Statistical methods in process management - Capability and performance - Part 3: Machine performance studies for measured data on discrete parts

This document describes the steps for conducting short-term performance studies that are typically performed on machines (including devices, appliances, apparatuses) where parts produced consecutively under repeatability conditions are considered. The number of observations to be analysed vary according to the patterns the data produce, or if the runs (the rate at which items are produced) on the machine are low in quantity. The methods are not considered suitable where the sample size produced is less than 30 observations. Methods for handling the data and carrying out the calculations are described. In addition, machine performance indices and the actions required at the conclusion of a machine performance study are described.
This document is not applicable when tool wear patterns are expected to be present during the duration of the study, nor if autocorrelation between observations is present. The situation where a machine has captured the data, sometimes thousands of data points collected in a minute, is not considered suitable for the application of this document.

Méthodes statistiques dans la gestion de processus - Aptitude et performance - Partie 3: Études de performance de machines pour des données mesurées sur des parties discrètes

Statistične metode za obvladovanje procesov - Sposobnost in delovanje - 3. del: Študije strojnega delovanja za izmerjene podatke na diskretnih delih

General Information

Status
Published
Publication Date
05-Sep-2021
Technical Committee
Current Stage
6060 - National Implementation/Publication (Adopted Project)
Start Date
01-Sep-2021
Due Date
06-Nov-2021
Completion Date
06-Sep-2021

Relations

Buy Standard

Standard
ISO 22514-3:2021
English language
24 pages
sale 10% off
Preview
sale 10% off
Preview
e-Library read for
1 day
Standard
ISO 22514-3:2020 - Statistical methods in process management -- Capability and performance
English language
19 pages
sale 15% off
Preview
sale 15% off
Preview
Draft
ISO/FDIS 22514-3:Version 13-okt-2020 - Statistical methods in process management -- Capability and performance
English language
19 pages
sale 15% off
Preview
sale 15% off
Preview
Draft
ISO/FDIS 22514-3 - Statistical methods in process management -- Capability and performance
English language
17 pages
sale 15% off
Preview
sale 15% off
Preview

Standards Content (Sample)

SLOVENSKI STANDARD
SIST ISO 22514-3:2021
01-oktober-2021
Nadomešča:
SIST ISO 22514-3:2010
Statistične metode za obvladovanje procesov - Sposobnost in delovanje - 3. del:
Študije strojnega delovanja za izmerjene podatke na diskretnih delih
Statistical methods in process management - Capability and performance - Part 3:
Machine performance studies for measured data on discrete parts
Méthodes statistiques dans la gestion de processus - Aptitude et performance - Partie 3:
Études de performance de machines pour des données mesurées sur des parties
discrètes
Ta slovenski standard je istoveten z: ISO 22514-3:2020
ICS:
03.120.30 Uporaba statističnih metod Application of statistical
methods
SIST ISO 22514-3:2021 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

---------------------- Page: 1 ----------------------
SIST ISO 22514-3:2021

---------------------- Page: 2 ----------------------
SIST ISO 22514-3:2021
INTERNATIONAL ISO
STANDARD 22514-3
Second edition
2020-12
Statistical methods in process
management — Capability and
performance —
Part 3:
Machine performance studies for
measured data on discrete parts
Méthodes statistiques dans la gestion de processus — Aptitude et
performance —
Partie 3: Études de performance de machines pour des données
mesurées sur des parties discrètes
Reference number
ISO 22514-3:2020(E)
©
ISO 2020

---------------------- Page: 3 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

COPYRIGHT PROTECTED DOCUMENT
© ISO 2020
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting
on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address
below or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii © ISO 2020 – All rights reserved

---------------------- Page: 4 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols . 1
5 Pre-conditions for application . 2
5.1 General . 2
5.2 Number of parts to be used in the study . 2
5.3 Materials to be used. 3
5.4 Measurement system. 3
5.5 Running the study . 3
5.6 Special circumstances . 3
6 Data collection . 3
6.1 Traceability of data . 3
6.2 Retention of specimens . 4
6.3 Data recording . 4
7 Analysis . 4
7.1 General . 4
7.2 Run chart . 4
7.2.1 Purpose . 4
7.2.2 Review the plot . 4
7.3 Analyse the pattern of the data . 5
7.3.1 Software approach . 5
7.3.2 Check the pattern of the data . 6
7.3.3 Summarize the data . 6
7.3.4 Manual approach . 6
7.4 Produce a probability plot . 9
7.4.1 General. 9
7.4.2 Analyse the data . 9
7.5 Special cases .10
7.5.1 Data indicate a skewed distribution .10
7.5.2 Bimodal data .11
7.5.3 Truncated data .12
7.5.4 Censored data .13
7.6 Calculation of machine performance indices .13
7.6.1 General procedure .13
7.6.2 Data following a normal distribution .14
8 Reporting .14
8.1 Test report .14
8.2 Confidence intervals .15
8.2.1 General.15
8.2.2 Indices calculated with the data following a normal distribution .15
8.2.3 Indices calculated with data following a non-normal distribution .16
9 Actions following a machine performance study .16
Annex A (informative) Tables and worksheets .17
Bibliography .19
© ISO 2020 – All rights reserved iii

---------------------- Page: 5 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(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.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/ directives).
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. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/ patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to
the World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see
www .iso .org/ iso/ foreword .html.
This document was prepared by Technical Committee ISO/TC 69, Applications of statistical methods,
Subcommittee SC 4, Applications of statistical methods in product and process management.
This second edition cancels and replaces the first edition (ISO 22514-3:2008), which has been
technically revised.
The main changes compared to the previous edition are as follows:
— updated and improved figures and computer outputs.
A list of all parts in the ISO 22514 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www .iso .org/ members .html.
iv © ISO 2020 – All rights reserved

---------------------- Page: 6 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

Introduction
This document has been prepared to provide guidance in circumstances where a study is necessary
to determine if the output from a machine, for example, is acceptable according to some criteria. Such
circumstances are common in engineering when the purpose for the study is part of an acceptance
trial. These studies can also be used when diagnosis is required concerning a machine’s current level of
performance or as part of a problem-solving effort. The method is very versatile and has been applied
to many situations.
Machine performance studies of this type provide information about the behaviour of a machine under
very restricted conditions such as limiting, as far as possible, external sources of variation that are
commonplace within a process, e.g. multi-factor and multi-level situations. The data gathered in a
study might come from items made consecutively, although this may be altered according to the study
requirements. The data are assumed to have been, generally, gathered manually.
The study procedure and reporting are of interest to engineers, supervisors and management wishing
to establish whether a machine should be purchased or put in for maintenance, to assist in problem-
solving or to understand the level of variation due to the machine itself.
© ISO 2020 – All rights reserved v

---------------------- Page: 7 ----------------------
SIST ISO 22514-3:2021

---------------------- Page: 8 ----------------------
SIST ISO 22514-3:2021
INTERNATIONAL STANDARD ISO 22514-3:2020(E)
Statistical methods in process management — Capability
and performance —
Part 3:
Machine performance studies for measured data on
discrete parts
1 Scope
This document describes the steps for conducting short-term performance studies that are typically
performed on machines (including devices, appliances, apparatuses) where parts produced
consecutively under repeatability conditions are considered. The number of observations to be analysed
vary according to the patterns the data produce, or if the runs (the rate at which items are produced)
on the machine are low in quantity. The methods are not considered suitable where the sample size
produced is less than 30 observations. Methods for handling the data and carrying out the calculations
are described. In addition, machine performance indices and the actions required at the conclusion of a
machine performance study are described.
This document is not applicable when tool wear patterns are expected to be present during the duration
of the study, nor if autocorrelation between observations is present. The situation where a machine has
captured the data, sometimes thousands of data points collected in a minute, is not considered suitable
for the application of this document.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
No terms and definitions are listed in this document.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at http:// www .electropedia .org/
4 Symbols
P probability
P machine performance index
m
P lower machine performance index
mk
L
P upper machine performance index
mk
U
P minimum machine performance index
mk
f absolute frequency
© ISO 2020 – All rights reserved 1

