ISO 22514-1:2009
(Main)Statistical methods in process management — Capability and performance — Part 1: General principles and concepts
Statistical methods in process management — Capability and performance — Part 1: General principles and concepts
ISO 22514-1:2009 describes the fundamental principles of capability and performance of manufacturing processes. It has been prepared to provide guidance about circumstances where a capability study is requested or is necessary to determine if the output from a manufacturing process or the production equipment (a production machine) is acceptable according to appropriate criteria. Such circumstances are common in quality control when the purpose for the study is part of some kind of production acceptance. These studies may also be used when diagnosis is required concerning a production output or as part of a problem solving effort. The methods are very versatile and have been applied for many situations. ISO 22514-1:2009 is applicable to the following: organizations seeking confidence that their product characteristics requirements are fulfilled; organizations seeking confidence from their suppliers that their product specifications are and will be satisfied; those internal or external to the organization who audit it for conformity with the product requirements; those internal to the organization who deal with analysing and evaluating the existing production situation to identify areas for process improvement.
Méthodes statistiques dans la gestion de processus — Aptitude et performance — Partie 1: Principes et concepts généraux
L'ISO 22514-1:2009 décrit les principes fondamentaux de l'aptitude et de la performance des processus de fabrication. Elle a été élaborée pour fournir des recommandations concernant les circonstances dans lesquelles une étude d'aptitude est requise ou nécessaire pour déterminer si le résultat d'un processus de fabrication ou le matériel de production (une machine de fabrication) est acceptable selon des critères appropriés. Ces circonstances sont courantes dans le processus de contrôle de la qualité, lorsque l'objet de l'étude fait partie intégrante d'un certain type d'acceptation de la production. Ces études peuvent également être utilisées lorsqu'un diagnostic est requis concernant le rendement d'une production ou comme partie intégrante d'une démarche de résolution de problèmes. Les méthodes utilisées, très polyvalentes, ont été appliquées dans de nombreuses situations. L'ISO 22514-1:2009 est applicable aux organismes qui cherchent à s'assurer que les exigences relatives aux caractéristiques de leurs produits sont satisfaites; aux organismes qui cherchent à s'assurer que leurs fournisseurs satisfont et satisferont aux spécifications de leurs produits; à ceux, en interne ou à l'extérieur de l'organisme, qui auditent ce dernier en termes de conformité aux exigences relatives au produit; à ceux, à l'intérieur de l'organisme, qui analysent et évaluent la situation de production existante pour identifier les secteurs d'amélioration du processus.
Statistične metode za obvladovanje procesov - Sposobnost in delovanje - 1. del: Splošna načela in pojmi
Ta del ISO 22514 opisuje temeljna načela sposobnosti in delovanja proizvodnih procesov. Pripravljen je bil z namenom nuditi vodilo o okoliščinah, kjer se zahteva študija sposobnosti oziroma je nujna za ugotavljanje, ali je rezultat proizvodnega procesa oziroma proizvodne opreme (proizvodnega stroja) sprejemljiv glede na ustrezna merila. Takšne okoliščine so običajne pri nadzoru kakovosti, kjer je namen študije del neke vrste prevzemanja proizvodnje. Te študije se lahko uporabljajo tudi, kadar se zahteva diagnoza glede proizvodnega rezultata ali kot del prizadevanj za reševanje problemov. Te metode so zelo vsestranske in so bile uporabljene za številne situacije. Ta del ISO 22514 velja za naslednje: organizacije, ki si prizadevajo za zaupanje, da so zahteve glede značilnosti njihovih proizvodov izpolnjene; organizacije, ki si prizadevajo za zaupanje njihovih dobaviteljev, da so in bodo njihove produktne specifikacije izpolnjene; tiste interne ali zunanje sodelavce organizacije, ki jo revidirajo glede skladnosti s produktnimi zahtevami; tiste interne ali zunanje sodelavce organizacije, ki opravljajo analize in vrednotenja obstoječih proizvodnih razmer za ugotavljanje področij za izboljšavo procesov.
General Information
Relations
Buy Standard
Standards Content (Sample)
INTERNATIONAL ISO
STANDARD 22514-1
First edition
2009-10-01
Statistical methods in process
management — Capability and
performance —
Part 1:
General principles and concepts
Méthodes statistiques dans la gestion de processus — Aptitude et
performance —
Partie 1: Principes et concepts généraux
Reference number
ISO 22514-1:2009(E)
©
ISO 2009
---------------------- Page: 1 ----------------------
ISO 22514-1:2009(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
COPYRIGHT PROTECTED DOCUMENT
© ISO 2009
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2009 – All rights reserved
---------------------- Page: 2 ----------------------
ISO 22514-1:2009(E)
Contents Page
Foreword .iv
Introduction.v
1 Scope.1
2 Terms and definitions .1
2.1 Basic terms .1
2.2 Performance — Measures and indices .7
2.3 Capability — Measures and indices .10
3 Symbols, abbreviated terms and subscripts.13
3.1 Symbols and abbreviated terms .13
3.2 Subscripts .14
4 Pre-conditions for application .14
4.1 Aspects about establishing specifications.14
4.2 Distribution and sample size.15
4.3 Materials used in studies.15
4.4 Special circumstances.15
5 Collection of data .15
5.1 Traceability of data.15
5.2 Measurement uncertainty.16
5.3 Data recording .16
5.4 Outliers .16
6 Performance, capability and process analysis .16
6.1 Six different types of performance and capability.16
6.2 Basic considerations .17
6.3 Machine performance .19
6.4 Process performance and process capability.19
6.5 Position performance.20
6.6 Measurement system analysis.21
6.7 Performance and capability indices (PCIs).22
7 Results of use .22
8 Benefits of use.23
9 Limitations of use.23
Bibliography.24
iii
© ISO 2009 – All rights reserved
---------------------- Page: 3 ----------------------
ISO 22514-1:2009(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO 22514-1 was prepared by Technical Committee ISO/TC 69, Applications of statistical methods,
Subcommittee SC 4, Applications of statistical methods in process management.
ISO 22514 consists of the following parts, under the general title Statistical methods in process
management — Capability and performance:
⎯ Part 1: General principles and concepts
⎯ Part 3: Machine performance studies for measured data on discrete parts
⎯ Part 4: Process capability estimates and performance measures [Technical Report]
The following parts are planned:
⎯ Part 5: Process capability statistics for attribute characteristics
⎯ Part 6: Process capability statistics for characteristics following a multivariate normal distribution
⎯ Part 7: Capability of measurement processes
It is planned to reissue ISO 21747, Statistical methods — Process performance and capability statistics for
measured quality characteristics, as part 2 of ISO 22514 in the future.
iv
© ISO 2009 – All rights reserved
---------------------- Page: 4 ----------------------
ISO 22514-1:2009(E)
Introduction
0.1 This introduction to capability treats the subjects “capability” and “performance” in a general way. To
fully understand the concepts, it would be helpful to consult ISO 22514-3, ISO/TR 22514-4 and ISO 21747.
These documents extend this introductory explanation to more specific uses of the procedures.
A process can either be a discrete process or a continuous process. A discrete process generates a sequence
of distinguishable items and a continuous process generates a continuous product (e.g. a lane of paper).
The purpose of a process is to manufacture a product or perform a service which satisfies a set of preset
specifications. The specifications for a process under investigation are defined for one or more characteristics
of the product or service. However, in process performance or capability, only one characteristic is considered
at a time. The characteristic can be measurable, countable or a property. The process thus generates either a
discrete or a continuous stochastic process.
⎯ The discrete process can be
⎯ a process of real numbers,
⎯ a process of natural numbers, or
⎯ a process telling which event from a set of events has occurred for the individual items.
As an example, the set of events for the individual items could be {colour acceptable; colour not
acceptable}.
In general, the notation for a discrete stochastic process is {X }, where X is the outcome of element i in
i i
the process. In the case where the characteristic is a property X , it is a value given to each of the events
i
in the set of events used for characterizing the process. For a discrete process, the index i is normally the
number of the item in the generated sequence of items. However, sometimes it may be more convenient
to use the time from a fixed point as the index.
⎯ When the process is continuous, a number of possibilities exist for the index, depending on the nature of
the product. When the product is e.g. a lane of paper, the index could be the length from a starting point
or it could be the time from a fixed point.
It should be noted that normally a serial correlation exists in a stochastic process.
A stochastic process is either stationary or non-stationary. The stringent definition of a stationary stochastic
process will not be given here. However, for a stationary process, a distribution exists for X which is
i
independent of i.
To obtain a process which satisfies the specifications, the stochastic process should be a stationary process
or a well-defined non-stationary process (e.g. a periodic process).
To evaluate a process, a performance study is performed. A performance study should, in fact, start as a
theoretical study of all the elements in the process before the process is physically implemented. When the
parameters of the various stages in the process have been analysed and redefined, the process is
implemented (this may be only as a test process).
