWMA Control Chart The Exponentially Weight Moving Average Control Chart
s c um S
E- der ocu

is an approach to detect small shifts in the process location.

gh

It uses as a statistic to monitor the process location the

Or is d

exponentially weighted moving average.

Th

Haphazard Sampling A sample scheme wherein the samples are collected using
an unsystematic, indiscriminant, unplanned, and/or chaotic
approach. This type of sampling is not appropriate for
process analysis or control because it can lead to a biased
result and consequently a possible erroneous decision.

Individual A single unit, or a single measurement of a characteristic,

often denoted by the symbol X.

194
 Appendix G
Glossary of Terms and Symbols

Inherent Variation See Variation.

Location A general term for the typical values of central tendency of

a distribution.

Loss Function A graphical representation of the relationship between the

customer’s sensitivity (loss) and deviations from the target
(design intent). This analysis is conducted without
considering the specifications.

din p
Mean A measure of location. The average of values in a group of

.
or ou
gly
measurements.

cc Gr
Inc

d a ion
rke Act
al
MCUSUM Control Chart The Multivariate Cumulative Sum Control Chart is the
on
application of the CUSUM Control Chart approach to

ma ry
ati

ter ust
multivariate situations.
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

Median A measure of location. The middle value in a group of

3/3
gn

measurements, when arranged from lowest to highest. If the

an e A n:
Ma

as m

number of values is even, by convention the average of the

o
o
i
d h ut
t

middle two values is used as the median. Subgroup

to:

ted y t ira

medians form the basis for a simple control chart for

xp
op en ite 426 sed

process location. Medians are designated by a tilde (~) over

ro ed E
n

t p wn se
yri t is Lic 2

the symbol for the individual values: X is the median of a

d i oc nt r: 4 lice

b
en

subgroup.
an is d ume be t is
5

tec
Th Doc Num men

MEWMA Control Chart The Multivariate Exponentially Weight Moving Average

s c um S
E- der ocu

Control Chart is the application of the EWMA Control

gh

Chart approach to multivariate situations.

Or is d
Th

Mode A measure of location defined by the value that occurs

most frequently in a distribution or data set (there may be
more than one mode within one data set).

Moving Range A measure of process spread. The difference between the

highest and lowest value among two or more successive
samples. As each additional datum point (sample) is
obtained, the range associated with that point is computed
by adding the new point and deleting the ‘oldest’
chronological point, so that each range calculation has at
least one shared point from the previous range calculation.

195
Appendix G
Glossary of Terms and Symbols

Typically, the moving range is used in concert with control

charts for individuals and uses two-point (consecutive
points) moving ranges.

Multivariate Control Chart The genre of control charts that have been developed to
monitor and control processes that are more appropriately
modeled with a multivariate distribution rather than
multiple univariate distributions.

Nonconforming Units Units which do not conform to a specification or other

inspection standard; p and np control charts are used to

din p
.
analyze systems producing nonconforming units.

or ou
gly
cc Gr
Inc

d a ion
Nonconformity A specific occurrence of a condition which does not

rke Act
al
conform to a specification or other inspection standard. An
on
individual nonconforming unit can have more than one

ma ry
ati

ter ust
nonconformity. For example, a door could have several
en e 11
rn

dents and dings plus a malfunctioning handle; a functional

be otiv /20
wa Ind
nte

check of a HVAC unit could reveal any of a number of

1
aI

3/3

potential discrepancies. c and u control charts are used to

gn

analyze systems producing nonconformities.

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Non-Normal Control Chart A control chart approach in which adjustments are made to
xp
op en ite 426 sed

the data or the control limits to allow process control

ro ed E
n

t p wn se

similar to that of Shewhart charts while compensating for

yri t is Lic 2
d i oc nt r: 4 lice

b
en

the characteristics of a non-normal distribution.

an is d ume be t is
5

tec
Th Doc Num men

Non-Normal Distribution A probability distribution that does not follow the normal
s c um S
E- der ocu

form; i.e., a distribution where the moments greater than

gh

order two are not all zero.

Or is d
Th

Normal Distribution A continuous, symmetrical, bell-shaped frequency

distribution for variables data that is the basis for the
control charts for variables.

Pre-Control An application of probabilistic analysis to product

(nonconformance) control using two data points within
each sample.

196
 Appendix G
Glossary of Terms and Symbols

Operational Definition A means of clearly communicating quality expectations

and performance; it consists of
(1) a criterion to be applied to an object or to a group,
(2) a test of the object or of the group,
(3) a decision: yes or no – the object or the group did or did
not meet the criterion.56

Over-adjustment Tampering; taking action on a process when the process is

actually in statistical control. Ascribing a variation or a
mistake to a special cause, when in fact the cause belongs
to the system (common causes).

din p
.
or ou
gly
cc Gr
Pareto Chart A simple tool for problem solving that involves ranking all

Inc

d a ion
potential problem areas or sources of variation according to
their contribution to cost or to total variation. Typically, a

rke Act
al
on
few causes account for most of the cost (or variation), so

ma ry
ati
problem-solving efforts are best prioritized to concentrate

ter ust
en e 11
rn

on the “vital few” causes, temporarily ignoring the “trivial

be otiv /20
wa Ind
nte

many”.
1
aI

3/3
gn

an e A n:
Ma

Point Estimate A statistic (single number) calculated from sample data

as m
o
o

(e.g., average or standard deviation) for which there is

i
d h ut
t
to:

ted y t ira

some expectation that it is “close” to the population

xp
op en ite 426 sed

parameter it estimates.
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Poisson Distribution A discrete probability distribution for attributes data that

an is d ume be t is
5

applies to nonconformities and underlies the c and u control

tec
Th Doc Num men

charts.
s c um S
E- der ocu

gh
Or is d

Prediction Interval Once a regression model is established for a population, the

Th

response, y , can be predicted for future values (samples) of

the regressor variable(s), x0, x1, … xn.
The interval for (100 - α) confidence in this prediction is
called the prediction interval.

Prevention A proactive (future-oriented) strategy that improves quality

and productivity by directing analysis and action toward
correcting the process itself. Prevention is consistent with a
philosophy of continual improvement (see also Detection).

56
See Deming (1982).

197
Appendix G
Glossary of Terms and Symbols

Probability Based Charts An approach which uses analysis and charts based on
categorical data and the probabilities related to the
categories for the control and analysis of products and
processes.
Probability Sampling See Random Sampling.

Problem Solving The process of moving from symptoms to causes (special

or common) to actions. Among the basic techniques that
can be used are Pareto charts, cause-and-effect diagrams
and statistical process control techniques.

din p
.
or ou
gly
Process The combination of people, equipment, materials, methods,

cc Gr
measurement and environment that produce output – a

Inc

d a ion
given product or service. A process can involve any aspect
of the business. “6M’s” is a catch phrase sometimes used

rke Act
al
on
to describe a process: Man, Material, Method, Machine,

ma ry
ati
Mother Nature, Measurement.

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

Process Average The location of the distribution of measured values of a

3/3
gn

particular process characteristic, usually designated as an

an e A n:
Ma

as m

overall average, X .
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

Process Capability The 6 σˆ range of inherent process variation.

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is

Variables Data Case

5

tec

This is defined as 6 σˆ C .
Th Doc Num men

o
s c um S
E- der ocu

Attributes Data Case

gh

This is usually defined as the average proportion or rate of

Or is d

nonconformances or nonconformities.
Th

Process Control See Statistical Process Control.

Process Performance The 6 σˆ range of total process variation.

Process Spread The extent to which the distribution of individual values of

the process characteristic vary; often shown as the process
average plus or minus some number of standard deviations
( X ± 3 σˆ ).