---------------------- Page: 9 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

Σf cumulative absolute frequency
Σf % cumulative relative frequency in percent
i control variable, subscript used to identify the values of a variable
L lower specification limit
n sample size
X α % distribution fractile, percentile
α %
th
X i value in a sample
i
σ standard deviation, population
S standard deviation, sample statistic
U upper specification limit
z fractile of the standardized normal distribution from −∞ to α
α
μ population mean value in relation to the machine location
arithmetic mean value, sample
X
2
fractile of the chi-squared distribution
χ
α
5 Pre-conditions for application
5.1 General
The pre-conditions given in 5.2 and 5.6 are the minimum and may be exceeded if needed. In this type
of study, it is important to maintain constant all factors, other than the machine, which can influence
the results, if the study is to properly represent the machine itself, e.g. the same operator, same batch of
material, etc.
5.2 Number of parts to be used in the study
The number specified is usually 100. However, if the pattern of variation is expected to form a non-
normal distribution, the number of parts should be at least 100. The methods given within this document
may also be used when conducting audits of a process, in which case the number of measurements
taken might be less than the above number, e.g. 50.
NOTE 1 This is to ensure that a reasonably narrow confidence interval can be calculated for the machine
performance indices when a normal distribution has been used. The interval is approximately ±12 % of the
estimated index with a confidence of 90 % for samples of 100.
Some machines have very slow cycle times and a ‘run’ cannot produce 100 parts. In such circumstances,
it is necessary to proceed with available data. The minimum number that this document recommends
with the methods described herein is 30.
NOTE 2 Special techniques beyond the scope of this document exist for smaller sample sizes.
By contrast, for a machine that produces parts at a very high rate, e.g. a rivet-making machine,
the sampling strategy can require alteration since 100 parts can be produced in a few seconds. In
circumstances such as these, several studies can be required each allowing a different sampling
approach to examine the machine’s behaviour.
2 © ISO 2020 – All rights reserved

---------------------- Page: 10 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

5.3 Materials to be used
Ensure all input materials to be used in the study have been checked, conform to specifications and
belong to the same batches. It is not advised that a study be conducted with materials that are outside
specification since this could lead to unrepresentative results.
Care should be exercised not to introduce any other sources of variation other than those to be studied.
A typical example is where a machine run has to change to another batch of a particular material within
a single process batch, and batch material variation is not included in the study. In this instance, only
data taken while the first batch of that particular material was in use should be used in the analysis.
5.4 Measurement system
Ensure the measurement system used during the study has adequate properties and is calibrated,
and the measurement system variation has been quantified and minimized. Special studies on
the measurement system should be undertaken to establish the amount of variation present due to
measuring. The measurement system should ideally have a combined standard uncertainty u of less
MS
than 10 % of the standard deviation of the characteristic that the machine study is to investigate, as
determined through a properly conducted measurement systems analysis. This analysis should address
the issues of bias, calibration, linearity and discrimination. The resolution shall be lower than 1/20 of
the specification interval.
It is appropriate to calculate the expanded uncertainty U of the measurement process and to express
MP
the result as a percentage of a given tolerance. If the expanded uncertainty U does not exceed
MP
15 % of the tolerance, it may be regarded as acceptable, dependent upon application. If it exceeds
15 %, the measurement process should be regarded as inappropriate. Should a study be performed
using a measurement process with an uncertainty worse than these recommendations, some wrong
conclusions can be drawn from the study. Refer to ISO 22514-7 for more information about the
calculation of the measurement system and measurement process capability. Users who prefer doing
measurement systems analysis and gauge repeatability and reproducibility can refer to ISO/TR 12888
for more information.
5.5 Running the study
Ensure an uninterrupted run takes place, under normal operating conditions. This includes any warm-
up time for the machine necessary to bring it up to its usual operating condition and with the machine
set at nominal for the characteristic to be studied. If the machine is stopped during the study for
whatever reason, either re-run the study or analyse the data collected, as long as sufficient data have
been collected and as long as the repeatability conditions have not been violated. Under no circumstance
shall less than 30 consecutive results be used, to conclude the acceptance of the machine performance.
5.6 Special circumstances
In a multiple fixture set-up, multiple-cavity or multi-stream situation, each station, fixture, cavity or
stream should be treated as a separate machine for machine performance purposes since those streams
can violate the repeatability conditions.
In the case of a multiple-cavity tool, some extra studies may be performed to examine the between-
cavity and within-cavity variation, see ISO 22514-8.
6 Data collection
6.1 Traceability of data
It is important for all data to be traceable so that unexpected values can be investigated. The
collection sequence should be preserved so that a time series can be plotted of the data that might
indicate unexpected variations. Such occurrences should be explained, and a decision taken about the
© ISO 2020 – All rights reserved 3

---------------------- Page: 11 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

admissibility of such data. A ‘log-book’ would be suitable for recording all machine settings, including
any prior work on the machine, e.g. maintenance, and for recording all events during the study, such as
adjustments.
6.2 Retention of specimens
Unless the tests performed are destructive in their nature, all specimens should be retained so that all
necessary examinations can be made. They should only be disposed of once the study is complete and
all conclusions determined.
6.3 Data recording
Data should be clearly recorded either electronically or on the appropriate analysis sheet in numerical
form to the appropriate number of significant digits, often one significant digit more than that of the
tolerance. This should be determined prior to the measuring process and is dependent on the resolution
of the measuring instrument.
7 Analysis
7.1 General
The analysis of the data generated in the study is often performed using computer programs, or by
manual means, examples of which are given within this clause.
7.2 Run chart
7.2.1 Purpose
When conducting a machine study, it is important to understand whether the data collected form a
single and stable pattern or not. There are occasions when the conditions within the machine under
study lead to a drift in its settings that influence the pattern of data produced. There might be occasions
when an unauthorized adjustment has been made to the machine, or data have been mixed in some way.
Such an event should stop the study and a new study should be begun. A run chart is helpful to identify
such circumstances. The pattern on the run chart in Figure 1 (see also Table 1) might have been caused
by a slight trend within the first 25 items or something might have gone wrong with the machine itself
or it is being used wrongly.
If such a systematic influence had been proven, it would have been necessary to take special measures
according to the circumstances. These might range between repeating the whole study to analysing the
data in its separate parts or eliminating certain results.
ISO 7870-1 contains guidance about the application of control charts and their associated statistical
tests that should be applied to plots such as that shown in Figure 1 to assist with the interpretation of
the plots.
7.2.2 Review the plot
Inspect the plot for evidence of instability. This can appear as a step change in the data. Other patterns
might appear such as a drift. It is possible to use control limits and control chart rules to assess, easily,
for any other assignable causes in the data. The data might be put into an individual and moving range
chart to check for potential outliers in the data. (See ISO 7870-2 for further information about such
limits and rules.)
There exists a number of software products that can replace the manual methods. These have become
popular because they produce the graphs mentioned above quickly and easily.
4 © ISO 2020 – All rights reserved

---------------------- Page: 12 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