Based on sampling from the implemented process, the numerical part of the performance study of the process
is started. A number of questions concerning the process must, beyond any reasonable doubt, be answered
correctly. The most important question to be answered is whether the process is a stationary process which is
v
© ISO 2009 – All rights reserved
---------------------- Page: 5 ----------------------
ISO 22514-1:2009(E)
stable or predictable for a reasonable period. For the process, it is then important to identify the probability
distribution of the process and to obtain estimates of the distribution parameters within a reasonable small
variance. Based on this information, the next stage in the performance study is to map the properties of the
characteristics under investigation and decide whether they are acceptable. If the properties cannot be
accepted, the parameters of the process itself must be changed in order to obtain a process with acceptable
properties.
Consider a well-defined and implemented process that has been accepted using a performance study. The
next stage for the process would then be to ensure that the parameters of the process and, thus, of the
stochastic process do not change, or change in a predicted way. This is performed by defining a suitable
capability study.
Studies of performance and capability indices are used more and more to assess production equipment, a
process, or even measurement equipment relative to specification criteria. Different types of studies are used
depending on the circumstances.
0.2 The concepts of performance and capability have been subject to large shifts of opinion. The most
fundamental shift has been to philosophically separate what is called “capability conditions” in this part of
ISO 22514-1 from “performance conditions”, the primary difference being whether statistical stability has been
obtained (capability) or not (performance). This naturally leads to the two sets of indices that are to be found in
2.2 and 2.3. It has become necessary to draw a firm distinction between these sets, since it has been
observed in industry that companies have been misled about their true capability position due to inappropriate
indices being calculated and published.
vi
© ISO 2009 – All rights reserved
---------------------- Page: 6 ----------------------
INTERNATIONAL STANDARD ISO 22514-1:2009(E)
Statistical methods in process management — Capability and
performance —
Part 1:
General principles and concepts
1 Scope
This part of ISO 22514 describes the fundamental principles of capability and performance of manufacturing
processes. It has been prepared to provide guidance about circumstances where a capability study is
requested or is necessary to determine if the output from a manufacturing process or the production
equipment (a production machine) is acceptable according to appropriate criteria. Such circumstances are
common in quality control when the purpose for the study is part of some kind of production acceptance.
These studies may also be used when diagnosis is required concerning a production output or as part of a
problem solving effort. The methods are very versatile and have been applied for many situations.
This part of ISO 22514 is applicable to the following:
⎯ organizations seeking confidence that their product characteristics requirements are fulfilled;
⎯ organizations seeking confidence from their suppliers that their product specifications are and will be
satisfied;
⎯ those internal or external to the organization who audit it for conformity with the product requirements;
⎯ those internal to the organization who deal with analysing and evaluating the existing production situation
to identify areas for process improvement.
2 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
2.1 Basic terms
2.1.1
requirement
need or expectation that is stated, generally implied or obligatory
[ISO 9000:2005, definition 3.1.2]
© ISO 2009 – All rights reserved 1
---------------------- Page: 7 ----------------------
ISO 22514-1:2009(E)
2.1.2
process
set of inter-related or interacting activities which transforms inputs into outputs
NOTE 1 Inputs to a process are generally outputs from other processes.
NOTE 2 Processes in an organization are generally planned and carried out under controlled conditions to add value.
NOTE 3 Adapted from ISO 3534-2:2006, definition 2.1.1.
2.1.3
system
set of interrelated or interacting elements
[ISO 9000:2005, definition 3.1.3]
2.1.4
product
result of a process
NOTE 1 Four generic product categories are:
⎯ services (e.g. transport);
⎯ software (e.g. computer program);
⎯ hardware (e.g. engine mechanical part);
⎯ processed materials (e.g. lubricant).
Many products comprise elements belonging to different generic product categories. What the product is then called
depends on the dominant element.
NOTE 2 In mathematics, the concept of product is limited to the result of multiplication.
[ISO 3534-2:2006, definition 1.2.32]
2.1.5
characteristic
distinguishing feature (of an item)
NOTE 1 Adapted from ISO 9000:2005, definition 3.5.1
NOTE 2 Item is defined in ISO 3534-2:2006, definition 1.2.11.
2.1.6
quality
degree to which a set of inherent characteristics (2.1.5) of a product (2.1.4) fulfils requirements (2.1.1) of
customers and other interested parties
NOTE In ISO 9000:2005 (3.1.1), quality is defined without specification of who defines the requirements.
2.1.7
product characteristic
inherent characteristic (2.1.5) of a product (2.1.4)
NOTE 1 Product characteristics can be either quantitative or qualitative.
NOTE 2 The product characteristic may be multidimensional.
2 © ISO 2009 – All rights reserved
---------------------- Page: 8 ----------------------
ISO 22514-1:2009(E)
2.1.8
process characteristic
inherent characteristic (2.1.5) of a process (2.1.2)
NOTE 1 Process characteristics can be either quantitative or qualitative.
NOTE 2 The process characteristic may be multidimensional.
2.1.9
quality characteristic
inherent characteristic (2.1.5) of a product (2.1.4), process (2.1.2) or system (2.1.3) related to a
requirement (2.1.1)
NOTE 1 Quality characteristics can be either quantitative or qualitative.
NOTE 2 The quality characteristic may be multidimensional.
NOTE 3 Often, there is a strong correlation between a process characteristic and a product characteristic realized by a
process. However, the individual requirements differ. For a process characteristic, the individual requirement is part of the
quality requirement for the process; for a product characteristic realized by the process, the individual requirement is part
of the quality requirement for a product.
2.1.10
specification
document stating requirements (2.1.1)
NOTE A specification can be related to activities (e.g. procedure document, process specification and test
specification), or products (e.g. product specification, performance specification and drawing).
[ISO 9000:2005, definition 3.7.3]
2.1.11
specification limit
limiting value stated for a characteristic (2.1.5)
[ISO 3534-2:2006, definition 3.1.3]
NOTE Sometimes specification limits are called “tolerance limits”.
2.1.12
upper specification limit
U
specification limit (2.1.11) that defines the highest value a quality characteristic may have and still be
considered conforming
NOTE 1 The preferred symbol for upper specification limit is U.
NOTE 2 Adapted from ISO 3534-2:2006, definition 3.1.4.
2.1.13
lower specification limit
L
specification limit (2.1.11) that defines the lowest value a quality characteristic may have and still be
considered conforming
NOTE 1 The preferred symbol for lower specification limit is L.
NOTE 2 Adapted from ISO 3534-2:2006, definition 3.1.5.
© ISO 2009 – All rights reserved 3
---------------------- Page: 9 ----------------------
ISO 22514-1:2009(E)
2.1.14
specification interval
tolerance interval
interval between upper and lower specification limits (2.1.11)
NOTE This term is completely different from a statistical tolerance interval, which is an interval with stochastic
borders.
2.1.15
tolerance zone
space limited by one or several geometrically perfect lines or surfaces, and characterized by a linear
dimension, called a tolerance
[ISO 1101:2004, definition 3.1]
2.1.16
target value
T
preferred or reference value of a characteristic (2.1.5) stated in a specification (2.1.10)
[ISO 3534-2:2006, definition 3.1.2]
2.1.17
nominal value
reference value of a characteristic (2.1.5) stated in a specification
NOTE In ISO 3534-2:2006, nominal value and target value are synonyms, with target value as the preferred term.
There is a need to distinguish the reference value in a specification and the preferred value used in production.
2.1.18
actual value
value of a quantity in a characteristic (2.1.5)
2.1.19
variation
difference between values of a characteristic (2.1.5)
NOTE Variation is often expressed as a variance or standard deviation.
[ISO 3534-2:2006, definition 2.2.1]
2.1.20
random cause
common cause
chance cause
〈process variation〉 source of process variation (2.1.19) that is inherent in a process (2.1.2) over time
NOTE In a process subject only to random cause variation, the variation is predictable within statistically established
limits.
2.1.21
product characteristic in control
product characteristic (2.1.7) parameter of the distribution of the characteristic values, which practically do
not change or do change only in a known manner or within known limits
4 © ISO 2009 – All rights reserved
---------------------- Page: 10 ----------------------
ISO 22514-1:2009(E)
2.1.22
stable process
process in a state of statistical control
〈constant mean〉 process (2.1.2) subject only to random causes (2.1.20)
NOTE 1 A production in control is a production with processes in control.
NOTE 2 A stable process will generally behave as though the samples from the process at any time are simple random
samples from the same population.
NOTE 3 This state does not imply that the random variation is large or small, within or outside of specification, but
rather that the variation is predictable using statistical techniques.
NOTE 4 Adapted from ISO 3534-2:2006, definition 2.2.7.