198
 Appendix G
Glossary of Terms and Symbols

Quadratic Of or pertaining to a second order mathematical model; a

common graphical example is a parabola.

Randomness A condition in which no pattern in the data can be

discerned.

Random Sampling A random sample is one in which every sample point has
the same chance (probability) of being selected. A random
sample is systematic and planned; that is, all sample points
are determined before any data are collected.

din p
.
or ou
gly
The process of selecting units for a sample of size n, in

cc Gr
such a manner that each n unit under consideration has an

Inc

d a ion
equal chance of being selected in the sample.

rke Act
al
on
Convenience Sampling:

ma ry
ati

ter ust
See Convenience Sampling
en e 11
rn

be otiv /20
wa Ind
nte

Haphazard Sampling:
1
aI

See Haphazard Sampling

3/3
gn

Range A measure of process spread. The difference between the

an e A n:
Ma

as m
o
o

highest and lowest values in a subgroup, a sample, or a

i
d h ut
t
to:

ted y t ira

population.
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Rational Subgroup A subgroup gathered in such a manner as to give the

b
en

maximum chance for the measurements in each subgroup

an is d ume be t is
5

to be alike and the maximum chance for the subgroups to

tec
Th Doc Num men

differ one from the other. This subgrouping scheme enables

s c um S

a determination of whether the process variation includes

E- der ocu

special cause variation.

gh
Or is d
Th

Regression Control Chart Regression Control Charts are used to monitor the
relationship between two correlated variables in order to
determine if and when deviation from the known
predictable relationship occurs.

Residuals Control Chart A chart that monitors a process using the residuals
(differences) between a fitted model and the data. A
process shift will cause a shift in the mean of the residuals.

199
Appendix G
Glossary of Terms and Symbols

Run A consecutive number of points consistently increasing or

decreasing, or above or below the centerline. This can be
evidence of the existence of special causes of variation.

Sample See Subgroup.

Shape A general concept for the overall pattern formed by a

distribution of values. Shape involves many characteristics
such as symmetry (skewness) and peakedness (kurtosis).

din p
Short Run Control Chart A control chart approach in which adjustments are made to

.
or ou
gly
the data or the control limits to allow process control

cc Gr
similar to that of Shewhart charts for processes that only

Inc

d a ion
produce a small number of products during a single run

rke Act
al
on

ma ry
ati
Sigma (σ )

ter ust
The Greek letter used to designate a standard deviation of a
en e 11
rn

be otiv /20
wa Ind
nte

population.
1
aI

3/3
gn

Special Cause A source of variation that affects only some of the output of
an e A n:
Ma

as m
o
o

the process; it is often intermittent and unpredictable. A

i
d h ut
t
to:

ted y t ira

special cause is sometimes called assignable cause. It is

xp
op en ite 426 sed

signaled by one or more points beyond the control limits or

h
ro ed E

a non-random pattern of points within the control limits.

n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

Specification The engineering requirement for judging acceptability of a

tec
Th Doc Num men

particular characteristic. A specification must never be

s c um S

confused with a control limit. Ideally, a specification ties

E- der ocu

directly to or is compatible with the customer’s (internal

gh
Or is d

and/or external) requirements and expectations.

Th

Bilateral:
A bilateral specification identifies requirements at both
extremes of the process range. Often referred to as a two-
sided specification or tolerance.
Unilateral:
A unilateral specification identifies requirements at only
one extreme of the process range. Often referred to as a
one-sided specification or tolerance.

200
 Appendix G
Glossary of Terms and Symbols

Spread The expected span of values from smallest to largest in a

distribution (see also Process Spread).

Stability The absence of special causes of variation; the property of

being in statistical control.

Stable Process A process that is in statistical control.

Standard Deviation A measure of the spread of the process output or the spread
of a sampling statistic from the process (e.g., of subgroup

din p
averages).

.
or ou
gly
cc Gr
Inc

d a ion
Statistic A value calculated from or based upon sample data (e.g., a

rke Act
al
subgroup average or range) used to make inferences about
on
the process that produced the output.

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Statistical Control The condition describing a process from which the effect of
1
aI

all special causes of variation have been eliminated and

3/3
gn

only that due to common causes remain; i.e., observed

an e A n:
Ma

as m

variation can be attributed to a constant system of chance

o
o
i
d h ut
t

causes. This is evidenced on a control chart by the absence

to:

ted y t ira

of points beyond the control limits and by the absence of

xp
op en ite 426 sed

non-random patterns within the control limits.

ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is

Statistical Inference Information about population parameters is estimated or

5

tec

inferred from data obtained from a sample of that

Th Doc Num men

population. These inferences can be in the form of a single

s c um S
E- der ocu

number (point estimate) or a pair of numbers (interval

gh

estimate).
Or is d
Th

Statistical Process Control The use of statistical techniques such as control charts to
analyze a process or its output so as to take appropriate
actions to achieve and maintain a state of statistical control
and to improve the process capability.

Statistical Tolerance Limits An interval or range of values that is expected to contain a

specified proportion of a population. See Montgomery
(1997) or Juran and Godfrey (1999) for calculation
methods. See Tolerance Interval.

201
Appendix G
Glossary of Terms and Symbols

Stoplight Control Chart A probability based chart approach to process control that
uses three categories and double sampling. In this approach
the target area is designated green, the warning areas as
yellow, and the stop zones as red. The use of these colors
gives rise to the “stoplight” designation.

Subgroup One or more observations or measurements used to analyze

the performance of a process. Rational subgroups are
usually chosen so that the variation represented within each
subgroup is as small as feasible for the process
(representing the variation from common causes), and so
that any changes in the process performance (i.e., special

din p
.
or ou
causes) will appear as differences between subgroups.

gly
cc Gr
Rational subgroups are typically made up of consecutive
pieces, although random samples are sometimes used.

Inc

d a ion
rke Act
al
Tolerance See Specification. on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Tolerance Interval See Statistical Tolerance Limits.

1
aI

3/3
gn

an e A n:
Ma

as m

Total Process Variation See Variation.

o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

Type I Error Rejecting an assumption that is true; e.g., taking action

ro ed E
n

t p wn se

appropriate for a special cause when in fact the process has

yri t is Lic 2
d i oc nt r: 4 lice

b
en

not changed (over-control). This is associated with the

an is d ume be t is

producer’s or alpha risk.

5

tec
Th Doc Num men

o
s c um S
E- der ocu

Type II Error Failing to reject an assumption that is false; e.g., not taking
gh

appropriate action when in fact the process is affected by

Or is d

special causes (under-control). This is associated with the

Th

consumer’s risk or beta risk.

Unimodal A distribution is said to be unimodal if it has only one

mode.

Variables Data Quantitative data, where measurements are used for

analysis. Examples include the diameter of a bearing
journal in millimeters, the closing effort of a door in
Newtons, the concentration of electrolyte in percent, or the
torque of a fastener in Newton-meters. X and R , X and
s , median and range, and individuals and moving range
control charts are used for variables data. See also

202
 Appendix G
Glossary of Terms and Symbols

Attributes Data. (The term “Variables”, although

awkward sounding, is used in order to distinguish the
difference between something that varies, and the control
chart used for data taken from a continuous variable).