Table 1 — Example 1 — Example of observed values
Sample
1 to 10 11 to 20 21 to 30 31 to 40 41 to 50 51 to 60 61 to 70 71 to 80 81 to 90 91 to 100
No.
10,006 9 10,007 3 10,007 3 10,006 9 10,007 0 10,007 1 10,006 8 10,007 2 10,007 7 10,006 9
10,007 5 10,007 0 10,007 3 10,007 0 10,006 7 10,006 8 10,007 2 10,007 5 10,007 8 10,007 3
10,007 5 10,007 5 10,006 4 10,007 7 10,006 7 10,006 5 10,007 4 10,006 6 10,007 4 10,006 9
10,007 6 10,007 1 10,007 0 10,007 3 10,007 2 10,006 5 10,006 6 10,007 6 10,006 9 10,007 2
10,006 6 10,007 5 10,006 8 10,007 5 10,007 4 10,006 9 10,006 7 10,007 0 10,007 3 10,007 2
Diameter
in mm
10,007 5 10,006 9 10,006 2 10,007 4 10,006 3 10,006 6 10,007 2 10,007 2 10,007 1 10,006 9
10,007 5 10,007 2 10,007 1 10,007 0 10,007 3 10,007 6 10,006 8 10,006 4 10,006 8 10,006 6
10,007 1 10,006 9 10,007 2 10,006 7 10,006 9 10,007 2 10,006 6 10,007 6 10,007 1 10,007 1
10,007 4 10,006 8 10,007 2 10,006 9 10,006 9 10,007 5 10,007 1 10,007 9 10,007 1 10,007 0
10,007 0 10,007 5 10,006 9 10,006 9 10,007 5 10,007 2 10,007 0 10,007 5 10,006 6 10,006 8

Key
X observation number (i)
Y diameter in mm
Figure 1 — Example 1 — Run chart
7.3 Analyse the pattern of the data
7.3.1 Software approach
The data should be entered into a software tool and a histogram produced of the data. There exist a
number of suitable software products that carry out such analysis. Figure 2 shows the histogram of the
data from Figure 1.
© ISO 2020 – All rights reserved 5

---------------------- Page: 13 ----------------------
SIST ISO 22514-3:2021
ISO 22514-3:2020(E)

Key
X diameter in mm
Y absolute frequency (f)
Figure 2 — Example 1 — Histogram for normally distributed data
7.3.2 Check the pattern of the data
Study the pattern of the data to see if it conforms to a known distribution. Investigate the cause if
the data appear to form a quite different pattern. If the data do not form a normal distribution, it can
become necessary to employ a different distribution model. An analysis carried out on non-normal
data using the normal distribution can produce inaccurate results. Non-normality can occur from
circumstances where the data are limited in some way, such as the results of measurements of stress
or of concentricity. There might be some anticipation of non-normal data if geometrical tolerances have
been specified for a dimension or characteristic, for example. Consult the Bibliography for assistance
in determining if the data follow a known distribution model (e.g. ISO 5479) as well as in using other
statistical procedures beyond the scope of this document.
Special cases, such as skewed distributions and bimodal data, are discussed in 7.5.
If similar studies have been conducted prior to the current one, there can be a certain expectation of
what the distribution might be. Engineering knowledge might also suggest what the pattern ought to
...

INTERNATIONAL ISO
STANDARD 22514-3
Second edition
2020-12
Statistical methods in process
management — Capability and
performance —
Part 3:
Machine performance studies for
measured data on discrete parts
Méthodes statistiques dans la gestion de processus — Aptitude et
performance —
Partie 3: Études de performance de machines pour des données
mesurées sur des parties discrètes
Reference number
ISO 22514-3:2020(E)
©
ISO 2020

---------------------- Page: 1 ----------------------
ISO 22514-3:2020(E)

COPYRIGHT PROTECTED DOCUMENT
© ISO 2020
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting
on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address
below or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii © ISO 2020 – All rights reserved

---------------------- Page: 2 ----------------------
ISO 22514-3:2020(E)

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols . 1
5 Pre-conditions for application . 2
5.1 General . 2
5.2 Number of parts to be used in the study . 2
5.3 Materials to be used. 3
5.4 Measurement system. 3
5.5 Running the study . 3
5.6 Special circumstances . 3
6 Data collection . 3
6.1 Traceability of data . 3
6.2 Retention of specimens . 4
6.3 Data recording . 4
7 Analysis . 4
7.1 General . 4
7.2 Run chart . 4
7.2.1 Purpose . 4
7.2.2 Review the plot . 4
7.3 Analyse the pattern of the data . 5
7.3.1 Software approach . 5
7.3.2 Check the pattern of the data . 6
7.3.3 Summarize the data . 6
7.3.4 Manual approach . 6
7.4 Produce a probability plot . 9
7.4.1 General. 9
7.4.2 Analyse the data . 9
7.5 Special cases .10
7.5.1 Data indicate a skewed distribution .10
7.5.2 Bimodal data .11
7.5.3 Truncated data .12
7.5.4 Censored data .13
7.6 Calculation of machine performance indices .13
7.6.1 General procedure .13
7.6.2 Data following a normal distribution .14
8 Reporting .14
8.1 Test report .14
8.2 Confidence intervals .15
8.2.1 General.15
8.2.2 Indices calculated with the data following a normal distribution .15
8.2.3 Indices calculated with data following a non-normal distribution .16
9 Actions following a machine performance study .16
Annex A (informative) Tables and worksheets .17
Bibliography .19
© ISO 2020 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO 22514-3:2020(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.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/ directives).
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. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/ patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to
the World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see
www .iso .org/ iso/ foreword .html.
This document was prepared by Technical Committee ISO/TC 69, Applications of statistical methods,
Subcommittee SC 4, Applications of statistical methods in product and process management.
This second edition cancels and replaces the first edition (ISO 22514-3:2008), which has been
technically revised.
The main changes compared to the previous edition are as follows:
— updated and improved figures and computer outputs.
A list of all parts in the ISO 22514 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www .iso .org/ members .html.
iv © ISO 2020 – All rights reserved

---------------------- Page: 4 ----------------------
ISO 22514-3:2020(E)

Introduction
This document has been prepared to provide guidance in circumstances where a study is necessary
to determine if the output from a machine, for example, is acceptable according to some criteria. Such
circumstances are common in engineering when the purpose for the study is part of an acceptance
trial. These studies can also be used when diagnosis is required concerning a machine’s current level of
performance or as part of a problem-solving effort. The method is very versatile and has been applied
to many situations.
Machine performance studies of this type provide information about the behaviour of a machine under
very restricted conditions such as limiting, as far as possible, external sources of variation that are
commonplace within a process, e.g. multi-factor and multi-level situations. The data gathered in a
study might come from items made consecutively, although this may be altered according to the study
requirements. The data are assumed to have been, generally, gathered manually.
The study procedure and reporting are of interest to engineers, supervisors and management wishing
to establish whether a machine should be purchased or put in for maintenance, to assist in problem-
solving or to understand the level of variation due to the machine itself.
© ISO 2020 – All rights reserved v

---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO 22514-3:2020(E)
Statistical methods in process management — Capability
and performance —
Part 3:
Machine performance studies for measured data on
discrete parts
1 Scope
This document describes the steps for conducting short-term performance studies that are typically
performed on machines (including devices, appliances, apparatuses) where parts produced
consecutively under repeatability conditions are considered. The number of observations to be analysed
vary according to the patterns the data produce, or if the runs (the rate at which items are produced)
on the machine are low in quantity. The methods are not considered suitable where the sample size
produced is less than 30 observations. Methods for handling the data and carrying out the calculations
are described. In addition, machine performance indices and the actions required at the conclusion of a
machine performance study are described.
This document is not applicable when tool wear patterns are expected to be present during the duration
of the study, nor if autocorrelation between observations is present. The situation where a machine has
captured the data, sometimes thousands of data points collected in a minute, is not considered suitable
for the application of this document.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
No terms and definitions are listed in this document.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at http:// www .electropedia .org/
4 Symbols
P probability
P machine performance index
m
P lower machine performance index
mk
L
P upper machine performance index
mk
U
P minimum machine performance index
mk
f absolute frequency
© ISO 2020 – All rights reserved 1

---------------------- Page: 6 ----------------------
ISO 22514-3:2020(E)