2.1.23
distribution of a product characteristic
information on the probabilistic behaviour of a product characteristic (2.1.7)
NOTE 1 The distribution contains the numerical information about the product characteristic, except for the serial order
in which the items have been produced.
NOTE 2 The distribution of a product characteristic exists whether the product characteristic is being recorded or not,
and it depends on technical conditions such as input batches, tools, operators, etc.
NOTE 3 If information about the distribution of a product characteristic is desired, data must be collected. The
distribution that is observed depends on the technical conditions (see Note 2) and on the following data collection
conditions:
⎯ the measurement;
⎯ the time interval over which the sampling takes place;
⎯ the frequency of sampling.
The technical conditions (see Note 2) and the conditions of the data collection must always be specified.
NOTE 4 The distribution of the product characteristic may be represented in any of the ways distributions and data
from distributions are represented. The histogram is frequently used for data from a distribution, whereas the density
function is frequently used for a model of the distribution of the product characteristic.
NOTE 5 In this part of ISO 22514, the distribution of the product characteristic will be considered under different but
well-defined conditions, such as performance and capability, where performance is the least restrictive.
2.1.24
class of distributions
particular family of distributions (2.1.23), each member of which has the same common attributes by which
the family is fully specified
EXAMPLE 1 The class of normal distributions where the unknown parameters are the mean and the standard
deviation. Often the class of normal distributions is referred to simply as the normal distribution.
EXAMPLE 2 Three parameter, multi-shaped, Weibull distribution with parameters, location, shape and scale.
EXAMPLE 3 The unimodal continuous distributions.
NOTE 1 The class of distributions can often be fully specified through the values of appropriate parameters.
NOTE 2 Adapted from ISO 3534-2:2006, definition 2.5.2.
© ISO 2009 – All rights reserved 5
---------------------- Page: 11 ----------------------
ISO 22514-1:2009(E)
2.1.25
distribution model of the product characteristic
specified distribution (2.1.23) or class of distributions (2.1.24)
EXAMPLE 1 A model for the distribution of a product characteristic, such as the diameter of a bolt, might be the
normal distribution with mean 15 mm and standard deviation 0,05 mm. Here the model is a fully specified distribution.
EXAMPLE 2 A model for the same situation as in Example 1 could be the class of normal distributions without
attempting to specify a particular distribution. Here the model is the class of normal distributions.
[ISO 3534-2:2006, definition 2.5.3]
2.1.26
reference limits of the product characteristic
X , X
0,135 % 99,865 %
quantile of the distribution of the product characteristic (2.1.23)
EXAMPLE If the distribution of the product characteristic is normal with mean µ and standard deviation σ, the limits
are µ ± 3σ if traditional 0,135 % and 99,865 % quantiles are used.
NOTE 1 The conditions of the distribution of the product characteristic must be specified, see Notes 2 and 3 of 2.1.23
distribution of the product characteristic.
NOTE 2 Traditionally, the 0,135 % and 99,865 % quantiles have been used.
2.1.27
reference interval of a product characteristic
interval bounded by the 99,865 % distribution quantile, X , and the 0,135 % distribution quantile,
99,865 %
X
0,135 %
EXAMPLE 1 In a normal distribution with mean µ and standard deviation σ, the reference interval corresponding to the
traditional 0,135 % and 99,865 % quantiles has limits µ ± 3σ, and has length 6σ.
EXAMPLE 2 For a non-normal distribution, the reference interval may be estimated by means of appropriate
probability papers (e.g. log-normal) or from the sample kurtosis and sample skewness using the methods described in
ISO/TR 22514-4.
NOTE 1 The interval can be expressed by X , X , quantiles, and the length of the interval is
0,135 % 99,865 %
X − X
99,865 % 0,135 %.
NOTE 2 This term is used only as an arbitrary, but standardized, basis for defining the process performance index,
(see 2.2.3, Notes 1, 2 and 3), and process capability index (see 2.3.6, Notes 1, 2 and 3). It is sometimes incorrectly
referred to as a “natural” interval.
NOTE 3 For a normal distribution, the length of the reference interval may be expressed in terms of six standard
deviations, 6σ, or 6S, when σ is estimated from a sample.
NOTE 4 For a non-normal distribution, the length of the reference interval may be estimated by means of appropriate
software or probability plot (e.g. log-normal) or from the sample kurtosis and sample skewness using the methods
described in ISO/TR 22514-4.
NOTE 5 A quantile or fractile indicates a division of a distribution into equal units or fractions, e.g. percentiles.
NOTE 6 Adapted from ISO 3534-2:2006, definition 2.5.7.
6 © ISO 2009 – All rights reserved
---------------------- Page: 12 ----------------------
ISO 22514-1:2009(E)
2.1.28
upper fraction nonconforming of the product characteristic
p
U
fraction of the distribution of the product characteristic (2.1.23) that exceeds the upper specification limit,
U (2.1.12)
EXAMPLE In a normal distribution with mean µ and standard deviation σ,
UU−−µµ
p =−1(Φ )=Φ( )
U
σσ
where Φ is the distribution function of the standard normal distribution.
NOTE Adapted from ISO 3534-2:2006, definition 2.5.4.
2.1.29
lower fraction nonconforming of the product characteristic
p
L
fraction of the distribution of the product characteristic (2.1.23) that is less than the lower specification
limit L (2.1.13)
EXAMPLE In a normal distribution with mean µ and standard deviation σ,
L − µ
p =Φ()
L
σ
where Φ is the distribution function of the standard normal distribution.
NOTE Adapted from ISO 3534-2:2006, definition 2.5.5.
2.1.30
fraction nonconforming of the product characteristic
p
t
sum of upper fraction nonconforming of the product characteristic (2.1.28) and lower fraction
nonconforming of the product characteristic (2.1.29)
p=+pp
t L U
EXAMPLE In a normal distribution with mean µ and standard deviation σ,
µµ−−UL
p =Φ()+Φ( )
t
σσ
where Φ is the distribution function of the standard normal distribution.
NOTE Adapted from ISO 3534-2:2006, definition 2.5.6.
2.2 Performance — Measures and indices
2.2.1
performance conditions
precisely defined external conditions under which the process is evaluated, and where statistical stability has
not been obtained
NOTE 1 Examples of external conditions include:
⎯ technical conditions (input batches, operators, tools, etc.);
© ISO 2009 – All rights reserved 7
---------------------- Page: 13 ----------------------
ISO 22514-1:2009(E)
⎯ the measurement process (resolution, trueness, repeatability, reproducibility, etc.);
⎯ data collection (duration, frequency).
NOTE 2 Performance conditions are the least restrictive allowed.
NOTE 3 It is irrelevant that the process is in state of statistical control in the period considered.
NOTE 4 See the Introduction, 0.2.
2.2.2
performance measure
statistical mea
...
SLOVENSKI STANDARD
SIST ISO 22514-1:2010
01-julij-2010
6WDWLVWLþQHPHWRGH]DREYODGRYDQMHSURFHVRY6SRVREQRVWLQGHORYDQMHGHO
6SORãQDQDþHODLQSRMPL
Statistical methods in process management - Capability and performance - Part 1:
General principles and concepts
Méthodes statistiques dans la gestion de processus - Aptitude et performance - Partie 1:
Principes et concepts généraux
Ta slovenski standard je istoveten z: ISO 22514-1:2009
ICS:
03.120.30 8SRUDEDVWDWLVWLþQLKPHWRG Application of statistical
methods
SIST ISO 22514-1:2010 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
SIST ISO 22514-1:2010
---------------------- Page: 2 ----------------------
SIST ISO 22514-1:2010
INTERNATIONAL ISO
STANDARD 22514-1
First edition
2009-10-01
Statistical methods in process
management — Capability and
performance —
Part 1:
General principles and concepts
Méthodes statistiques dans la gestion de processus — Aptitude et
performance —
Partie 1: Principes et concepts généraux
Reference number
ISO 22514-1:2009(E)
©
ISO 2009
---------------------- Page: 3 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
COPYRIGHT PROTECTED DOCUMENT
© ISO 2009
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2009 – All rights reserved
---------------------- Page: 4 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
Contents Page
Foreword .iv
Introduction.v
1 Scope.1
2 Terms and definitions .1
2.1 Basic terms .1
2.2 Performance — Measures and indices .7
2.3 Capability — Measures and indices .10
3 Symbols, abbreviated terms and subscripts.13
3.1 Symbols and abbreviated terms .13
3.2 Subscripts .14
4 Pre-conditions for application .14
4.1 Aspects about establishing specifications.14
4.2 Distribution and sample size.15
4.3 Materials used in studies.15
4.4 Special circumstances.15
5 Collection of data .15
5.1 Traceability of data.15
5.2 Measurement uncertainty.16
5.3 Data recording .16
5.4 Outliers .16
6 Performance, capability and process analysis .16
6.1 Six different types of performance and capability.16
6.2 Basic considerations .17
6.3 Machine performance .19
6.4 Process performance and process capability.19
6.5 Position performance.20
6.6 Measurement system analysis.21
6.7 Performance and capability indices (PCIs).22
7 Results of use .22
8 Benefits of use.23
9 Limitations of use.23
Bibliography.24
iii
© ISO 2009 – All rights reserved
---------------------- Page: 5 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO 22514-1 was prepared by Technical Committee ISO/TC 69, Applications of statistical methods,
Subcommittee SC 4, Applications of statistical methods in process management.