Variation The inevitable differences among individual outputs of a

process; the sources of variation can be grouped into two
major classes: Common Causes and Special Causes.
Inherent Variation:
That process variation due to common causes only.
Within-subgroup Variation:

din p
.
This is the variation due only to the variation within the

or ou
gly
subgroups. If the process is in statistical control this variation is

cc Gr
a good estimate of the inherent process variation. It can be

Inc

d a ion
R s
estimated from control charts by d 2 or c4 .

rke Act
al
on

ma ry
ati
Between-subgroup Variation:

ter ust
en e 11
rn

This is the variation due to the variation between

be otiv /20
wa Ind
nte

subgroups. If the process is in statistical control this

1
aI

variation should be zero.

3/3
gn

Total Process Variation:

an e A n:
Ma

as m

This is the variation due to both within-subgroup and

o
o
i
d h ut
t
to:

between-subgroup variation. If the process is not in

ted y t ira

statistical control the total process variation will include the

xp
op en ite 426 sed

effect of the special cause(s) as well as the common causes.

ro ed E
n

t p wn se

This variation may be estimated by s, the sample standard

yri t is Lic 2
d i oc nt r: 4 lice

b
en

deviation, using all of the individual readings obtained

an is d ume be t is

from either a detailed control chart or a process study:

5

tec
Th Doc Num men

( xi − x )2
n
σˆ P = s = ∑i n − 1 where xi is an individual reading,
s c um S
E- der ocu

gh

x is the average of the individual readings, and n is the

Or is d

total number of individual readings.

Th

Within-subgroup Variation See Variation.

Zone Analysis This is a method of detailed analysis of a Shewhart control

chart which divides the X chart between the control limits
into three equidistant zones above the mean and three
equidistant zones below the mean.57

57
See also AT&T. (1984) .

203
Appendix G
Glossary of Terms and Symbols

Symbols
as Used in This Manual
A2 A multiplier of R used to calculate the control limits for
averages; tabled in Appendix E.

A2 A multiplier of R used to calculate the control limits for

medians; tabled in Appendix E.

din p
.
or ou
A3 A multiplier of s used to calculate the control limits for

gly
cc Gr
averages; tabled in Appendix E.

Inc

d a ion
rke Act
al
on
B3 , B4 Multipliers of s used to calculate the lower and upper

ma ry
ati
control limits, respectively, for sample standard deviations;

ter ust
en e 11
rn

tabled in Appendix E.
be otiv /20
wa Ind
nte

1
aI

3/3
gn

c The number of nonconformities in a sample. The c chart is

an e A n:
Ma

as m

described in Chapter II, Section C.

o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

c The average number of nonconformities in samples of

h
ro ed E
n

constant size n.
t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec

c4 The divisor of s used to estimate the process standard

Th Doc Num men

deviation; tabled in Appendix E.

s c um S
E- der ocu

gh
Or is d

Cp The capability index for a stable process, typically defined

Th

(USL − LSL)
as .
6σˆ C

C pk The capability index for a stable process, typically defined

as the minimum of CPU or CPL.

CPL The lower capability index, typically defined as

( X − LSL)
.
3σˆ C

204
 Appendix G
Glossary of Terms and Symbols

CPU The upper capability index, typically defined as

(USL − X )
.
3σˆ C

CR The capability ratio for a stable process, typically defined

1
as .
Cp

din p
d2

.
A divisor of R used to estimate the process standard

or ou
gly
cc Gr
deviation; tabled in Appendix E.

Inc

d a ion
rke Act
al
D3 , D4 Multipliers of R used to calculate the lower and upper
on

ma ry
ati
control limits, respectively, for ranges; tabled in Appendix

ter ust
en e 11
rn

E. be otiv /20
wa Ind
nte

1
aI

3/3
gn

E2 A multiplier of the average moving range, MR ,used to

an e A n:
Ma

as m

calculate control limits for individuals; tabled in Appendix

o
o
i
d h ut
t
to:

ted y t ira

E.
xp
op en ite 426 sed

h
ro ed E
n

t p wn se

k
yri t is Lic 2

The number of subgroups being used to calculate control

d i oc nt r: 4 lice

b
en

limits.
an is d ume be t is
5

tec
Th Doc Num men

LCL The lower control limit; LCLX , LCLR , LCLp , etc., are,
s c um S
E- der ocu

respectively, the lower control limits for averages, ranges,

gh
Or is d

proportion nonconforming, etc.

Th

LSL The lower engineering specification limit.

MR The moving range of a series of data points used primarily

on a chart for individuals.

n The number of individuals in a subgroup; the subgroup

sample size.

205
Appendix G
Glossary of Terms and Symbols

n The average subgroup sample size; typically used in

attributes charts with varying subgroup sample sizes

np The number of nonconforming items in a sample of size n.

The np chart is described in Chapter II, Section C.

np The average number of nonconforming items in samples of

constant size n.

p The proportion of units nonconforming in a sample. The p-

din p
chart is discussed in Chapter II, Section C.

.
or ou
gly
cc Gr
Inc

d a ion
p The average proportion of units nonconforming in a series

rke Act
al
of samples.
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20 (USL − LSL)

wa Ind
nte

Pp The performance index, typically defined as .

6σˆ P
1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i

Ppk The performance index, typically defined as the minimum

d h ut
t
to:

ted y t ira

of PPU or PPL.
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

PPL The lower performance index, typically defined as

an is d ume be t is
5

( X − LSL)
tec

.
Th Doc Num men

3σˆ P
s c um S
E- der ocu

gh
Or is d

PPU The upper performance index, typically defined as

Th

(USL − X )
.
3σˆ P

1
PR The performance ratio, typically defined as .
Pp

pz The proportion of output beyond a point of interest, such as

a particular specification limit, z standard deviation units
away from the process average.

206
 Appendix G
Glossary of Terms and Symbols

R The subgroup range (highest minus lowest value); the R

chart is discussed in Chapter II, Section C.

R The average range of a series of subgroups of constant size.

s The sample standard deviation for subgroups; the s-chart is

discussed in Chapter II, Section C.

din p
The sample standard deviation for processes; s is discussed

.
or ou
gly
in Chapter IV, Section A.

cc Gr
Inc

d a ion
rke Act
s The average sample standard deviation of a series of

al
on
subgroups, weighted if necessary by sample size.

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

SL A unilateral engineering specification limit.

1
aI

3/3
gn

an e A n:
Ma

as m

u The number of nonconformities per unit in a sample which

o
o

may contain more than one unit. The u chart is discussed in

i
d h ut
t
to:

ted y t ira

Chapter II, Section C.

xp
op en ite 426 sed

h
ro ed E

u The average number of nonconformities per unit in samples

n

t p wn se
yri t is Lic 2

not necessarily of the same size.

d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

UCL The upper control limit; UCLX , UCLR , UCLp , etc., are,
o
s c um S

respectively, the upper control limits for averages, ranges,

E- der ocu

proportion nonconforming, etc.

gh
Or is d
Th

USL The upper engineering specification limit.

X An individual value. The chart for individuals is discussed

in Chapter II, Section C.

X The average of values in a subgroup. The X -chart is

discussed in Chapter II, Section C.

207
Appendix G
Glossary of Terms and Symbols

X The average of subgroup averages (weighted if necessary

by sample size); the measured process average.

X The median of values in a subgroup; the chart for medians

is discussed in Chapter II, Section C. This is pronounced as
“x tilde”.

X The average of subgroup medians; the estimated process

median. This is pronounced as “x tilde bar”.

din p
.
or ou
gly
cc Gr
z The number of standard deviation units from the process

Inc

d a ion
average to a value of interest such as an engineering

rke Act
al
specification. When used in capability assessment, zUSL is
on

ma ry
the distance to the upper specification limit, zLSL is the
ati

ter ust
en e 11
rn

distance to the lower specification limit, and zmin is the

be otiv /20
wa Ind
nte

distance to the nearest specification limit.