Σf cumulative absolute frequency
Σf % cumulative relative frequency in percent
i control variable, subscript used to identify the values of a variable
L lower specification limit
n sample size
X α % distribution fractile, percentile
α %
th
X i value in a sample
i
σ standard deviation, population
S standard deviation, sample statistic
U upper specification limit
z fractile of the standardized normal distribution from −∞ to α
α
μ population mean value in relation to the machine location
arithmetic mean value, sample
X
2
fractile of the chi-squared distribution
χ
α
5 Pre-conditions for application
5.1 General
The pre-conditions given in 5.2 and 5.6 are the minimum and may be exceeded if needed. In this type
of study, it is important to maintain constant all factors, other than the machine, which can influence
the results, if the study is to properly represent the machine itself, e.g. the same operator, same batch of
material, etc.
5.2 Number of parts to be used in the study
The number specified is usually 100. However, if the pattern of variation is expected to form a non-
normal distribution, the number of parts should be at least 100. The methods given within this document
may also be used when conducting audits of a process, in which case the number of measurements
taken might be less than the above number, e.g. 50.
NOTE 1 This is to ensure that a reasonably narrow confidence interval can be calculated for the machine
performance indices when a normal distribution has been used. The interval is approximately ±12 % of the
estimated index with a confidence of 90 % for samples of 100.
Some machines have very slow cycle times and a ‘run’ cannot produce 100 parts. In such circumstances,
it is necessary to proceed with available data. The minimum number that this document recommends
with the methods described herein is 30.
NOTE 2 Special techniques beyond the scope of this document exist for smaller sample sizes.
By contrast, for a machine that produces parts at a very high rate, e.g. a rivet-making machine,
the sampling strategy can require alteration since 100 parts can be produced in a few seconds. In
circumstances such as these, several studies can be required each allowing a different sampling
approach to examine the machine’s behaviour.
2 © ISO 2020 – All rights reserved

---------------------- Page: 7 ----------------------
ISO 22514-3:2020(E)

5.3 Materials to be used
Ensure all input materials to be used in the study have been checked, conform to specifications and
belong to the same batches. It is not advised that a study be conducted with materials that are outside
specification since this could lead to unrepresentative results.
Care should be exercised not to introduce any other sources of variation other than those to be studied.
A typical example is where a machine run has to change to another batch of a particular material within
a single process batch, and batch material variation is not included in the study. In this instance, only
data taken while the first batch of that particular material was in use should be used in the analysis.
5.4 Measurement system
Ensure the measurement system used during the study has adequate properties and is calibrated,
and the measurement system variation has been quantified and minimized. Special studies on
the measurement system should be undertaken to establish the amount of variation present due to
measuring. The measurement system should ideally have a combined standard uncertainty u of less
MS
than 10 % of the standard deviation of the characteristic that the machine study is to investigate, as
determined through a properly conducted measurement systems analysis. This analysis should address
the issues of bias, calibration, linearity and discrimination. The resolution shall be lower than 1/20 of
the specification interval.
It is appropriate to calculate the expanded uncertainty U of the measurement process and to express
MP
the result as a percentage of a given tolerance. If the expanded uncertainty U does not exceed
MP
15 % of the tolerance, it may be regarded as acceptable, dependent upon application. If it exceeds
15 %, the measurement process should be regarded as inappropriate. Should a study be performed
using a measurement process with an uncertainty worse than these recommendations, some wrong
conclusions can be drawn from the study. Refer to ISO 22514-7 for more information about the
calculation of the measurement system and measurement process capability. Users who prefer doing
measurement systems analysis and gauge repeatability and reproducibility can refer to ISO/TR 12888
for more information.
5.5 Running the study
Ensure an uninterrupted run takes place, under normal operating conditions. This includes any warm-
up time for the machine necessary to bring it up to its usual operating condition and with the machine
set at nominal for the characteristic to be studied. If the machine is stopped during the study for
whatever reason, either re-run the study or analyse the data collected, as long as sufficient data have
been collected and as long as the repeatability conditions have not been violated. Under no circumstance
shall less than 30 consecutive results be used, to conclude the acceptance of the machine performance.
5.6 Special circumstances
In a multiple fixture set-up, multiple-cavity or multi-stream situation, each station, fixture, cavity or
stream should be treated as a separate machine for machine performance purposes since those streams
can violate the repeatability conditions.
In the case of a multiple-cavity tool, some extra studies may be performed to examine the between-
cavity and within-cavity variation, see ISO 22514-8.
6 Data collection
6.1 Traceability of data
It is important for all data to be traceable so that unexpected values can be investigated. The
collection sequence should be preserved so that a time series can be plotted of the data that might
indicate unexpected variations. Such occurrences should be explained, and a decision taken about the
© ISO 2020 – All rights reserved 3

---------------------- Page: 8 ----------------------
ISO 22514-3:2020(E)

admissibility of such data. A ‘log-book’ would be suitable for recording all machine settings, including
any prior work on the machine, e.g. maintenance, and for recording all events during the study, such as
adjustments.
6.2 Retention of specimens
Unless the tests performed are destructive in their nature, all specimens should be retained so that all
necessary examinations can be made. They should only be disposed of once the study is complete and
all conclusions determined.
6.3 Data recording
Data should be clearly recorded either electronically or on the appropriate analysis sheet in numerical
form to the appropriate number of significant digits, often one significant digit more than that of the
tolerance. This should be determined prior to the measuring process and is dependent on the resolution
of the measuring instrument.
7 Analysis
7.1 General
The analysis of the data generated in the study is often performed using computer programs, or by
manual means, examples of which are given within this clause.
7.2 Run chart
7.2.1 Purpose
When conducting a machine study, it is important to understand whether the data collected form a
single and stable pattern or not. There are occasions when the conditions within the machine under
study lead to a drift in its settings that influence the pattern of data produced. There might be occasions
when an unauthorized adjustment has been made to the machine, or data have been mixed in some way.
Such an event should stop the study and a new study should be begun. A run chart is helpful to identify
such circumstances. The pattern on the run chart in Figure 1 (see also Table 1) might have been caused
by a slight trend within the first 25 items or something might have gone wrong with the machine itself
or it is being used wrongly.
If such a systematic influence had been proven, it would have been necessary to take special measures
according to the circumstances. These might range between repeating the whole study to analysing the
data in its separate parts or eliminating certain results.
ISO 7870-1 contains guidance about the application of control charts and their associated statistical
tests that should be applied to plots such as that shown in Figure 1 to assist with the interpretation of
the plots.
7.2.2 Review the plot
Inspect the plot for evidence of instability. This can appear as a step change in the data. Other patterns
might appear such as a drift. It is possible to use control limits and control chart rules to assess, easily,
for any other assignable causes in the data. The data might be put into an individual and moving range
chart to check for potential outliers in the data. (See ISO 7870-2 for further information about such
limits and rules.)
There exists a number of software products that can replace the manual methods. These have become
popular because they produce the graphs mentioned above quickly and easily.
4 © ISO 2020 – All rights reserved

---------------------- Page: 9 ----------------------
ISO 22514-3:2020(E)

Table 1 — Example 1 — Example of observed values
Sample
1 to 10 11 to 20 21 to 30 31 to 40 41 to 50 51 to 60 61 to 70 71 to 80 81 to 90 91 to 100
No.
10,006 9 10,007 3 10,007 3 10,006 9 10,007 0 10,007 1 10,006 8 10,007 2 10,007 7 10,006 9
10,007 5 10,007 0 10,007 3 10,007 0 10,006 7 10,006 8 10,007 2 10,007 5 10,007 8 10,007 3
10,007 5 10,007 5 10,006 4 10,007 7 10,006 7 10,006 5 10,007 4 10,006 6 10,007 4 10,006 9
10,007 6 10,007 1 10,007 0 10,007 3 10,007 2 10,006 5 10,006 6 10,007 6 10,006 9 10,007 2
10,006 6 10,007 5 10,006 8 10,007 5 10,007 4 10,006 9 10,006 7 10,007 0 10,007 3 10,007 2
Diameter
in mm
10,007 5 10,006 9 10,006 2 10,007 4 10,006 3 10,006 6 10,007 2 10,007 2 10,007 1 10,006 9
10,007 5 10,007 2 10,007 1 10,007 0 10,007 3 10,007 6 10,006 8 10,006 4 10,006 8 10,006 6
10,007 1 10,006 9 10,007 2 10,006 7 10,006 9 10,007 2 10,006 6 10,007 6 10,007 1 10,007 1
10,007 4 10,006 8 10,007 2 10,006 9 10,006 9 10,007 5 10,007 1 10,007 9 10,007 1 10,007 0
10,007 0 10,007 5 10,006 9 10,006 9 10,007 5 10,007 2 10,007 0 10,007 5 10,006 6 10,006 8