ISO 22514 consists of the following parts, under the general title Statistical methods in process
management — Capability and performance:
⎯ Part 1: General principles and concepts
⎯ Part 3: Machine performance studies for measured data on discrete parts
⎯ Part 4: Process capability estimates and performance measures [Technical Report]
The following parts are planned:
⎯ Part 5: Process capability statistics for attribute characteristics
⎯ Part 6: Process capability statistics for characteristics following a multivariate normal distribution
⎯ Part 7: Capability of measurement processes
It is planned to reissue ISO 21747, Statistical methods — Process performance and capability statistics for
measured quality characteristics, as part 2 of ISO 22514 in the future.
iv
© ISO 2009 – All rights reserved
---------------------- Page: 6 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
Introduction
0.1 This introduction to capability treats the subjects “capability” and “performance” in a general way. To
fully understand the concepts, it would be helpful to consult ISO 22514-3, ISO/TR 22514-4 and ISO 21747.
These documents extend this introductory explanation to more specific uses of the procedures.
A process can either be a discrete process or a continuous process. A discrete process generates a sequence
of distinguishable items and a continuous process generates a continuous product (e.g. a lane of paper).
The purpose of a process is to manufacture a product or perform a service which satisfies a set of preset
specifications. The specifications for a process under investigation are defined for one or more characteristics
of the product or service. However, in process performance or capability, only one characteristic is considered
at a time. The characteristic can be measurable, countable or a property. The process thus generates either a
discrete or a continuous stochastic process.
⎯ The discrete process can be
⎯ a process of real numbers,
⎯ a process of natural numbers, or
⎯ a process telling which event from a set of events has occurred for the individual items.
As an example, the set of events for the individual items could be {colour acceptable; colour not
acceptable}.
In general, the notation for a discrete stochastic process is {X }, where X is the outcome of element i in
i i
the process. In the case where the characteristic is a property X , it is a value given to each of the events
i
in the set of events used for characterizing the process. For a discrete process, the index i is normally the
number of the item in the generated sequence of items. However, sometimes it may be more convenient
to use the time from a fixed point as the index.
⎯ When the process is continuous, a number of possibilities exist for the index, depending on the nature of
the product. When the product is e.g. a lane of paper, the index could be the length from a starting point
or it could be the time from a fixed point.
It should be noted that normally a serial correlation exists in a stochastic process.
A stochastic process is either stationary or non-stationary. The stringent definition of a stationary stochastic
process will not be given here. However, for a stationary process, a distribution exists for X which is
i
independent of i.
To obtain a process which satisfies the specifications, the stochastic process should be a stationary process
or a well-defined non-stationary process (e.g. a periodic process).
To evaluate a process, a performance study is performed. A performance study should, in fact, start as a
theoretical study of all the elements in the process before the process is physically implemented. When the
parameters of the various stages in the process have been analysed and redefined, the process is
implemented (this may be only as a test process).
Based on sampling from the implemented process, the numerical part of the performance study of the process
is started. A number of questions concerning the process must, beyond any reasonable doubt, be answered
correctly. The most important question to be answered is whether the process is a stationary process which is
v
© ISO 2009 – All rights reserved
---------------------- Page: 7 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
stable or predictable for a reasonable period. For the process, it is then important to identify the probability
distribution of the process and to obtain estimates of the distribution parameters within a reasonable small
variance. Based on this information, the next stage in the performance study is to map the properties of the
characteristics under investigation and decide whether they are acceptable. If the properties cannot be
accepted, the parameters of the process itself must be changed in order to obtain a process with acceptable
properties.
Consider a well-defined and implemented process that has been accepted using a performance study. The
next stage for the process would then be to ensure that the parameters of the process and, thus, of the
stochastic process do not change, or change in a predicted way. This is performed by defining a suitable
capability study.
Studies of performance and capability indices are used more and more to assess production equipment, a
process, or even measurement equipment relative to specification criteria. Different types of studies are used
depending on the circumstances.
0.2 The concepts of performance and capability have been subject to large shifts of opinion. The most
fundamental shift has been to philosophically separate what is called “capability conditions” in this part of
ISO 22514-1 from “performance conditions”, the primary difference being whether statistical stability has been
obtained (capability) or not (performance). This naturally leads to the two sets of indices that are to be found in
2.2 and 2.3. It has become necessary to draw a firm distinction between these sets, since it has been
observed in industry that companies have been misled about their true capability position due to inappropriate
indices being calculated and published.
vi
© ISO 2009 – All rights reserved
---------------------- Page: 8 ----------------------
SIST ISO 22514-1:2010
INTERNATIONAL STANDARD ISO 22514-1:2009(E)
Statistical methods in process management — Capability and
performance —
Part 1:
General principles and concepts
1 Scope
This part of ISO 22514 describes the fundamental principles of capability and performance of manufacturing
processes. It has been prepared to provide guidance about circumstances where a capability study is
requested or is necessary to determine if the output from a manufacturing process or the production
equipment (a production machine) is acceptable according to appropriate criteria. Such circumstances are
common in quality control when the purpose for the study is part of some kind of production acceptance.
These studies may also be used when diagnosis is required concerning a production output or as part of a
problem solving effort. The methods are very versatile and have been applied for many situations.
This part of ISO 22514 is applicable to the following:
⎯ organizations seeking confidence that their product characteristics requirements are fulfilled;
⎯ organizations seeking confidence from their suppliers that their product specifications are and will be
satisfied;
⎯ those internal or external to the organization who audit it for conformity with the product requirements;
⎯ those internal to the organization who deal with analysing and evaluating the existing production situation
to identify areas for process improvement.
2 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
2.1 Basic terms
2.1.1
requirement
need or expectation that is stated, generally implied or obligatory
[ISO 9000:2005, definition 3.1.2]
© ISO 2009 – All rights reserved 1
---------------------- Page: 9 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
2.1.2
process
set of inter-related or interacting activities which transforms inputs into outputs
NOTE 1 Inputs to a process are generally outputs from other processes.
NOTE 2 Processes in an organization are generally planned and carried out under controlled conditions to add value.
NOTE 3 Adapted from ISO 3534-2:2006, definition 2.1.1.
2.1.3
system
set of interrelated or interacting elements
[ISO 9000:2005, definition 3.1.3]
2.1.4
product
result of a process
NOTE 1 Four generic product categories are:
⎯ services (e.g. transport);
⎯ software (e.g. computer program);
⎯ hardware (e.g. engine mechanical part);
⎯ processed materials (e.g. lubricant).
Many products comprise elements belonging to different generic product categories. What the product is then called
depends on the dominant element.
NOTE 2 In mathematics, the concept of product is limited to the result of multiplication.
[ISO 3534-2:2006, definition 1.2.32]
2.1.5
characteristic
distinguishing feature (of an item)
NOTE 1 Adapted from ISO 9000:2005, definition 3.5.1
NOTE 2 Item is defined in ISO 3534-2:2006, definition 1.2.11.
2.1.6
quality
degree to which a set of inherent characteristics (2.1.5) of a product (2.1.4) fulfils requirements (2.1.1) of
customers and other interested parties
NOTE In ISO 9000:2005 (3.1.1), quality is defined without specification of who defines the requirements.
2.1.7
product characteristic
inherent characteristic (2.1.5) of a product (2.1.4)
NOTE 1 Product characteristics can be either quantitative or qualitative.
NOTE 2 The product characteristic may be multidimensional.
2 © ISO 2009 – All rights reserved
---------------------- Page: 10 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
2.1.8
process characteristic
inherent characteristic (2.1.5) of a process (2.1.2)
NOTE 1 Process characteristics can be either quantitative or qualitative.
NOTE 2 The process characteristic may be multidimensional.
2.1.9
quality characteristic
inherent characteristic (2.1.5) of a product (2.1.4), process (2.1.2) or system (2.1.3) related to a
requirement (2.1.1)
NOTE 1 Quality characteristics can be either quantitative or qualitative.
NOTE 2 The quality characteristic may be multidimensional.
NOTE 3 Often, there is a strong correlation between a process characteristic and a product characteristic realized by a
process. However, the individual requirements differ. For a process characteristic, the individual requirement is part of the
quality requirement for the process; for a product characteristic realized by the process, the individual requirement is part
of the quality requirement for a product.