1
aI

3/3
gn

an e A n:
Ma

as m

σ The Greek letter sigma used to designate a standard

o
o
i
d h ut
t

deviation of a population.
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E

σ x , σ R , σ np , etc.
n

t p wn se

The standard deviation of a statistic based on sample

yri t is Lic 2
d i oc nt r: 4 lice

b
en

process output, such as the standard deviation of the

an is d ume be t is
5

distribution of subgroup averages the standard deviation of

tec
Th Doc Num men

the distribution of subgroup ranges, the standard deviation

s c um S

of the distribution of number of nonconforming items, etc.

E- der ocu

gh
Or is d

σˆ An estimate of the standard deviation of a process

Th

characteristic.

σˆ P The estimate of the standard deviation of a process using

the sample standard deviation of a set of individuals about
the average of the set. This is an estimate of the total
process variation of the process.

σˆ C The estimate of the standard deviation of a stable process

using the average range of subgrouped samples taken from
the process, usually within the context of control charts,

208
 Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

209
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
variation of the process.

wa Ind
ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
Appendix G

where the d 2 factor is tabled in Appendix E. This is the

Glossary of Terms and Symbols

within-subgroup variation and an estimate of the inherent

 Appendix G

Th
Or is d
Glossary of Terms and Symbols

E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h t i o rn

210
an e A n: ati
d h ut on
as m o 3/3
1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 APPENDIX H
References and Suggested Readings

APPENDIX H
References and Suggested Readings58

American National Standards Committee Z-1 (1996). Guide for Quality Control Charts; Control Chart
Method of Analyzing Data; Control Chart Method of Controlling Quality During Production, American
Society for Quality (ASQ), Milwaukee, WI.

ASQC Statistics Division, Statistical “How-To” Techniques Series, ASQC Quality Press (15 Volumes),
1979-1991.

din p
.
or ou
gly
ASQC (1987). Definitions, Symbols, Formulas, and Tables for Control Charts, ANSI/ASQC A1-1987.

cc Gr
Inc

d a ion
ASTM International (2002). Manual on Presentation of Data and Control Chart Analysis (STP-15D),

rke Act
al
7th edition.59
on

ma ry
ati

ter ust
AT&T Technologies, Inc. (1984). Statistical Quality Control Handbook. AT&T Technologies, Inc.

en e 11
rn

(originally published by Western Electric Co., Inc., 1956).

be otiv /20
wa Ind
nte

1
aI

Bhote, K.R. (1991). World Class Quality. AMACOM, New York.

3/3
gn

an e A n:
Ma

as m

Bothe, D. (2001). Measuring Process Capability. Landmark Publishing, Cedarburg, WI.

o
o
i
d h ut
t
to:

ted y t ira

Box, G.E.P., Jenkins, G.M. and Reinsel, G.C. (1994). Time Series Analysis, Forecasting and Control.
xp
op en ite 426 sed

Third Edition. Prentice Hall.

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Bissell, B.A.F. (1990). “How Reliable Is Your Capability Index?”, Applied Statistics, Vol. 39, 1990, pp.
an is d ume be t is

331-340.
5

tec
Th Doc Num men

Boyles, R.A. (1991). “The Taguchi Capability Index”, Journal of Quality Technology, Vol. 23, 1991, pp.
s c um S
E- der ocu

17-26.
gh
Or is d

Box, G.E.P. and Cox, D.R. (1964). “An Analysis of Transformations”, J. Royal Stat. Soc., Series B,
Th

V26, p. 211

Brase, C.H., and Brase, C.P. (1999). Understanding Statistics, 6th edition, Houghton Mifflin Company,
New York.

Burr, I.W. (1942). “Cumulative Frequency Distributions”, Annals of Mathematical Statistics, Vol. 13,
pp. 215-232.

Burr, I.W. (1967). “The Effect Of Non-Normality On Constants For Charts”, Industrial Quality Control,
May, pp. 563-569.

58
See Freund and Williams (1966) for an extensive listing of statistical terms and definitions.
59
Available from the ASTM International, 100 Barr Drive, West Conshohocken, PA 19428-2959.

211
APPENDIX H
References and Suggested Readings

Chan, L.J., Cheng, S.W. and. Spiring, F.A. (1988). “A New Measure of Process Capability: Cpm”,
Journal of Quality Technology, Vol. 20, No. 3, 1988, pp. 162-175.

Chan, L.K., and Cui, H.J. (2003). “Skewness Correction charts for Skewed Distributions”, Naval
Research Logistics, Volume 50, Issue 6, pp. 555 - 573.

Champ, C.W., and Rigdon, S.E. (1997). “An Analysis of the Run Sum Control Chart”, Journal of
Quality Technology, Vol. 29, No. 4.

Charbonneau, H.C. and Gordon, L.W. (1978). Industrial Quality Control, Prentice-Hall, Inc.

Chrysler, Ford, and General Motors (1995). Advanced Product Quality Planning Manual, AIAG.

din p
.
or ou
gly
DaimlerChrysler, Ford, and General Motors (2003). Measurement Systems Analysis Reference Manual,

cc Gr
AIAG.

Inc

d a ion
Davis, R.B., Homer, A., and Woodall, W.H. (1990). “Performance of the Zone Control Chart”,

rke Act
al
Communications in Statistics − Theory and Methods, 19, pp. 1581-1587.
on

ma ry
ati

ter ust
en e 11
rn

Deming, W. Edwards (1950). Some Theory of Sampling, Dover Publications Inc., New York.
be otiv /20
wa Ind
nte

1
aI

Deming, W. Edwards (1967). “What Happened in Japan?”, Industrial Quality Control, Vol. 24, No. 3,
3/3
gn

August, pp. 89-93.

an e A n:
Ma

as m
o
o

Deming, W. Edwards (1982). Quality, Productivity and Competitive Position, Massachusetts Institute of
i
d h ut
t
to:

ted y t ira

Technology, Center for Advanced Engineering Study.

xp
op en ite 426 sed

h
ro ed E

Deming, W. Edwards (1989). Out of the Crisis, Massachusetts Institute of Technology, Center for
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Advanced Engineering Study.

en
an is d ume be t is
5

Deming, W. Edwards (1994). New Economics: for Industry, Government, Education, Massachusetts
tec
Th Doc Num men

Institute of Technology, Center for Advanced Engineering Study.

s c um S
E- der ocu

Dixon, W.J and Massey, F.J., Jr. (1969). Introduction to Statistical Analysis, Third Edition,
gh
Or is d

McGraw-Hill Book Co., New York.

Th

Doty, L.A. (1991). Statistical Process Control, ASQ Quality Press, Milwaukee WI.

Dovich, R.A. (1991). Statistical Terrorists, Quality in Manufacturing Magazine, March-April, 1991.

Duncan, A.J. (1974). Quality Control and Industrial Statistics, 5th ed., Richard D. Irwin, Inc.

English, J.R., Lee, S., Martin, T.W., Tilmon, C. (2000). “Detecting Changes In Autoregressive Processes
with X and EWMA charts”, IIE Transactions, December.

Farnum, N.R. (1992). “Control Charts for Short Runs: Nonconstant Process and Measurement Error”,
Journal of Quality Technology, Vol. 24.

212
 APPENDIX H
References and Suggested Readings

Fellers, G., (1991). SPC for Practitioners: Special Cases and Continuous Processes, ASQ Quality Press,
Milwaukee, WI.

Freund, J.E. and Williams, F.J. (1966). Dictionary/Outline of Basic Statistics, Reprint. Originally
published: McGraw-Hill, New York. Dover edition 1991.

General Motors Corporation (1991). Key Characteristics Designation System, GM-1805 QN.