Key
X observation number (i)
Y diameter in mm
Figure 1 — Example 1 — Run chart
7.3 Analyse the pattern of the data
7.3.1 Software approach
The data should be entered into a software tool and a histogram produced of the data. There exist a
number of suitable software products that carry out such analysis. Figure 2 shows the histogram of the
data from Figure 1.
© ISO 2020 – All rights reserved 5

---------------------- Page: 10 ----------------------
ISO 22514-3:2020(E)

Key
X diameter in mm
Y absolute frequency (f)
Figure 2 — Example 1 — Histogram for normally distributed data
7.3.2 Check the pattern of the data
Study the pattern of the data to see if it conforms to a known distribution. Investigate the cause if
the data appear to form a quite different pattern. If the data do not form a normal distribution, it can
become necessary to employ a different distribution model. An analysis carried out on non-normal
data using the normal distribution can produce inaccurate results. Non-normality can occur from
circumstances where the data are limited in some way, such as the results of measurements of stress
or of concentricity. There might be some anticipation of non-normal data if geometrical tolerances have
been specified for a dimension or characteristic, for example. Consult the Bibliography for assistance
in determining if the data follow a known distribution model (e.g. ISO 5479) as well as in using other
statistical procedures beyond the scope of this document.
Special cases, such as skewed distributions and bimodal data, are discussed in 7.5.
If similar studies have been conducted prior to the current one, there can be a certain expectation of
what the distribution might be. Engineering knowledge might also suggest what the pattern ought to
be and this can be an important reference should the pattern appear unusual. It can be that something
has happened to induce a non-random pattern and an investigation should be conducted.
Misleading results can occur if the computer program used does not check for normality.
7.3.3 Summarize the data
Report the sample mean X and the sample standard deviation (S). For the mean value, this is usually
()
one decimal place more than the resolution of the raw data, and three more digits for the standard
deviation. If the distribution is non-normal, report the sample statistics corresponding to the relevant
parameters for the assumed distribution.
7.3.4 Manual approach
A simple manner to begin analysing the shape of the frequency distribution is to construct a tally chart.
The data are arranged into ‘classes’. If the number of classes is not pre-determined by resolution of the
measurement device, the number of classes should be between 5 and 20.
6 © ISO 2020 – All rights reserved

---------------------- Page: 11 ----------------------
ISO 22514-3:2020(E)

To find the ‘class width’, the range of the data shall be determined and divided by the number of classes.
A common recommendation is to use a number of classes of about n . The result shall be rounded to
the next
...

FINAL
INTERNATIONAL ISO/FDIS
DRAFT
STANDARD 22514-3
ISO/TC 69/SC 4
Statistical methods in process
Secretariat: DIN
management — Capability and
Voting begins on:
2020-09-01 performance —
Voting terminates on:
Part 3:
2020-10-27
Machine performance studies for
measured data on discrete parts
Méthodes statistiques dans la gestion de processus — Aptitude et
performance —
Partie 3: Études de performance de machines pour des données
mesurées sur des parties discrètes
RECIPIENTS OF THIS DRAFT ARE INVITED TO
SUBMIT, WITH THEIR COMMENTS, NOTIFICATION
OF ANY RELEVANT PATENT RIGHTS OF WHICH
THEY ARE AWARE AND TO PROVIDE SUPPOR TING
DOCUMENTATION.
IN ADDITION TO THEIR EVALUATION AS
Reference number
BEING ACCEPTABLE FOR INDUSTRIAL, TECHNO-
ISO/FDIS 22514-3:2020(E)
LOGICAL, COMMERCIAL AND USER PURPOSES,
DRAFT INTERNATIONAL STANDARDS MAY ON
OCCASION HAVE TO BE CONSIDERED IN THE
LIGHT OF THEIR POTENTIAL TO BECOME STAN-
DARDS TO WHICH REFERENCE MAY BE MADE IN
©
NATIONAL REGULATIONS. ISO 2020

---------------------- Page: 1 ----------------------
ISO/FDIS 22514-3:2020(E)

COPYRIGHT PROTECTED DOCUMENT
© ISO 2020
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting
on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address
below or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii © ISO 2020 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/FDIS 22514-3:2020(E)

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols . 1
5 Pre-conditions for application . 2
5.1 General . 2
5.2 Number of parts to be used in the study . 2
5.3 Materials to be used. 3
5.4 Measurement system. 3
5.5 Running the study . 3
5.6 Special circumstances . 3
6 Data collection . 3
6.1 Traceability of data . 3
6.2 Retention of specimens . 4
6.3 Data recording . 4
7 Analysis . 4
7.1 General . 4
7.2 Run chart . 4
7.2.1 Purpose . 4
7.2.2 Review the plot . 4
7.3 Analyse the pattern of the data . 5
7.3.1 Software approach . 5
7.3.2 Check the pattern of the data . 6
7.3.3 Summarize the data . 6
7.3.4 Manual approach . 6
7.4 Produce a probability plot . 9
7.4.1 General. 9
7.4.2 Analyse the data . 9
7.5 Special cases .10
7.5.1 Data indicate a skewed distribution .10
7.5.2 Bimodal data .11
7.5.3 Truncated data .12
7.5.4 Censored data .13
7.6 Calculation of machine performance indices .13
7.6.1 General procedure .13
7.6.2 Data following a normal distribution .14
8 Reporting .14
8.1 Test report .14
8.2 Confidence intervals .15
8.2.1 General.15
8.2.2 Indices calculated with the data following a normal distribution .15
8.2.3 Indices calculated with data following a non-normal distribution .16
9 Actions following a machine performance study .16
Annex A (informative) Tables and worksheets .17
Bibliography .19
© ISO 2020 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO/FDIS 22514-3:2020(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.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/ directives).
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. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/ patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to
the World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see
www .iso .org/ iso/ foreword .html.
This document was prepared by Technical Committee ISO/TC 69, Applications of statistical methods,
Subcommittee SC 4, Applications of statistical methods in product and process management.
This second edition cancels and replaces the first edition (ISO 22514-3:2008), which has been
technically revised.
The main changes compared to the previous edition are as follows:
— updated and improved figures and computer outputs.
A list of all parts in the ISO 22514 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www .iso .org/ members .html.
iv © ISO 2020 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/FDIS 22514-3:2020(E)

Introduction
This document has been prepared to provide guidance in circumstances where a study is necessary
to determine if the output from a machine, for example, is acceptable according to some criteria. Such
circumstances are common in engineering when the purpose for the study is part of an acceptance
trial. These studies can also be used when diagnosis is required concerning a machine’s current level of
performance or as part of a problem-solving effort. The method is very versatile and has been applied
to many situations.
Machine performance studies of this type provide information about the behaviour of a machine under
very restricted conditions such as limiting, as far as possible, external sources of variation that are
commonplace within a process, e.g. multi-factor and multi-level situations. The data gathered in a
study might come from items made consecutively, although this may be altered according to the study
requirements. The data are assumed to have been, generally, gathered manually.
The study procedure and reporting are of interest to engineers, supervisors and management wishing
to establish whether a machine should be purchased or put in for maintenance, to assist in problem-
solving or to understand the level of variation due to the machine itself.
© ISO 2020 – All rights reserved v