2.1.10
specification
document stating requirements (2.1.1)
NOTE A specification can be related to activities (e.g. procedure document, process specification and test
specification), or products (e.g. product specification, performance specification and drawing).
[ISO 9000:2005, definition 3.7.3]
2.1.11
specification limit
limiting value stated for a characteristic (2.1.5)
[ISO 3534-2:2006, definition 3.1.3]
NOTE Sometimes specification limits are called “tolerance limits”.
2.1.12
upper specification limit
U
specification limit (2.1.11) that defines the highest value a quality characteristic may have and still be
considered conforming
NOTE 1 The preferred symbol for upper specification limit is U.
NOTE 2 Adapted from ISO 3534-2:2006, definition 3.1.4.
2.1.13
lower specification limit
L
specification limit (2.1.11) that defines the lowest value a quality characteristic may have and still be
considered conforming
NOTE 1 The preferred symbol for lower specification limit is L.
NOTE 2 Adapted from ISO 3534-2:2006, definition 3.1.5.
© ISO 2009 – All rights reserved 3
---------------------- Page: 11 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
2.1.14
specification interval
tolerance interval
interval between upper and lower specification limits (2.1.11)
NOTE This term is completely different from a statistical tolerance interval, which is an interval with stochastic
borders.
2.1.15
tolerance zone
space limited by one or several geometrically perfect lines or surfaces, and characterized by a linear
dimension, called a tolerance
[ISO 1101:2004, definition 3.1]
2.1.16
target value
T
preferred or reference value of a characteristic (2.1.5) stated in a specification (2.1.10)
[ISO 3534-2:2006, definition 3.1.2]
2.1.17
nominal value
reference value of a characteristic (2.1.5) stated in a specification
NOTE In ISO 3534-2:2006, nominal value and target value are synonyms, with target value as the preferred term.
There is a need to distinguish the reference value in a specification and the preferred value used in production.
2.1.18
actual value
value of a quantity in a characteristic (2.1.5)
2.1.19
variation
difference between values of a characteristic (2.1.5)
NOTE Variation is often expressed as a variance or standard deviation.
[ISO 3534-2:2006, definition 2.2.1]
2.1.20
random cause
common cause
chance cause
〈process variation〉 source of process variation (2.1.19) that is inherent in a process (2.1.2) over time
NOTE In a process subject only to random cause variation, the variation is predictable within statistically established
limits.
2.1.21
product characteristic in control
product characteristic (2.1.7) parameter of the distribution of the characteristic values, which practically do
not change or do change only in a known manner or within known limits
4 © ISO 2009 – All rights reserved
---------------------- Page: 12 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
2.1.22
stable process
process in a state of statistical control
〈constant mean〉 process (2.1.2) subject only to random causes (2.1.20)
NOTE 1 A production in control is a production with processes in control.
NOTE 2 A stable process will generally behave as though the samples from the process at any time are simple random
samples from the same population.
NOTE 3 This state does not imply that the random variation is large or small, within or outside of specification, but
rather that the variation is predictable using statistical techniques.
NOTE 4 Adapted from ISO 3534-2:2006, definition 2.2.7.
2.1.23
distribution of a product characteristic
information on the probabilistic behaviour of a product characteristic (2.1.7)
NOTE 1 The distribution contains the numerical information about the product characteristic, except for the serial order
in which the items have been produced.
NOTE 2 The distribution of a product characteristic exists whether the product characteristic is being recorded or not,
and it depends on technical conditions such as input batches, tools, operators, etc.
NOTE 3 If information about the distribution of a product characteristic is desired, data must be collected. The
distribution that is observed depends on the technical conditions (see Note 2) and on the following data collection
conditions:
⎯ the measurement;
⎯ the time interval over which the sampling takes place;
⎯ the frequency of sampling.
The technical conditions (see Note 2) and the conditions of the data collection must always be specified.
NOTE 4 The distribution of the product characteristic may be represented in any of the ways distributions and data
from distributions are represented. The histogram is frequently used for data from a distribution, whereas the density
function is frequently used for a model of the distribution of the product characteristic.
NOTE 5 In this part of ISO 22514, the distribution of the product characteristic will be considered under different but
well-defined conditions, such as performance and capability, where performance is the least restrictive.
2.1.24
class of distributions
particular family of distributions (2.1.23), each member of which has the same common attributes by which
the family is fully specified
EXAMPLE 1 The class of normal distributions where the unknown parameters are the mean and the standard
deviation. Often the class of normal distributions is referred to simply as the normal distribution.
EXAMPLE 2 Three parameter, multi-shaped, Weibull distribution with parameters, location, shape and scale.
EXAMPLE 3 The unimodal continuous distributions.
NOTE 1 The class of distributions can often be fully specified through the values of appropriate parameters.
NOTE 2 Adapted from ISO 3534-2:2006, definition 2.5.2.
© ISO 2009 – All rights reserved 5
---------------------- Page: 13 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
2.1.25
distribution model of the product characteristic
specified distribution (2.1.23) or class of distributions (2.1.24)
EXAMPLE 1 A model for the distribution of a product characteristic, such as the diameter of a bolt, might be the
normal distribution with mean 15 mm and standard deviation 0,05 mm. Here the model is a fully specified distribution.
EXAMPLE 2 A model for the same situation as in Example 1 could be the class of normal distributions without
attempting to specify a particular distribution. Here the model is the class of normal distributions.
[ISO 3534-2:2006, definition 2.5.3]
2.1.26
reference limits of the product characteristic
X , X
0,135 % 99,865 %
quantile of the distribution of the product characteristic (2.1.23)
EXAMPLE If the distribution of the product characteristic is normal with mean µ and standard deviation σ, the limits
are µ ± 3σ if traditional 0,135 % and 99,865 % quantiles are used.
NOTE 1 The conditions of the distribution of the product characteristic must be specified, see Notes 2 and 3 of 2.1.23
distribution of the product characteristic.
NOTE 2 Traditionally, the 0,135 % and 99,865 % quantiles have been used.
2.1.27
reference interval of a product characteristic
interval bounded by the 99,865 % distribution quantile, X , and the 0,135 % distribution quantile,
99,865 %
X
0,135 %
EXAMPLE 1 In a normal distribution with mean µ and standard deviation σ, the reference interval corresponding to the
traditional 0,135 % and 99,865 % quantiles has limits µ ± 3σ, and has length 6σ.
EXAMPLE 2 For a non-normal distribution, the reference interval may be estimated by means of appropriate
probability papers (e.g. log-normal) or from the sample kurtosis and sample skewness using the methods described in
ISO/TR 22514-4.
NOTE 1 The interval can be expressed by X , X , quantiles, and the length of the interval is
0,135 % 99,865 %
X − X
99,865 % 0,135 %.
NOTE 2 This term is used only as an arbitrary, but standardized, basis for defining the process performance index,
(see 2.2.3, Notes 1, 2 and 3), and process capability index (see 2.3.6, Notes 1, 2 and 3). It is sometimes incorrectly
referred to as a “natural” interval.
NOTE 3 For a normal distribution, the length of the reference interval may be expressed in terms of six standard
deviations, 6σ, or 6S, when σ is estimated from a sample.
NOTE 4 For a non-normal distribution, the length of the reference interval may be estimated by means of appropriate
software or probability plot (e.g. log-normal) or from the sample kurtosis and sample skewness using the methods
described in ISO/TR 22514-4.
NOTE 5 A quantile or fractile indicates a division of a distribution into equal units or fractions, e.g. percentiles.
NOTE 6 Adapted from ISO 3534-2:2006, definition 2.5.7.
6 © ISO 2009 – All rights reserved
---------------------- Page: 14 ----------------------
SIST ISO 22514-1:2010
ISO 22514-1:2009(E)
2.1.28
upper fraction nonconforming of the product characteristic
p
U
fraction of the distribution of the product characteristic (2.1.23) that exceeds the upper specification limit,
U (2.1.12)
EXAMPLE In a normal distribution with mean µ and standard deviation σ,
UU−−µµ
p =−1(Φ )=Φ( )
U
σσ
where Φ is the distribution function of the standard normal distribution.
NOTE Adapted from ISO 3534-2:2006, definition 2.5.4.
2.1.29
lower fraction nonconforming of the product characteristic
p
L
fraction of the distribution of the product characteristic (2.1.23) that is less than the lower specification
limit L (2.1.13)
EXAMPLE In a normal distribution with mean µ and standard deviation σ,
L − µ
p =Φ()
L
σ
where Φ is the distribution function of the standard normal distribution.
NOTE Adapted from ISO 3534-2:2006, definition 2.5.5.