Grant, E.L. and. Leavenworth, R.S. (1980). Statistical Quality Control, 7th ed., McGraw-Hill, Inc.

Gruska, G.F., Mirkhani, K., and Lamberson, L.R. (1973). Point Estimation Samples, The Third
Generation, Inc. Troy, MI.

din p
.
or ou
gly
cc Gr
Gruska, G.F. (2004). Enumerative vs. Analytic Studies, Omnex, Ann Arbor, MI.

Inc

d a ion
Gunter, B. (1989). “Use and Abuse of Cpk”, 4 parts, Quality Progress, January 1989, March 1989, May

rke Act
al
on
1989 and July 1989.

ma ry
ati

ter ust
en e 11
rn

Heaphy, M.S. and Gruska, G.F., (1982). “Markovian Analysis of Sequenced Sampling”, 36th AQC
be otiv /20
wa Ind
nte

Transactions, ASQC.
1
aI

3/3

Herman, J.T. (1989). “Capability Index-Enough for Process Industries?”, Proceedings, ASQC 43rd
gn

AQC.
an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira

Ishikawa, K. (1976). Guide to Quality Control, Asian Productivity Organization, Revised Edition.
xp
op en ite 426 sed

h
ro ed E

Jaehn, A.H. (1991). “The Zone Control Chart”, Quality Progress, Vol. 24, No. 7, pp. 65-68.
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en

Johnson, N.L. (1949). “Systems of Frequency Curves Generated by Methods of Translation,”

an is d ume be t is
5

Biometrika, Vol 36, pp. 149-176.

tec
Th Doc Num men

o
s c um S

Juran, J. and Godfrey A.B. (1999). Quality Handbook, 5th ed., McGraw-Hill, Inc.
E- der ocu

gh
Or is d

Kane, V.E. (1989). Defect Prevention- Use of Simple Statistical Tools, Marcel Dekker, Inc. and ASQC
Quality Press.
Th

Keats, J.B. and Montgomery D. C. (1991). Statistical Process Control in Manufacturing, ASQ Quality
Press, Milwaukee, WI.

Kourti, T., MacGregor, J.F. (1996). “Multivariate SPC Methods for Process and Product Monitoring”,
Journal of Quality Technology, Vol. 28, No. 4.

Lowry, C.A., Woodall, W.H., Champ, C.W., and Rigdon, S.E. (1992). “A Multivariate Exponentially
Weighted Moving Average Control Chart”, Technometrics, 34, pp. 46-53.

Lowry, C.A. and Montgomery, D.C. (1995). “A Review Of Multivariate Control Charts”, IIE
Transactions, 27, pp. 800.

213
APPENDIX H
References and Suggested Readings

Mauch, P.D. (1991). Basic SPC: A Guide For the Service Industries, ASQ Quality Press, Milwaukee,
WI.

Mason, R.I. and Young, J.C. (2001). “Implementing Multivariate Statistical Process Control Using
Hotelling’s T2 Statistic”, Quality Progress.

Montgomery, D.C. (1997). Introduction to Statistical Quality Control, 3rd ed., John Wiley, New York.

Ott, E.R. (1975). Process Quality Control, McGraw-Hill, Inc.

Pham, H., (2001). Recent Advances In Reliability And Quality Engineering, Series on Quality,
Reliability and Engineering Statistics - Vol. 2, World Scientific.

din p
.
or ou
gly
Reynolds, J.H. (1971). “The Run Sum Control Chart Procedure”, Journal of Quality Technology Vol 3,

cc Gr
pp. 23-27.

Inc

d a ion
Roberts, S.W. (1966). “A Comparison of Some Control Chart Procedures”, Technometrics, Vol 8, pp.

rke Act
al
on
411-430.

ma ry
ati

ter ust
en e 11
rn

Scherkenbach, W.W. (1991). Deming's Road to Continual Improvement, SPC Press, Knoxville TN.
be otiv /20
wa Ind
nte

1
aI

Shewhart, Walter A. (1931). Economic Control of Quality of Manufactured Product, Van Nostrand;
3/3

republished ASQ Quality Press , (1980), Milwaukee, WI..

gn

an e A n:
Ma

as m
o
o

Spiring, F.A (1991). Assessing Process Capability in the Presence of Systematic Assignable Cause,
i
d h ut
t
to:

ted y t ira

Journal of Quality Technology, Vol. 23, No. 2, April, 1991.

xp
op en ite 426 sed

h
ro ed E

Wadsworth, H.M. (1989). Handbook of Statistical Methods for Engineers and Scientists, McGraw-Hill,
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

New York.
b
en
an is d ume be t is
5

Wheeler, D.J. (1991). Short run SPC, SPC Press, Knoxville, TN.
tec
Th Doc Num men

o
s c um S

Wheeler, D.J. (1995). Advanced Topics in Statistical Process Control, SPC Press, Knoxville TN.
E- der ocu

gh
Or is d

Wheeler, D.J. and Chambers, D.S. (1986). Understanding Statistical Process Control, 2nd ed., SPC
Press, Knoxville, TN.
Th

Wheeler, D.J. (1999). Beyond Capability Confusion, SPC Press, Knoxville, TN.

Wise, S.A. and Fair, D.C. (1998). Innovative Control Charting, ASQ Quality Press, Milwaukee, WI.

214
APPENDIX I
Standard Normal Tables

APPENDIX I
Standard Normal Tables
z x.x0 x.x1 x.x2 x.x3 x.x4 x.x5 x.x6 x.x7 x.x8 x.x9
0.0 0.50000000 0.50398940 0.50797830 0.51196650 0.51595340 0.51993880 0.52392220 0.52790320 0.53188140 0.53585640
0.1 0.53982780 0.54379530 0.54775840 0.55171680 0.55567000 0.55961770 0.56355950 0.56749490 0.57142370 0.57534540
0.2 0.57925970 0.58316620 0.58706440 0.59095410 0.59483490 0.59870630 0.60256810 0.60641990 0.61026120 0.61409190

din p
0.3 0.61791140 0.62171950 0.62551580 0.62930000 0.63307170 0.63683070 0.64057640 0.64430880 0.64802730 0.65173170

.
or ou
gly
0.4 0.65542170 0.65909700 0.66275730 0.66640220 0.67003140 0.67364480 0.67724190 0.68082250 0.68438630 0.68793310

cc Gr
0.5 0.69146250 0.69497430 0.69846820 0.70194400 0.70540150 0.70884030 0.71226030 0.71566120 0.71904270 0.72240470

Inc

d a ion
0.6 0.72574690 0.72906910 0.73237110 0.73565270 0.73891370 0.74215390 0.74537310 0.74857110 0.75174780 0.75490290

rke Act
al
0.7 0.75803630 0.76114790 0.76423750 0.76730490 0.77035000 0.77337260 0.77637270 0.77935010 0.78230460 0.78523610

on

ma ry
ati
0.8 0.78814460 0.79102990 0.79389190 0.79673060 0.79954580 0.80233750 0.80510550 0.80784980 0.81057030 0.81326710

ter ust
1
rn

1
0.9 0.81593990 0.81858870 0.82121360 0.82381450 0.82639120 0.82894390 0.83147240 0.83397680 0.83645690 0.83891290

be tiv 0
wa nd
nte

1/2
1.0 0.84134470 0.84375240 0.84613580 0.84849500 0.85083000 0.85314090 0.85542770 0.85769030 0.85992890 0.86214340

I
aI

en e
3
1.1 0.86433390 0.86650050 0.86864310 0.87076190 0.87285680 0.87492810 0.87697560 0.87899950 0.88099990 0.88297680