---------------------- Page: 5 ----------------------
FINAL DRAFT INTERNATIONAL STANDARD ISO/FDIS 22514-3:2020(E)
Statistical methods in process management — Capability
and performance —
Part 3:
Machine performance studies for measured data on
discrete parts
1 Scope
This document describes the steps for conducting short-term performance studies that are typically
performed on machines (including devices, appliances, apparatuses) where parts produced
consecutively under repeatability conditions are considered. The number of observations to be analysed
vary according to the patterns the data produce, or if the runs (the rate at which items are produced)
on the machine are low in quantity. The methods are not considered suitable where the sample size
produced is less than 30 observations. Methods for handling the data and carrying out the calculations
are described. In addition, machine performance indices and the actions required at the conclusion of a
machine performance study are described.
This document is not applicable when tool wear patterns are expected to be present during the duration
of the study, nor if autocorrelation between observations is present. The situation where a machine has
captured the data, sometimes thousands of data points collected in a minute, is not considered suitable
for the application of this document.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
No terms and definitions are listed in this document.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at http:// www .electropedia .org/
4 Symbols
P probability
P machine performance index
m
P lower machine performance index
mk
L
P upper machine performance index
mk
U
P minimum machine performance index
mk
f absolute frequency
© ISO 2020 – All rights reserved 1

---------------------- Page: 6 ----------------------
ISO/FDIS 22514-3:2020(E)

Σf cumulative absolute frequency
Σf % cumulative relative frequency in percent
i control variable, subscript used to identify the values of a variable
L lower specification limit
n sample size
X α % distribution fractile, percentile
α %
th
X i value in a sample
i
σ standard deviation, population
S standard deviation, sample statistic
U upper specification limit
z fractile of the standardized normal distribution from −∞ to α
α
μ population mean value in relation to the machine location
arithmetic mean value, sample
X
2
fractile of the chi-squared distribution
χ
α
5 Pre-conditions for application
5.1 General
The pre-conditions given in 5.2 and 5.6 are the minimum and may be exceeded if needed. In this type
of study, it is important to maintain constant all factors, other than the machine, which can influence
the results, if the study is to properly represent the machine itself, e.g. the same operator, same batch of
material, etc.
5.2 Number of parts to be used in the study
The number specified is usually 100. However, if the pattern of variation is expected to form a non-
normal distribution, the number of parts should be at least 100. The methods given within this document
may also be used when conducting audits of a process, in which case the number of measurements
taken might be less than the above number, e.g. 50.
NOTE 1 This is to ensure that a reasonably narrow confidence interval can be calculated for the machine
performance indices when a normal distribution has been used. The interval is approximately ±12 % of the
estimated index with a confidence of 90 % for samples of 100.
Some machines have very slow cycle times and a ‘run’ cannot produce 100 parts. In such circumstances,
it is necessary to proceed with available data. The minimum number that this document recommends
with the methods described herein is 30.
NOTE 2 Special techniques beyond the scope of this document exist for smaller sample sizes.
By contrast, for a machine that produces parts at a very high rate, e.g. a rivet-making machine,
the sampling strategy can require alteration since 100 parts can be produced in a few seconds. In
circumstances such as these, several studies can be required each allowing a different sampling
approach to examine the machine’s behaviour.
2 © ISO 2020 – All rights reserved

---------------------- Page: 7 ----------------------
ISO/FDIS 22514-3:2020(E)

5.3 Materials to be used
Ensure all input materials to be used in the study have been checked, conform to specifications and
belong to the same batches. It is not advised that a study be conducted with materials that are outside
specification since this could lead to unrepresentative results.
Care should be exercised not to introduce any other sources of variation other than those to be studied.
A typical example is where a machine run has to change to another batch of a particular material within
a single process batch, and batch material variation is not included in the study. In this instance, only
data taken while the first batch of that particular material was in use should be used in the analysis.
5.4 Measurement system
Ensure the measurement system used during the study has adequate properties and is calibrated,
and the measurement system variation has been quantified and minimized. Special studies on
the measurement system should be undertaken to establish the amount of variation present due to
measuring. The measurement system should ideally have a combined standard uncertainty u of less
MS
than 10 % of the standard deviation of the characteristic that the machine study is to investigate, as
determined through a properly conducted measurement systems analysis. This analysis should address
the issues of bias, calibration, linearity and discrimination. The resolution shall be lower than 1/20 of
the specification interval.
It is appropriate to calculate the expanded uncertainty U of the measurement process and to express
MP
the result as a percentage of a given tolerance. If the expanded uncertainty U does not exceed
MP
15 % of the tolerance, it may be regarded as acceptable, dependent upon application. If it exceeds
15 %, the measurement process should be regarded as inappropriate. Should a study be performed
using a measurement process with an uncertainty worse than these recommendations, some wrong
conclusions can be drawn from the study. Refer to ISO 22514-7 for more information about the
calculation of the measurement system and measurement process capability. Users who prefer doing
measurement systems analysis and gauge repeatability and reproducibility can refer to ISO/TR 12888
for more information.
5.5 Running the study
Ensure an uninterrupted run takes place, under normal operating conditions. This includes any warm-
up time for the machine necessary to bring it up to its usual operating condition and with the machine
set at nominal for the characteristic to be studied. If the machine is stopped during the study for
whatever reason, either re-run the study or analyse the data collected, as long as sufficient data have
been collected and as long as the repeatability conditions have not been violated. Under no circumstance
shall less than 30 consecutive results be used, to conclude the acceptance of the machine performance.
5.6 Special circumstances
In a multiple fixture set-up, multiple-cavity or multi-stream situation, each station, fixture, cavity or
stream should be treated as a separate machine for machine performance purposes since those streams
can violate the repeatability conditions.
In the case of a multiple-cavity tool, some extra studies may be performed to examine the between-
cavity and within-cavity variation, see ISO 22514-8.
6 Data collection
6.1 Traceability of data
It is important for all data to be traceable so that unexpected values can be investigated. The
collection sequence should be preserved so that a time series can be plotted of the data that might
indicate unexpected variations. Such occurrences should be explained, and a decision taken about the
© ISO 2020 – All rights reserved 3

---------------------- Page: 8 ----------------------
ISO/FDIS 22514-3:2020(E)

admissibility of such data. A ‘log-book’ would be suitable for recording all machine settings, including
any prior work on the machine, e.g. maintenance, and for recording all events during the study, such as
adjustments.
6.2 Retention of specimens
Unless the tests performed are destructive in their nature, all specimens should be retained so that all
necessary examinations can be made. They should only be disposed of once the study is complete and
all conclusions determined.
6.3 Data recording
Data should be clearly recorded either electronically or on the appropriate analysis sheet in numerical
form to the appropriate number of significant digits, often one significant digit more than that of the
tolerance. This should be determined prior to the measuring process and is dependent on the resolution
of the measuring instrument.
7 Analysis
7.1 General
The analysis of the data generated in the study is often performed using computer programs, or by
manual means, examples of which are given within this clause.
7.2 Run chart
7.2.1 Purpose
When conducting a machine study, it is important to understand whether the data collected form a
single and stable pattern or not. There are occasions when the conditions within the machine under
study lead to a drift in its settings that influence the pattern of data produced. There might be occasions
when an unauthorized adjustment has been made to the machine, or data have been mixed in some way.
Such an event should stop the study and a new study should be begun. A run chart is helpful to identify
such circumstances. The pattern on the run chart in Figure 1 (see also Table 1) might have been caused
by such an adjustment or something might have gone wrong with the machine itself or it is being used
wrongly.
If such a change has occurred, it is necessary to take special measures according to the circumstances.
These might range between repeating the whole study to analysing the data in its separate parts or
eliminating certain results.
ISO 7870-1 contains guidance about the application of control charts and their associated statistical
tests that should be applied to plots such as that shown in Figure 1 to assist with the interpretation of
the plots.
7.2.2 Review the plot
Inspect the plot for evidence of instability. This can appear as a step change in the data. Other patterns
might appear such as a drift. It is possible to use control limits and control chart rules to assess, easily,
for any other assignable causes in the data. The data might be put into an individual and moving range
chart to check for potential outliers in the data. (See ISO 7870-2 for further information about such
limits and rules.)
There exists a number of software products that can replace the manual methods. These have become
popular because they produce the graphs mentioned above quickly and easily.
4 © ISO 2020 – All rights reserved