2.1.30
fraction nonconforming of the product characteristic
p
t
sum of upper fraction nonconforming of the product characteristic (2.1.28) and low
...
NORME ISO
INTERNATIONALE 22514-1
Première édition
2009-10-01
Méthodes statistiques dans la gestion de
processus — Aptitude et performance —
Partie 1:
Principes et concepts généraux
Statistical methods in process management — Capability and
performance —
Part 1: General principles and concepts
Numéro de référence
ISO 22514-1:2009(F)
©
ISO 2009
---------------------- Page: 1 ----------------------
ISO 22514-1:2009(F)
PDF – Exonération de responsabilité
Le présent fichier PDF peut contenir des polices de caractères intégrées. Conformément aux conditions de licence d'Adobe, ce fichier
peut être imprimé ou visualisé, mais ne doit pas être modifié à moins que l'ordinateur employé à cet effet ne bénéficie d'une licence
autorisant l'utilisation de ces polices et que celles-ci y soient installées. Lors du téléchargement de ce fichier, les parties concernées
acceptent de fait la responsabilité de ne pas enfreindre les conditions de licence d'Adobe. Le Secrétariat central de l'ISO décline toute
responsabilité en la matière.
Adobe est une marque déposée d'Adobe Systems Incorporated.
Les détails relatifs aux produits logiciels utilisés pour la création du présent fichier PDF sont disponibles dans la rubrique General Info
du fichier; les paramètres de création PDF ont été optimisés pour l'impression. Toutes les mesures ont été prises pour garantir
l'exploitation de ce fichier par les comités membres de l'ISO. Dans le cas peu probable où surviendrait un problème d'utilisation,
veuillez en informer le Secrétariat central à l'adresse donnée ci-dessous.
DOCUMENT PROTÉGÉ PAR COPYRIGHT
© ISO 2009
Droits de reproduction réservés. Sauf prescription différente, aucune partie de cette publication ne peut être reproduite ni utilisée sous
quelque forme que ce soit et par aucun procédé, électronique ou mécanique, y compris la photocopie et les microfilms, sans l'accord écrit
de l'ISO à l'adresse ci-après ou du comité membre de l'ISO dans le pays du demandeur.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Publié en Suisse
ii © ISO 2009 – Tous droits réservés
---------------------- Page: 2 ----------------------
ISO 22514-1:2009(F)
Sommaire Page
Avant-propos .iv
Introduction.v
1 Domaine d'application .1
2 Termes et définitions .1
2.1 Termes fondamentaux .1
2.2 Performance — Mesures et indices.7
2.3 Aptitude — Mesures et indices .10
3 Symboles, termes abrégés et indices .13
3.1 Symboles et termes abrégés.13
3.2 Indices .14
4 Conditions préalables à l'application.14
4.1 Aspects relatifs à la détermination des spécifications .14
4.2 Distribution et effectif d'échantillon .15
4.3 Équipements utilisés dans les études .15
4.4 Circonstances particulières .15
5 Collecte des données.15
5.1 Traçabilité des données.15
5.2 Incertitude de mesure .16
5.3 Enregistrement des données .16
5.4 Valeurs aberrantes .16
6 Analyse de la performance, de l'aptitude et du processus.16
6.1 Six types différents de performance et d'aptitude.16
6.2 Éléments de base pris en considération.17
6.3 Performance de la machine.19
6.4 Performance du processus et aptitude du processus .19
6.5 Performance de position .20
6.6 Analyse du système de mesure.21
6.7 Indices de performance et d'aptitude (PCI) .22
7 Résultats de l'utilisation des indices.22
8 Avantages de l'utilisation des indices.23
9 Limites d'utilisation.23
Bibliographie.24
© ISO 2009 – Tous droits réservés iii
---------------------- Page: 3 ----------------------
ISO 22514-1:2009(F)
Avant-propos
L'ISO (Organisation internationale de normalisation) est une fédération mondiale d'organismes nationaux de
normalisation (comités membres de l'ISO). L'élaboration des Normes internationales est en général confiée
aux comités techniques de l'ISO. Chaque comité membre intéressé par une étude a le droit de faire partie du
comité technique créé à cet effet. Les organisations internationales, gouvernementales et non
gouvernementales, en liaison avec l'ISO participent également aux travaux. L'ISO collabore étroitement avec
la Commission électrotechnique internationale (CEI) en ce qui concerne la normalisation électrotechnique.
Les Normes internationales sont rédigées conformément aux règles données dans les Directives ISO/CEI,
Partie 2.
La tâche principale des comités techniques est d'élaborer les Normes internationales. Les projets de Normes
internationales adoptés par les comités techniques sont soumis aux comités membres pour vote. Leur
publication comme Normes internationales requiert l'approbation de 75 % au moins des comités membres
votants.
L'attention est appelée sur le fait que certains des éléments du présent document peuvent faire l'objet de
droits de propriété intellectuelle ou de droits analogues. L'ISO ne saurait être tenue pour responsable de ne
pas avoir identifié de tels droits de propriété et averti de leur existence.
L'ISO 22514-1 a été élaborée par le comité technique ISO/TC 69, Application des méthodes statistiques,
sous-comité SC 4, Application de méthodes statistiques au management de processus.
L'ISO 22514 comprend les parties suivantes, présentées sous le titre général Méthodes statistiques dans la
gestion de processus — Aptitude et performance:
⎯ Partie 1: Principes et concepts généraux
⎯ Partie 3: Études de performance de machines pour des données mesurées sur des parties discrètes
⎯ Partie 4: Estimations de l'aptitude de processus et mesures de performance [Rapport technique]
Les parties suivantes sont prévues:
⎯ Partie 5: Statistiques d'aptitude d’un processus pour les caractéristiques d’attribut
⎯ Partie 6: Statistiques de capacité opérationnelle d'un processus pour les caractéristiques qui suivent une
distribution normale à plusieurs variables
⎯ Partie 7: Aptitude des processus de mesure
Il est prévu de republier ultérieurement l'ISO 21747, Méthodes statistiques — Performances de processus et
statistiques d'aptitude pour les caractéristiques de qualité mesurées, en tant que partie 2 de l'ISO 22514.
iv © ISO 2009 – Tous droits réservés
---------------------- Page: 4 ----------------------
ISO 22514-1:2009(F)
Introduction
0.1 La présente introduction au concept de l'aptitude traite de l'«aptitude» et de la «performance» de
manière générale. La consultation des ISO 22514-3, ISO/TR 22514-4 et ISO 21747 se révèlerait utile pour
appréhender pleinement ces concepts. Ces documents étendent le présent exposé introductif à des
utilisations plus spécifiques des procédures.
Un processus peut être discontinu ou continu. Un processus discontinu génère une séquence d'individus
différenciables tandis qu'un processus continu génère un produit continu (par exemple une bobine de papier).
Un processus a pour objet de fabriquer un produit ou d'exécuter un service, qui satisfait à un ensemble de
spécifications préétablies. Les spécifications relatives au processus étudié sont définies pour une ou plusieurs
caractéristiques du produit ou du service. Les performances ou l'aptitude d'un processus ne tiennent toutefois
compte que d'une seule caractéristique à la fois. Cette caractéristique peut être mesurable, dénombrable ou
être une propriété. Le processus génère ainsi un processus stochastique discontinu ou continu.
⎯ Le processus discontinu peut être
⎯ un processus de nombres réels,
⎯ un processus de nombres naturels, ou
⎯ un processus qui indique l'occurrence d'un événement donné issu d'un ensemble d'événements pour
les individus.
À titre d'exemple, l'ensemble d'événements pour les individus pourrait être du type {de couleur
acceptable, de couleur non acceptable}.
En général, la notation applicable à un processus stochastique discontinu est {X }, où X est le résultat de
i i
l'élément i dans le processus. Dans le cas où la caractéristique est une propriété X , il s'agit d'une valeur
i
attribuée à chacun des événements de l'ensemble des événements qui sert à caractériser le processus.
Pour un processus discontinu, l'indice i est normalement le numéro de l'individu dans la séquence
d'individus générée. Cependant, il peut parfois se révéler plus approprié d'utiliser comme indice le temps
par rapport à un point fixe.
⎯ Lorsque le processus est continu, il existe un grand nombre de possibilités pour l'indice, selon la nature
du produit. Lorsque le produit est, par exemple, une bobine de papier, l'indice pourrait être la longueur
effective par rapport à un point de départ ou il pourrait être le temps par rapport à un point fixe.
Il convient de noter qu'un processus stochastique comporte normalement une corrélation propre.
Un processus stochastique est stationnaire ou non stationnaire. Le présent document ne donne pas une
définition rigoureuse d'un processus stochastique stationnaire. Toutefois, un processus stationnaire comporte
une répartition de X , qui est indépendante de i.
i
Pour obtenir un processus qui satisfait aux spécifications, il convient que le processus stochastique soit
stationnaire ou qu'il soit non stationnaire et bien défini (par exemple un processus périodique).