/
gn

3
o
1.2 0.88493030 0.88686060 0.88876760 0.89065140 0.89251230 0.89435020 0.89616530 0.89795770 0.89972740 0.90147470

an he A tion:
Ma

as m o
1.3 0.90319950 0.90490210 0.90658250 0.90824090 0.90987730 0.91149200 0.91308500 0.91465650 0.91620670 0.91773560

d h ut
to:

ted by t pira
1.4 0.91924330 0.92073020 0.92219620 0.92364150 0.92506630 0.92647070 0.92785500 0.92921910 0.93056340 0.93188790

op en ite 426 sed

x
1.5 0.93319280 0.93447830 0.93574450 0.93699160 0.93821980 0.93942920 0.94062010 0.94179240 0.94294660 0.94408260

ro ed E
n
1.6 0.94520070 0.94630110 0.94738390 0.94844930 0.94949740 0.95052850 0.95154280 0.95254030 0.95352130 0.95448600

t p own nse
yri t is Lic 2
d i oc nt r: 4 lice

1.7 0.95543450 0.95636710 0.95728380 0.95818490 0.95907050

e 0.95994080 0.96079610 0.96163640 0.96246200 0.96327300
an is d ume be t is
5

1.8 0.96406970 0.96485210 0.96562050 0.96637500 0.96711590 0.96784320 0.96855720 0.96925810 0.96994600 0.97062100
tec
Th Doc Num men

1.9 0.97128340 0.97193340 0.97257110 0.97319660 0.97381020 0.97441190 0.97500210 0.97558080 0.97614820 0.97670450
s c um S

2.0 0.97724990 0.97778440 0.97830830 0.97882170 0.97932480 0.97981780 0.98030070 0.98077380 0.98123720 0.98169110
E- der ocu

gh

2.1 0.98213560 0.98257080 0.98299700 0.98341420 0.98382260 0.98422240 0.98461370 0.98499660 0.98537130 0.98573790
Or is d

2.2 0.98609660 0.98644740 0.98679060 0.98712630 0.98745450 0.98777550 0.98808940 0.98839620 0.98869620 0.98898930
Th

2.3 0.98927590 0.98955590 0.98982960 0.99009690 0.99035810 0.99061330 0.99086250 0.99110600 0.99134370 0.99157580
2.4 0.99180250 0.99202370 0.99223970 0.99245060 0.99265640 0.99285720 0.99305310 0.99324430 0.99343090 0.99361280
2.5 0.99379030 0.99396340 0.99413230 0.99429690 0.99445740 0.99461390 0.99476640 0.99491510 0.99506000 0.99520120
2.6 0.99533880 0.99547290 0.99560350 0.99573080 0.99585470 0.99597540 0.99609300 0.99620740 0.99631890 0.99642740
2.7 0.99653300 0.99663580 0.99673590 0.99683330 0.99692800 0.99702020 0.99710990 0.99719720 0.99728210 0.99736460
2.8 0.99744490 0.99752290 0.99759880 0.99767260 0.99774430 0.99781400 0.99788180 0.99794760 0.99801160 0.99807380
2.9 0.99813420 0.99819290 0.99824980 0.99830520 0.99835890 0.99841110 0.99846180 0.99851100 0.99855880 0.99860510
3.0 0.99865010 0.99869380 0.99873610 0.99877720 0.99881710 0.99885580 0.99889330 0.99892970 0.99896500 0.99899920

215
APPENDIX I
Standard Normal Tables

z x.x0 x.x1 x.x2 x.x3 x.x4 x.x5 x.x6 x.x7 x.x8 x.x9
3.1 0.99903240 0.99906460 0.99909570 0.99912600 0.99915530 0.99918360 0.99921120 0.99923780 0.99926360 0.99928860
3.2 0.99931290 0.99933630 0.99935900 0.99938100 0.99940240 0.99942300 0.99944290 0.99946230 0.99948100 0.99949910
3.3 0.99951660 0.99953350 0.99954990 0.99956580 0.99958110 0.99959590 0.99961030 0.99962420 0.99963760 0.99965050
3.4 0.99966310 0.99967520 0.99968690 0.99969820 0.99970910 0.99971970 0.99972990 0.99973980 0.99974930 0.99975850
3.5 0.99976740 0.99977590 0.99978420 0.99979220 0.99979990 0.99980740 0.99981460 0.99982150 0.99982820 0.99983470
3.6 0.99984090 0.99984690 0.99985270 0.99985830 0.99986370 0.99986890 0.99987390 0.99987870 0.99988340 0.99988790
3.7 0.99989220 0.99989640 0.99990040 0.99990430 0.99990800 0.99991160 0.99991500 0.99991840 0.99992160 0.99992470

din p
3.8 0.99992770 0.99993050 0.99993330 0.99993590 0.99993850 0.99994090 0.99994330 0.99994560 0.99994780 0.99994990

.
or ou
gly
3.9 0.99995190 0.99995390 0.99995570 0.99995750 0.99995930 0.99996090 0.99996250 0.99996410 0.99996550 0.99996700

cc Gr
4.0 0.99996830 0.99996960 0.99997090 0.99997210 0.99997330 0.99997440 0.99997550 0.99997650 0.99997750 0.99997840

Inc

d a ion
4.1 0.99997930 0.99998020 0.99998110 0.99998190 0.99998260 0.99998340 0.99998410 0.99998480 0.99998540 0.99998610

rke Act
al
on
4.2 0.99998670 0.99998720 0.99998780 0.99998830 0.99998880 0.99998930 0.99998980 0.99999020 0.99999070 0.99999110

ma ry
ati
4.3 0.99999150 0.99999180 0.99999220 0.99999250 0.99999290 0.99999320 0.99999350 0.99999380 0.99999410 0.99999430

ter ust
1
rn

1
4.4 0.99999460 0.99999480 0.99999510 0.99999530 0.99999550 0.99999570 0.99999590 0.99999610 0.99999630 0.99999640

be tiv 0
wa nd
nte

1/2
I
aI

en e
3
4.5 0.9999966023 0.9999967586 0.9999969080 0.9999970508 0.9999971873 0.9999973177 0.9999974423 0.9999975614 0.9999976751 0.9999977838

/
gn

3
o
4.6 0.9999978875 0.9999979867 0.9999980813 0.9999981717 0.9999982580 0.9999983403 0.9999984190 0.9999984940 0.9999985656 0.9999986340

an he A tion:
Ma

as m o
4.7 0.9999986992 0.9999987614 0.9999988208 0.9999988774 0.9999989314 0.9999989829 0.9999990320 0.9999990789 0.9999991235 0.9999991661

d h ut
to:

ted by t pira
4.8 0.9999992067 0.9999992453 0.9999992822 0.9999993173 0.9999993508 0.9999993827 0.9999994131 0.9999994420 0.9999994696 0.9999994958

op en ite 426 sed

x
4.9 0.9999995208 0.9999995446 0.9999995673 0.9999995889 0.9999996094 0.9999996289 0.9999996475 0.9999996652 0.9999996821 0.9999996981

ro ed E
n
5.0 0.9999997133 0.9999997278 0.9999997416 0.9999997548 0.9999997672 0.9999997791 0.9999997904 0.9999998011 0.9999998113 0.9999998210

t p own nse
yri t is Lic 2
d i oc nt r: 4 lice

5.1 0.9999998302 0.9999998389 0.9999998472 0.9999998551 0.9999998626

e 0.9999998698 0.9999998765 0.9999998830 0.9999998891 0.9999998949
an is d ume be t is
5