---------------------- Page: 9 ----------------------
ISO/FDIS 22514-3:2020(E)

Table 1 — Example 1 — Example of observed values
Sample
1 to 10 11 to 20 21 to 30 31 to 40 41 to 50 51 to 60 61 to 70 71 to 80 81 to 90 91 to 100
No.
10,006 9 10,007 3 10,007 3 10,006 9 10,007 0 10,007 1 10,006 8 10,007 2 10,007 7 10,006 9
10,007 5 10,007 0 10,007 3 10,007 0 10,006 7 10,006 8 10,007 2 10,007 5 10,007 8 10,007 3
10,007 5 10,007 5 10,006 4 10,007 7 10,006 7 10,006 5 10,007 4 10,006 6 10,007 4 10,006 9
10,007 6 10,007 1 10,007 0 10,007 3 10,007 2 10,006 5 10,006 6 10,007 6 10,006 9 10,007 2
10,006 6 10,007 5 10,006 8 10,007 5 10,007 4 10,006 9 10,006 7 10,007 0 10,007 3 10,007 2
Diameter
in mm
10,007 5 10,006 9 10,006 2 10,007 4 10,006 3 10,006 6 10,007 2 10,007 2 10,007 1 10,006 9
10,007 5 10,007 2 10,007 1 10,007 0 10,007 3 10,007 6 10,006 8 10,006 4 10,006 8 10,006 6
10,007 1 10,006 9 10,007 2 10,006 7 10,006 9 10,007 2 10,006 6 10,007 6 10,007 1 10,007 1
10,007 4 10,006 8 10,007 2 10,006 9 10,006 9 10,007 5 10,007 1 10,007 9 10,007 1 10,007 0
10,007 0 10,007 5 10,006 9 10,006 9 10,007 5 10,007 2 10,007 0 10,007 5 10,006 6 10,006 8

Key
X observation number (i)
Y diameter in mm
Figure 1 — Example 1 — Run chart
7.3 Analyse the pattern of the data
7.3.1 Software approach
The data should be entered into a software tool and a histogram produced of the data. There exist a
number of suitable software products that carry out such analysis. Figure 2 shows the histogram of the
data from Figure 1.
© ISO 2020 – All rights reserved 5

---------------------- Page: 10 ----------------------
ISO/FDIS 22514-3:2020(E)

Key
X diameter in mm
Y absolute frequency (f)
Figure 2 — Example 1 — Histogram for normally distributed data
7.3.2 Check the pattern of the data
Study the pattern of the data to see if it conforms to a known distribution. Investigate the cause if
the data appear to form a quite different pattern. If the data do not form a normal distribution, it can
become necessary to employ a different distribution model. An analysis carried out on non-normal
data using the normal distribution can produce inaccurate results. Non-normality can occur from
circumstances where the data are limited in some way, such as the results of measurements of stress
or of concentricity. There might be some anticipation of non-normal data if geometrical tolerances have
been specified for a dimension or characteristic, for example. Consult the Bibliography for assistance
in determining if the data follow a known distribution model (e.g. ISO 5479) as well as in using other
statistical procedures beyond the scope of this document.
Special cases, such as skewed distributions and bimodal data, are discussed in 7.5.
If similar studies have been conducted prior to the current one, there can be a certain expectation of
what the distribution might be. Engineering knowledge might also suggest what the pattern ought to
be and this can be an important reference should the pattern appear unusual. It can be that something
has happened to induce a non-random pattern and an investigation should be conducted.
Misleading results can occur if the computer program used does not check for normality.
7.3.3 Summarize the data
Report the sample mean X and the sample standard deviation (S). For the mean value, this is usually
()
one decimal place more than the resolution of the raw data, and three more digits for the standard
deviation. If the distribution is non-normal, report the sample statistics corresponding to the relevant
parameters for the assumed distribution.
7.3.4 Manual approach
A simple manner to begin analys
...

FINAL
INTERNATIONAL ISO/FDIS
DRAFT
STANDARD 22514-3
ISO/TC 69/SC 4
Statistical methods in process
Secretariat: DIN
management — Capability and
Voting begins on:
2020-04-17 performance —
Voting terminates on:
Part 3:
2020-06-12
Machine performance studies for
measured data on discrete parts
Méthodes statistiques dans la gestion de processus — Aptitude et
performance —
Partie 3: Études de performance de machines pour des données
mesurées sur des parties discrètes
IMPORTANT — Please use this updated version dated 2020-04-06, and discard
any previous version of this FDIS.
RECIPIENTS OF THIS DRAFT ARE INVITED TO
SUBMIT, WITH THEIR COMMENTS, NOTIFICATION
OF ANY RELEVANT PATENT RIGHTS OF WHICH
THEY ARE AWARE AND TO PROVIDE SUPPOR TING
DOCUMENTATION.
IN ADDITION TO THEIR EVALUATION AS
Reference number
BEING ACCEPTABLE FOR INDUSTRIAL, TECHNO-
ISO/FDIS 22514-3:2020(E)
LOGICAL, COMMERCIAL AND USER PURPOSES,
DRAFT INTERNATIONAL STANDARDS MAY ON
OCCASION HAVE TO BE CONSIDERED IN THE
LIGHT OF THEIR POTENTIAL TO BECOME STAN-
DARDS TO WHICH REFERENCE MAY BE MADE IN
©
NATIONAL REGULATIONS. ISO 2020

---------------------- Page: 1 ----------------------
ISO/FDIS 22514-3:2020(E)

COPYRIGHT PROTECTED DOCUMENT
© ISO 2020
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting
on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address
below or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Fax: +41 22 749 09 47
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii © ISO 2020 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/FDIS 22514-3:2020(E)

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviations . 1
5 Pre-conditions for application . 2
5.1 General . 2
5.2 Number of parts to be used in the study . 2
5.3 Materials to be used. 3
5.4 Measurement system. 3
5.5 Running the study . 3
5.6 Special circumstances . 3
6 Data collection . 3
6.1 Traceability of data . 3
6.2 Retention of specimens . 4
6.3 Data recording . 4
7 Analysis . 4
7.1 General . 4
7.2 Run chart . 4
7.2.1 Purpose . 4
7.2.2 Review the plot . 4
7.3 Analyse the pattern of the data . 5
7.3.1 Software approach . 5
7.3.2 Check the pattern of the data . 5
7.3.3 Summarize the data . 6
7.3.4 Manual approach . 6
7.4 Produce a probability plot . 8
7.4.1 General. 8
7.4.2 Analyse the data . 8
7.5 Special cases . 9
7.5.1 Data indicate a skewed distribution . 9
7.5.2 Bimodal data .10
7.5.3 Truncated data .11
7.5.4 Censored data .11
7.6 Calculation of machine performance indices .12
7.6.1 General.12
7.6.2 Data following a normal distribution .12
8 Reporting .13
8.1 Test report .13
8.2 Confidence intervals .14
8.2.1 General.14
8.2.2 Indices calculated with the data following a normal distribution .14
8.2.3 Indices calculated with data following a non-normal distribution .14
9 Actions following a machine performance study .14
Annex A (informative) Estimated proportion beyond a specification limit —
Normal distribution .16
Bibliography .17
© ISO 2020 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO/FDIS 22514-3:2020(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.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/ directives).
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. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/ patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to the
World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www .iso .org/
iso/ foreword .html.
This document was prepared by Technical Committee ISO/TC 69, Applications of statistical methods,
Subcommittee SC 4, Applications of statistical methods in product and process management.
This second edition cancels and replaces the first edition (ISO 22514-3:2008), which has been
technically revised.
The main changes compared to the previous edition are as follows:
— updated and improved figures and computer outputs.
A list of all parts in the ISO 22514 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www .iso .org/ members .html.
iv © ISO 2020 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/FDIS 22514-3:2020(E)