L'évaluation d'un processus requiert une étude de la performance. Il convient en fait qu'une étude de la
performance débute comme une étude théorique de tous les éléments contenus dans le processus avant la
mise en œuvre physique dudit processus. Lorsque les paramètres des différentes phases du processus ont
été analysés et redéfinis, le processus est mis en œuvre (peut-être uniquement comme un processus d'essai).
© ISO 2009 – Tous droits réservés v
---------------------- Page: 5 ----------------------
ISO 22514-1:2009(F)
L'échantillonnage du processus mis en œuvre constitue la base d'initiation de la partie numérique de l'étude
de la performance. Il convient de répondre correctement à un certain nombre de questions concernant le
processus, et ce, au-delà de tout doute raisonnable. La question la plus importante à laquelle il doit être
répondu consiste à déterminer si le processus est un processus stationnaire stable ou prévisible pendant une
période raisonnable. Il est alors important, pour le processus, d'identifier sa loi de probabilité et d'obtenir des
estimations des paramètres de répartition avec une variance faible raisonnable. Sur la base de ces
informations, la phase suivante de l'étude de la performance consisterait à représenter les propriétés des
caractéristiques examinées et à déterminer si elles sont acceptables. Si les propriétés ne peuvent pas être
acceptées, les paramètres du processus proprement dit doivent être modifiés de manière à obtenir un
processus ayant des propriétés acceptables.
Considérons, dans un premier temps, un processus bien défini et mis en œuvre qui a été accepté au moyen
d'une étude de la performance. La phase suivante du processus consisterait alors à s'assurer que les
paramètres du processus et, ainsi, du processus stochastique, ne changent pas, ou changent de manière
prévisible. Pour ce faire, il y a lieu de définir une étude d'aptitude appropriée.
Les études portant sur les indices de performance et d'aptitude sont de plus en plus utilisées pour évaluer le
matériel de production, un processus, voire un équipement de mesure, par rapport aux critères de
spécification. Différents types d'études sont utilisés selon les circonstances.
0.2 Les concepts d'aptitude et de performance ont fait l'objet de profonds changements d'opinion. Le
changement le plus important a consisté à procéder à une distinction philosophique entre ce que la présente
partie de l'ISO 22514 désigne comme les «conditions d'aptitude» et les «conditions de performance», la
différence principale consistant dans le fait de déterminer si une stabilité statistique a été obtenue (aptitude)
ou non (performance). Cette distinction conduit naturellement aux deux ensembles d'indices spécifiés en 2.2
et 2.3. Il est devenu nécessaire d'établir une distinction nette entre ces ensembles, dans la mesure où il a été
constaté que les entreprises du secteur industriel ont été induites en erreur quant à leur potentiel d'aptitude
réel, en raison du calcul et de la publication d'indices inappropriés.
vi © ISO 2009 – Tous droits réservés
---------------------- Page: 6 ----------------------
NORME INTERNATIONALE ISO 22514-1:2009(F)
Méthodes statistiques dans la gestion de processus — Aptitude
et performance —
Partie 1:
Principes et concepts généraux
1 Domaine d'application
La présente partie de l'ISO 22514 décrit les principes fondamentaux de l'aptitude et de la performance des
processus de fabrication. Elle a été élaborée pour fournir des recommandations concernant les circonstances
dans lesquelles une étude d'aptitude est requise ou nécessaire pour déterminer si le résultat d'un processus
de fabrication ou le matériel de production (une machine de fabrication) est acceptable selon des critères
appropriés. Ces circonstances sont courantes dans le processus de contrôle de la qualité, lorsque l'objet de
l'étude fait partie intégrante d'un certain type d'acceptation de la production. Ces études peuvent également
être utilisées lorsqu'un diagnostic est requis concernant le rendement d'une production ou comme partie
intégrante d'une démarche de résolution de problèmes. Les méthodes utilisées, très polyvalentes, ont été
appliquées dans de nombreuses situations.
La présente partie de l'ISO 22514 est applicable:
⎯ aux organismes qui cherchent à s'assurer que les exigences relatives aux caractéristiques de leurs
produits sont satisfaites;
⎯ aux organismes qui cherchent à s'assurer que leurs fournisseurs satisfont et satisferont aux
spécifications de leurs produits;
⎯ à ceux, en interne ou à l'extérieur de l'organisme, qui auditent ce dernier en termes de conformité aux
exigences relatives au produit;
⎯ à ceux, à l'intérieur de l'organisme, qui analysent et évaluent la situation de production existante pour
identifier les secteurs d'amélioration du processus.
2 Termes et définitions
Pour les besoins du présent document, les termes et définitions suivants s'appliquent.
2.1 Termes fondamentaux
2.1.1
exigence
besoin ou attente formulés, habituellement implicites, ou imposés
[ISO 9000:2005, définition 3.1.2]
© ISO 2009 – Tous droits réservés 1
---------------------- Page: 7 ----------------------
ISO 22514-1:2009(F)
2.1.2
processus
ensemble d'activités corrélées ou interactives qui transforme des éléments d'entrée en éléments de sortie
NOTE 1 Les éléments d'entrée d'un processus sont généralement les éléments de sortie d'autres processus.
NOTE 2 Les processus d'un organisme sont généralement planifiés et mis en œuvre dans des conditions maîtrisées
afin d'apporter une valeur ajoutée.
NOTE 3 Adapté de l'ISO 3534-2:2006, définition 2.1.1.
2.1.3
système
ensemble d'éléments corrélés ou interactifs
[ISO 9000:2005, définition 3.1.3]
2.1.4
produit
résultat d'un processus
NOTE 1 Il existe quatre catégories génériques de produits:
⎯ les services (par exemple, le transport);
⎯ les logiciels (par exemple, programme informatique);
⎯ les [produits] matériels (par exemple, pièces mécaniques de moteur);
⎯ les produits issus de processus à caractère continu (par exemple, lubrifiant).
De nombreux produits sont constitués d'éléments appartenant à différentes catégories génériques de produits.
L'appellation du produit dépend alors de l'élément dominant.
NOTE 2 En mathématique, le concept de produit est limité au résultat de la multiplication.
[ISO 3534-2:2006, définition 1.2.32]
2.1.5
caractéristique
trait distinctif (d'un individu)
NOTE 1 Аdapté de l'ISO 9000:2005, définition 3.5.1.
NOTE 2 L'individu est défini dans l'ISO 3534-2:2006, définition 1.2.11.
2.1.6
qualité
aptitude d'un ensemble de caractéristiques (2.1.5) intrinsèques d'un produit (2.1.4) à satisfaire les
exigences (2.1.1) des clients et des autres parties intéressées
NOTE Dans l'ISO 9000:2005 (3.1.1), la définition du terme qualité ne précise pas qui définit les exigences.
2.1.7
caractéristique de produit
caractéristique (2.1.5) intrinsèque d'un produit (2.1.4)
NOTE 1 Les caractéristiques de produit peuvent être quantitatives ou qualitatives.
NOTE 2 La caractéristique de produit peut être multidimensionnelle.
2 © ISO 2009 – Tous droits réservés
---------------------- Page: 8 ----------------------
ISO 22514-1:2009(F)
2.1.8
caractéristique de processus
caractéristique (2.1.5) intrinsèque d'un processus (2.1.2)
NOTE 1 Les caractéristiques de processus peuvent être quantitatives ou qualitatives.
NOTE 2 La caractéristique de processus peut être multidimensionnelle.
2.1.9
caractéristique qualité
caractéristique (2.1.5) intrinsèque d'un produit (2.1.4), d'un processus (2.1.2) ou d'un système (2.1.3)
relative à une exigence (2.1.1)
NOTE 1 Les caractéristiques qualité peuvent être quantitatives ou qualitatives.
NOTE 2 La caractéristique qualité peut être multidimensionnelle.
NOTE 3 Souvent, il existe une corrélation forte entre une caractéristique de processus et une caractéristique de produit,
effective du fait du processus. Toutefois, les exigences individuelles sont différentes. Pour une caractéristique de
processus, l'exigence individuelle fait partie intégrante de l'exigence qualité relative au processus; pour une
caractéristique de produit effective du fait du processus, l'exigence individuelle fait partie intégrante de l'exigence qualité
relative au produit.
2.1.10
spécification
document formulant des exigences (2.1.1)
NOTE Une spécification peut être liée à des activités (par exemple document de procédure, spécification de
processus et spécification d'essai), ou à des produits (par exemple spécification de produit, spécification de performance
et plan).
[ISO 9000:2005, définition 3.7.3]
2.1.11
limite de spécification
valeur limite spécifiée pour une caractéristique (2.1.5)
[ISO 3534-2:2006, définition 3.1.3]
NOTE Parfois, les limites de spécification sont appelées «limites de tolérance».