5.2 0.9999999004 0.9999999056 0.9999999105 0.9999999152 0.9999999197 0.9999999240 0.9999999280 0.9999999318 0.9999999354 0.9999999388
tec
Th Doc Num men

5.3 0.9999999421 0.9999999452 0.9999999481 0.9999999509 0.9999999535 0.9999999560 0.9999999584 0.9999999606 0.9999999628 0.9999999648
s c um S

5.4 0.9999999667 0.9999999685 0.9999999702 0.9999999718 0.9999999734 0.9999999748 0.9999999762 0.9999999775 0.9999999787 0.9999999799
E- der ocu

gh

5.5 0.9999999810 0.9999999821 0.9999999831 0.9999999840 0.9999999849 0.9999999857 0.9999999865 0.9999999873 0.9999999880 0.9999999886
Or is d

5.6 0.9999999893 0.9999999899 0.9999999905 0.9999999910 0.9999999915 0.9999999920 0.9999999924 0.9999999929 0.9999999933 0.9999999936
Th

5.7 0.9999999940 0.9999999944 0.9999999947 0.9999999950 0.9999999953 0.9999999955 0.9999999958 0.9999999960 0.9999999963 0.9999999965
5.8 0.9999999967 0.9999999969 0.9999999971 0.9999999972 0.9999999974 0.9999999975 0.9999999977 0.9999999978 0.9999999979 0.9999999981
5.9 0.9999999982 0.9999999983 0.9999999984 0.9999999985 0.9999999986 0.9999999987 0.9999999987 0.9999999988 0.9999999989 0.9999999990
6.0 0.9999999990

The tabled values are 1 - PZ = the proportion of process output beyond a particular value
of interest (such as a specification limit) that is z standard deviation units away from the
process average (for a process that is in statistical control and is normally distributed). For
example, if z = -2.17, PZ = 1 - 0.98499660 = 0.0150 or 1.5%. In any actual situation, this
proportion is only approximate.

216
 Index

INDEX
American National Standards Committee Z-1, Median and Range Chart, 84, 85
211 MEWMA, 113, 116, 195
ASQ, 211, 213, 214 np chart, 183, 192, 206
ASQ Statistics Division, 211 p chart, 70, 183
ASTM, 181, 182, 211 Regression Chart, 118
Autocorrelation, 159, 160, 191 Residuals Chart, 118, 120
Average (See also Mean), 43, 60, 62, 63, 71, 76, Short Run Chart, 176
78, 79, 82, 83, 85, 87, 89, 93, 95, 97, 109, Stoplight Control Chart, 202
111, 116, 119, 191, 194, 195, 198, 213 u chart, 108, 183, 207
Average and Range Chart, 63, 78, 79 Zone Chart, 121
Average and Standard Deviation Chart, 82, 83 Control Limit, 30, 55, 56, 59, 61, 62, 64, 65, 69,

din p
Average Run Length, 76, 111, 191 70, 80, 83, 85, 87, 90, 93, 95, 97, 181, 182,

.
or ou
gly
Bhote, K.R, 104, 211 183, 193

cc Gr
Binomial Distribution, 192 Control Statistic, 58, 59, 193

Inc

d a ion
Bissell, B.A.F., 139, 211 Convenience Sampling, 193, 199

rke Act
Bothe, D., 145, 174, 179, 211 Correlation, 53, 193

al
on
Box, G.E.P., 115, 120, 141, 211 Correlation Matrix, 193

ma ry
ati
Boyles, R.A., 139, 179, 211 Cox, D.R., 115, 141, 211

ter ust
en e 11
rn
Brase, 211 Cui, H., 114, 212
be otiv /20
wa Ind
nte

Burr, I.W., 114, 211 CUSUM, 109, 110, 111, 112, 122, 194, 195
c chart, 183, 204 Davis, R.B., 123, 212
1
aI

3/3

Capability, 19, 20, 128, 185, 211, 213, 214 Deming, W. Edwards, 17, 19, 29, 57, 171, 174,
gn

Cause and Effect Diagram, 192 197, 212, 214

an e A n:
Ma

as m

Centerline, 32, 48, 59, 80, 83, 85, 87, 90, 93, 95, Detection, 7, 194, 197
o
o
i
d h ut
t
to:

97, 181, 182, 183, 192 Dispersion, 194

ted y t ira

Champ, C.W., 212, 213 Distribution, 192, 194, 196, 197

xp
op en ite 426 sed

Chan, 114, 179, 212 Dixon, W.J., 182, 212

h
ro ed E
n

Characteristic, 151, 192 Doty, L.A., 212

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Charbonneau, H.C., 63, 176, 212 Dovich, R.A., 139, 212

en

Cheng, S.W., 179, 212 Duncan, A.J., 63, 176, 212

an is d ume be t is
5

tec

Common Cause, 12, 192, 203 English, J.R.,, 212

Th Doc Num men

Common Cause (See also Special Cause), 13 EWMA, 109, 111, 112, 174, 194, 195, 212
s c um S

Confidence Interval, 192 Fair, D.C., 47, 109, 214

E- der ocu

Consecutive, 192 Farnum, N.R., 109, 212

gh
Or is d

Control, 7, 9, 19, 20, 25-34, 37, 38, 41, 43, 45, Fellers, G., 213
47-74, 79, 80, 83, 85, 87, 89, 90, 93, 95, 97, Freund, J.E., 211, 213
Th

99-108, 113, 117, 118, 121, 128, 157, 176, Godfrey A.B., 63, 109, 192, 201, 213
177, 181-183, 188, 191-202, 211-214 Gordon, L. W., 63, 212
Control Chart, 28, 29, 32, 37, 41, 45-55, 58, 59, Grant, E.L., 57, 63, 109, 176, 213
71, 72, 74, 79, 89, 99, 100, 107, 108, 113, Gruska, G.F., 57, 102, 213
117, 121, 176, 177, 181-183, 191-196, 199, Gunter, B., 147, 213
200, 202, 211- 214 Haphazard Sampling, 194, 199, 200
Average and Range Chart, 63, 78, 79 Heaphy, M.S., 102, 213
Average and Standard Deviation Chart, 82, 83 Herman, J.T., 147, 213
c chart, 183, 204 Homer, A., 212
CUSUM, 109, 110, 111, 112, 122, 194, 195 Index (See Process Capabiltiy), 185, 211, 213
EWMA, 109, 111, 112, 174, 194, 195, 212 Individual, 86, 87, 89, 93, 95, 97, 110, 194
Individuals and Moving Range Chart, 87, 89, Individuals and Moving Range Chart, 87, 89,
174 174
MCUSUM, 113, 116, 195 Inherent Variation, 195, 203

217
Index

Ishikawa, K, 61, 62, 63, 192, 213 135, 136, 147, 152, 153, 154, 162, 194, 198,
Jaehn, A.H., 213 201-203, 211- 214
Jenkins, G.M., 119, 120, 211 Process Average, 198
Johnson, N.L., 115, 141, 142, 213 Process Capability, 18, 19, 125, 131, 198, 211,
Juran, J., 17, 63, 109, 192, 201, 213 212, 214
Kane, V. E., 213 Variables Data Case, 198
Keats, J. B., 213 Process Control (See Statistical Process
Kourti, T., 117, 213 Control), 1, 4, 8, 9, 18, 19, 25, 29, 152, 198,
Lamberson, L.R., 213 201, 212, 213, 214
Leavenworth, RS, 57, 63, 109, 176, 213 Process Performance, 125, 131, 198
Lee, S., 212 Process Spread, 194, 198, 201
Location, 13, 194, 195 Quadratic, 199
Loss Function, 104, 148, 150, 151, 179, 195 Randomness, 199
Lowry, C.A., 113, 213 Range, 31, 43, 60, 62, 72, 79, 85, 87, 158, 160,

din p
.
MacGregor, J.F., 117, 213 164, 195, 199

or ou
gly
Martin, T.W., v, 212 Rational Subgroup, 199

cc Gr
Mason, R.I., 117, 214 Regression Chart, 118

Inc

d a ion
Massey, F.J Jr., 182, 212 Reinsel, G.C., 120, 211

rke Act
al
Mauch, P. D., 214 Residuals Chart, 118, 120
on
MCUSUM, 113, 116, 195 Reynolds, J.H, 121, 214

ma ry
ati
Mean (See also Average), 191, 195 Rigdon, S.E, 212, 213

ter ust
en e 11
rn

Median, 84, 85, 182, 195 Roberts, S.W., 121, 214

be otiv /20
wa Ind
nte

Median and Range Chart, 84, 85 Run, 76, 107, 111, 191, 200, 212, 214
1
aI

MEWMA, 113, 116, 195 Sample, 51, 52, 58, 85, 163, 168, 188, 200
3/3

Mirkhani, K., 213 Sampling

gn

Mode, 195 Convenience Sampling, 193, 199

an e A n:
Ma

as m
o

Montgomery, D.C., 47, 57, 109, 110, 111, 113, Haphazard Sampling, 194, 199, 200
o
i
d h ut
t
to:

ted y t ira

117, 118, 121, 192, 201, 213, 214 Probability Sampling, 198
Moving Range, 43, 86, 87, 89, 107, 110, 160, Random Sampling, 198, 199
xp
op en ite 426 sed

174, 195 Rational Subgroup, 199

ro ed E
n

t p wn se

Multivariate Distribution, 140, 144 Scherkenbach, W.W., 153, 214

yri t is Lic 2
d i oc nt r: 4 lice

Nonconforming Units, 196 Shape, 13, 200

en
an is d ume be t is

Nonconformity, 196 Shewhart, Walter A., 19, 29, 30, 31, 76, 109,
5

tec

Non-Normal Chart, 113 111, 113, 114, 115, 122, 123, 196, 200, 203,
Th Doc Num men

Non-Normal Distribution, 140, 142, 196 214

s c um S

Normal Distribution, 140, 142, 196 Short Run Chart, 176

E- der ocu

np chart, 183, 192, 206 Sigma (σ ) , 200

gh
Or is d

Operational Definition, 197

Ott, E.R., 63, 214 Special Cause (See also Common Cause), 12,
Th

p chart, 70, 183 13, 60, 62, 75, 171, 200, 203
Pareto Chart, 197 Specification, 67, 188, 200, 202
Performance, 9, 128, 212 Bilateral, 132
Pham, H, 114, 214 Unilateral, 137
Point Estimate, 197 Spiring, F.A., 127, 179, 212, 214
Poisson Distribution, 197 Spread (See also Variation), 13, 127, 198, 201
Prediction Interval, 197 Stability, 201
Prevention, 7, 194, 197, 213 Stable Process, 201
Probability based charts, 101, 198 Standard Deviation, 79, 83, 85, 87, 160, 201
Probability Sampling, 198 Statistic, 58, 59, 193, 201, 214
Problem Solving, 198 Statistical Control, 20, 55, 60, 193, 201
Process, 1, 4, 8, 9, 18, 19, 21, 24, 25, 26, 29, 31, Statistical Inference, 201
33, 34, 53, 67, 103, 107, 125, 127, 131, 132, Statistical Process Control, i, 4, 198, 201, 212,
213, 214

218
 Index

Statistical Tolerance Limits, 201, 202 Variation, 12, 13, 67, 83, 130, 131, 190, 192,
Stoplight Control Chart, 202 193, 195, 202, 203
Subgroup, 48, 55, 57, 58, 79, 83, 85, 130, 181, Inherent Variation, 195, 203
182, 188, 195, 199, 200, 202 Inherent Variation:, 203
Tilmon, C., 212 Total Process Variation:, 203
Tolerance (See Specification), 104, 201, 202 Wadsworth, H.M.,, 214
Total Process Variation, 131, 202, 203 Wheeler, D.J., 47, 63, 89, 107, 111, 117, 121,
Type I Error, 202 134, 145, 161, 174, 214
Type II Error, 202 Williams, F.J., 211, 213
u chart, 108, 183, 207 Wise, S.A., 47, 109, 214
Unimodal, 202 Woodall, W.H, 212, 213
Variables Data (See also Attribute Data), 44, Young, J.C., 117, 214
125, 191, 198, 202 Zone Analysis, 203
Zone Chart, 121

din p
.
or ou
gly
cc Gr
Inc

d a ion
rke Act
al
on

ma ry
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

1
aI

3/3
gn

an e A n:
Ma

as m
o
o
i
d h ut
t
to:

ted y t ira
xp
op en ite 426 sed

h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

b
en
an is d ume be t is
5

tec
Th Doc Num men

o
s c um S
E- der ocu

gh
Or is d
Th

219
 Index

Th
Or is d
E- der ocu
Th Doc Num men
an is d ume be t is
d i oc nt r: 4 lice
s c um S 5 n
op en ite 426 sed
yri t is Lic 2 to:
gh o
t p wn se
en Ma
ro ed E gn
tec b xp aI
ted y t ira nte
h
an e A n: t i o rn
d h ut ati
o 3/3 on
as m 1 al
be otiv /20 Inc
en e 11
wa Ind
This page intentionally left blank

ter ust
ma ry

220
rke Act
d a ion
cc Gr
or ou
din p
gly
.
 Feedback

S.P.C. Manual User Feedback Process

Consistent with the concept of continual improvement, this automotive industry Statistical Process
Control (SPC) manual is being subjected to a formal periodic review/revision process. In line with the
concept of customer satisfaction, this review will entail consideration of not only any applicable vehicle
manufacturer requirement changes from year to year but also of feedback from users of the manual for the
purpose of making it more value-added and effective to the automotive industry and user communities.
Accordingly, please feel free to offer, in writing, your feedback comments, both pro and con, relative to
the manual’s understandability, “user-friendliness,” etc., in the area indicated below. Please indicate
specific manual page numbers where appropriate. Forward your feedback to the address indicated below:

Your Name __________________________________________________________________

din p
.
or ou
gly
Representing ________________________________________________________________

cc Gr
Company/Division Name

Inc

d a ion
Address ________________________________________________________________

rke Act
al
________________________________________________________________
on

ma ry
________________________________________________________________
ati

ter ust
en e 11
rn

be otiv /20
wa Ind
nte

Phone ___(____)________________
1
aI

Please list your top three automotive customers and their locations.
3/3
gn

________________________________________ ___________________________________
an e A n:

Customer Location
Ma

as m

________________________________________ ___________________________________
o
o
i
d h ut
t
to:

Customer Location
ted y t ira

________________________________________ ___________________________________
xp
op en ite 426 sed

Customer Location
h
ro ed E
n

t p wn se
yri t is Lic 2
d i oc nt r: 4 lice

Feedback Comments (attach additional sheets if needed)

en

___________________________________________________________________________
an is d ume be t is
5

tec
Th Doc Num men

___________________________________________________________________________
o
s c um S

___________________________________________________________________________
E- der ocu

gh

___________________________________________________________________________
Or is d

___________________________________________________________________________
Th

Send Comments To:

Automotive Industry Action Group
Suite 200 SPC, 2nd Edition
26200 Lahser Road
Southfield, Michigan 48034

Please access www.aiag.org to submit your feedback electronically.

221