2.1.12
limite de spécification supérieure
U
limite de spécification (2.1.11) qui définit la valeur la plus élevée pouvant être attribuée à une caractéristique
qualité et pouvant par ailleurs être considérée conforme
NOTE 1 U est le symbole préférentiel pour la limite de spécification supérieure.
NOTE 2 Adapté de l'ISO 3534-2:2006, définition 3.1.4.
2.1.13
limite de spécification inférieure
L
limite de spécification (2.1.11) qui définit la valeur la moins élevée pouvant être attribuée à une
caractéristique qualité et pouvant par ailleurs être considérée conforme
NOTE 1 L est le symbole préférentiel pour la limite de spécification inférieure.
NOTE 2 Adapté de l'ISO 3534-2:2006, définition 3.1.5.
© ISO 2009 – Tous droits réservés 3
---------------------- Page: 9 ----------------------
ISO 22514-1:2009(F)
2.1.14
intervalle de spécification
intervalle de tolérance
intervalle entre les limites de spécification (2.1.11) supérieure et inférieure
NOTE Ce terme est complètement différent d'un intervalle de tolérance statistique, qui est un intervalle comportant
des limites stochastiques.
2.1.15
zone de tolérance
espace limité par une ou plusieurs lignes ou surfaces géométriquement parfaites, et caractérisé par une
dimension linéaire, appelée tolérance
[ISO 1101:2004, définition 3.1]
2.1.16
valeur cible
T
valeur préférentielle ou de référence d'une caractéristique (2.1.5) indiquée dans une spécification (2.1.10)
[ISO 3534-2:2006, définition 3.1.2]
2.1.17
valeur nominale
valeur de référence d'une caractéristique (2.1.5) indiquée dans une spécification
NOTE Dans l'ISO 3534-2, la valeur nominale et la valeur cible sont synonymes, la valeur cible étant le terme
préférentiel. Il est nécessaire de distinguer la valeur de référence définie dans une spécification et la valeur préférentielle
utilisée dans le processus de production.
2.1.18
valeur réelle
valeur d'une grandeur dans une caractéristique (2.1.5)
2.1.19
variation
différence entre les valeurs d'une caractéristique (2.1.5)
NOTE La variation est souvent exprimée comme une variance ou un écart-type.
[ISO 3534-2:2006, définition 2.2.1]
2.1.20
cause aléatoire
cause commune
cause fortuite
〈variation de processus〉 source de variation du processus (2.1.19) intrinsèque à un processus (2.1.2) dans
le temps
NOTE Dans un processus soumis uniquement à une variation de cause aléatoire, la variation est prévisible dans les
limites statistiquement établies.
2.1.21
caractéristique de produit maîtrisée
paramètre de caractéristique de produit (2.1.7) de la distribution des valeurs de caractéristique qui ne
changent pratiquement pas ou qui changent uniquement de manière connue ou dans des limites connues
4 © ISO 2009 – Tous droits réservés
---------------------- Page: 10 ----------------------
ISO 22514-1:2009(F)
2.1.22
processus stable
processus en état de maîtrise statistique
〈sens général〉 processus (2.1.2) uniquement soumis à des causes aléatoires (2.1.20)
NOTE 1 Une production maîtrisée est une production dont les processus sont maîtrisés.
NOTE 2 Un processus stable se comportera généralement comme si les échantillons issus du processus sont, à tout
moment, de simples échantillons aléatoires issus de la même population.
NOTE 3 Cet état n'implique pas que la variation aléatoire est grande ou petite, qu'elle s'inscrit ou non dans la
spécification, mais indique que la variation est prévisible au moyen de techniques statistiques.
NOTE 4 Adapté de l'ISO 3534-2:2006, définition 2.2.7.
2.1.23
distribution d'une caractéristique de produit
information sur le comportement probabiliste d'une caractéristique de produit (2.1.7)
NOTE 1 La distribution contient l'information numérique concernant la caractéristique de produit, à l'exception de
l'ordre dans lequel les individus ont été produits.
NOTE 2 La distribution de la caractéristique de produit existe, que la caractéristique de produit soit ou non enregistrée.
Cette distribution dépend de conditions techniques telles que les lots d'entrée, les outils, les opérateurs, etc.
NOTE 3 Des données doivent être collectées si l'on souhaite obtenir des informations sur la distribution de la
caractéristique de produit. La distribution observée dépend, en plus des conditions techniques (voir Note 2), des
conditions suivantes afférentes à la collecte des données:
⎯ le mesurage;
⎯ l'intervalle de temps pendant lequel l'échantillonnage a lieu;
⎯ la fréquence d'échantillonnage.
Les conditions techniques (voir Note 2) et les conditions de collecte des données doivent toujours être spécifiées.
NOTE 4 La distribution de la caractéristique de produit peut être représentée selon l'une des voies de représentation
des distributions et des données issues de ces distributions. L'histogramme est utilisé fréquemment pour les données
issues d'une distribution, tandis que la fonction de densité est utilisée fréquemment pour un modèle de distribution de la
caractéristique de produit.
NOTE 5 Dans la présente partie de l'ISO 22514, la distribution de la caractéristique de produit sera examinée dans des
conditions différentes mais bien définies, telles que la performance et l'aptitude, où la performance est la valeur la moins
restrictive.
2.1.24
classe de distributions
famille particulière de distributions (2.1.23), dont chacun des membres a les mêmes attributs communs,
lesquels spécifient entièrement ladite famille
EXEMPLE 1 La classe des distributions normales, où la moyenne et l'écart-type représentent les paramètres inconnus.
Il est souvent fait référence à la classe des distributions normales simplement en tant que loi normale.
EXEMPLE 2 Loi de Weibull à plusieurs formes et trois paramètres, avec paramètres, valeur centrale, forme et échelle.
EXEMPLE 3 Les distributions continues unimodales.
NOTE 1 La classe de distributions peut souvent être spécifiée totalement par les valeurs de paramètres appropriés.
NOTE 2 Adapté de l'ISO 3534-2:2006, définition 2.5.2.
© ISO 2009 – Tous droits réservés 5
---------------------- Page: 11 ----------------------
ISO 22514-1:2009(F)
2.1.25
modèle de distribution de la caractéristique de produit
distribution spécifiée (2.1.23) ou classe de distributions (2.1.24)
EXEMPLE 1 Un modèle de distribution pour une caractéristique de produit telle que le diamètre d'un boulon peut être
la loi normale avec une moyenne de 15 mm et un écart-type de 0,05 mm. Il s'agit d'un modèle de distribution totalement
spécifié.
EXEMPLE 2 Un modèle applicable à la même situation que dans l'Exemple 1 peut être la classe de distributions
normales sans spécification d'une distribution particulière. Il s'agit d'un modèle à classe de distributions normales.
[ISO 3534-2:2006, définition 2.5.3]
2.1.26
limites de référence de la caractéristique de produit
X , X
0,135 % 99,865 %
quantile de la distribution de la caractéristique de produit (2.1.23)
EXEMPLE Si la distribution de la caractéristique de produit est normale avec une moyenne µ et un écart-type σ, les
limites sont µ ± 3σ, si les quantiles 0,135 % et 99,865 % traditionnels sont utilisés.
NOTE 1 Les conditions de la distribution de la caractéristique de produit doivent être spécifiées, voir distribution
d'une caractéristique de produit (2.1.23), Notes 2 et 3.
NOTE 2 Les quantiles 0,135 % et 99,865 % sont traditionnellement utilisés.
2.1.27
intervalle de référence d'une caractéristique de produit
intervalle compris entre le quantile de distribution 99,865 %, X , et le quantile de distribution 0,135 %,
99,865 %
X
0,135 %
EXEMPLE 1 Dans une distribution normale avec une moyenne µ et un écart-type σ, l'intervalle de référence
correspondant aux quantiles traditionnels 0,135 % et 99,865 % a des limites µ ± 3σ, et une longueur 6σ.
EXEMPLE 2 Pour une loi non normale, l'intervalle de référence peut être estimé au moyen de papiers à échelle de
probabilité appropriés (par exemple log-normale) ou sur la base de l'aplatissement et de l'asymétrie de l'échantillon en
utilisant les méthodes décrites dans l'ISO/TR 22514-4.
NOTE 1 L'intervalle peut être exprimé par les quantiles X , X , sa longueur étant égale à
0,135 % 99,865 %
X − X .
99,865 % 0,135 %
NOTE 2 Ce terme n'est utilisé que comme base arbitraire mais réduite pour définir l'indice de performance du
processus (voir 2.2.3, Notes 1, 2 et 3) et l'indice d'aptitude du processus (voir 2.3.6, Notes 1, 2 et 3).
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.