Introduction
This document has been prepared to provide guidance in circumstances where a study is necessary
to determine if the output from a machine, for example, is acceptable according to some criteria. Such
circumstances are common in engineering when the purpose for the study is part of an acceptance
trial. These studies may also be used when diagnosis is required concerning a machine’s current level
of performance or as part of a problem solving effort. The method is very versatile and has been applied
to many situations.
Machine performance studies of this type provide information about the behaviour of a machine under
very restricted conditions such as limiting, as far as possible, external sources of variation that are
commonplace within a process, e.g. multi-factor and multi-level situations. The data gathered in a
study might come from items made consecutively, although this may be altered according to the study
requirements. The data are assumed to have been, generally, gathered manually.
The study procedure and reporting are of interest to engineers, supervisors and management wishing
to establish whether a machine should be purchased or put in for maintenance, to assist in problem
solving or to understand the level of variation due to the machine itself.
© ISO 2020 – All rights reserved v

---------------------- Page: 5 ----------------------
FINAL DRAFT INTERNATIONAL STANDARD ISO/FDIS 22514-3:2020(E)
Statistical methods in process management — Capability
and performance —
Part 3:
Machine performance studies for measured data on
discrete parts
1 Scope
This document describes the steps for conducting short-term performance studies that are typically
performed on machines where parts produced consecutively under repeatability conditions are
considered. The number of observations to be analysed vary according to the patterns the data produce,
or if the runs (the rate at which items are produced) on the machine are low in quantity. The methods
are not considered suitable where the sample size produced is less than 30 observations. Methods for
handling the data and carrying out the calculations are described. In addition, machine performance
indices and the actions required at the conclusion of a machine performance study are described.
This document is not applicable when tool wear patterns are expected to be present during the duration
of the study, nor if autocorrelation between observations is present. The situation where a machine has
captured the data, sometimes thousands of data points collected in a minute, is not considered suitable
for the application of this document.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
No terms and definitions are listed in this document.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at http:// www .electropedia .org/
4 Symbols and abbreviations
P machine performance index
m
P lower machine performance index
mk
L
P upper machine performance index
mk
U
P minimum machine performance index
mk
f frequency
Σf cumulative frequency
© ISO 2020 – All rights reserved 1

---------------------- Page: 6 ----------------------
ISO/FDIS 22514-3:2020(E)

i subscript used to identify values of a variable
L lower specification limit
N total sample size
X α % distribution fractile
α %
th
X i value in a sample
i
σ standard deviation, population
S standard deviation, sample statistic,
U upper specification limit
z fractile of the standardized normal distribution from −∞ to α
α
μ population mean value in relation to the machine location
arithmetic mean value, sample,
X
2
fractile of the chi-squared distribution
X
α
5 Pre-conditions for application
5.1 General
The pre-conditions given in 5.2 and 5.6 are the minimum and may be exceeded when needed. In this
type of study, it is important to maintain constant all factors, other than the machine, which can
influence the results, if the study is to properly represent the machine itself, e.g. the same operator,
same batch of material, etc.
5.2 Number of parts to be used in the study
The number specified is usually 100. However, if the pattern of variation is expected to form a non-
normal distribution, the number of parts should be at least 100. The methods given within this document
may also be used when conducting audits of a process, in which case the number of measurements
taken might be less than the above number, e.g. 50.
NOTE 1 This is to ensure that a reasonably narrow confidence interval can be calculated for the machine
performance indices when a normal distribution has been used. The interval is approximately ± 12 % of the
estimated index with a confidence of 90 % for samples of 100.
Some machines have very slow cycle times and a ‘run’ cannot produce 100 parts. In such circumstances,
it is necessary to proceed with available data. The minimum number that this document recommends
with the methods described herein is 30.
NOTE 2 Special techniques beyond the scope of this document exist for circumstances when there are fewer
samples.
By contrast, for a machine that produces parts at a very high rate, e.g. a rivet-making machine,
the sampling strategy can require alteration since 100 parts can be produced in a few seconds. In
circumstances such as these, several studies can be required each allowing a different sampling
approach to examine the machine’s behaviour.
2 © ISO 2020 – All rights reserved

---------------------- Page: 7 ----------------------
ISO/FDIS 22514-3:2020(E)

5.3 Materials to be used
Ensure all input materials to be used in the study have been checked, conform to specifications and
belong to the same batches. It is not advised that a study be conducted with materials that are outside
specification since this could lead to unrepresentative results.
Care should be exercised not to introduce any other sources of variation other than those to be studied.
A typical example is where a machine run has to change to another batch of a particular material within
a single process batch, and batch material variation is not included in the study. In this instance, only
data taken while the first batch of that particular material was in use should be used in the analysis.
5.4 Measurement system
Ensure the measurement system used during the study has adequate properties and is calibrated,
and the measurement process variation has been quantified and minimized. Special studies on
the measurement process should be undertaken to establish the amount of variation present due to
measuring. The measurement process should ideally have a combined repeatability and reproducibility
(GRR, gauge repeatability and reproducibility) of less than 10 % of the process spread of the
characteristic that the machine study is to investigate as determined through a properly conducted
measurement process analysis. This analysis should address the issues of bias, stability, linearity and
discrimination, as well as GRR and other influence factors.
It is appropriate to calculate the total measurement uncertainty instead of only MSA (measurement
system analysis) and GRR, and to express the result as a percentage of a given specification tolerance.
If the measurement process has between 10 % and 30 % of the tolerance, it may still be regarded as
acceptable, dependent upon application. If it exceeds 30 %, the measurement process should be regarded
as inappropriate. Should a study be performed using a measurement process with a performance worse
than these recommendations, some erroneous conclusions to the study might be reached. Refer to
ISO 22514-7 for more information about the calculation of measurement capability.
5.5 Running the study
Ensure an uninterrupted run takes place, under normal operating conditions. This includes any
warm-up time for the machine necessary to bring it up to its usual operating condition and with the
machine set at nominal for the characteristic to be studied. If the machine is stopped during the study
for whatever reason, either re-run the study again or analyse the data collected, as long as sufficient
data has been collected and as long as the repeatability conditions have not been violated. Under no
circumstance shall less than 30 consecutive results be used, to conclude the acceptance of the machine
performance.
5.6 Special circumstances
In a multiple fixture set-up, multiple-cavity or multi-stream situation, each station, fixture, cavity or
stream should be treated as a separate machine for machine performance purposes since those streams
can violate the repeatability conditions.
In the case of a multiple-cavity tool, some extra studies may be performed to examine the between-
cavity and within-cavity variation, see ISO 22514-8.
6 Data collection
6.1 Traceability of data
It is important for all data to be traceable so that unexpected values can be investigated. The
collection sequence should be preserved so that a time series can be plotted of the data that might
indicate unexpected variations. Such occurrences should be explained and a decision taken about the
admissibility of such data. A ‘log-book’ would be suitable for recording all machine settings, including
© ISO 2020 – All rights reserved 3

---------------------- Page: 8 ----------------------
ISO/FDIS 22514-3:2020(E)

any prior work on the machine, e.g. maintenance, and for recording all events during the study, such as
adjustments.
6.2 Retention of specimens
Unless the tests performed are destructive in their nature, all specimens should be retained so that all
necessary examinations can be made. They should only be disposed of once the study is complete and
all conclusions determined.
6.3 Data recording
Data should be clearly recorded either electronically or on the appropriate analysis sheet in numerical
form to the appropriate number of significant digits, often one significant digit more than that of
tolerance. This should be determined prior to the measuring process and is dependent on the resolution
of the measuring instrument.
7 Analysis
7.1 General
The analysis of the data generated in the study is often performed using computer programs, or by
manual means, examples of which are given within this clause.
7.2 Run chart
7.2.1 Purpose
When conducting a machine study, it is important to understand whether the data collected form a
single and stable pattern or not. There are occasions when the conditions within the machine under
study lead to a drift in its settings that influence the pattern of data produced. There might be occasions
when an unauthorized adjustment has been made to the machine or data have been mixed in some
way. Such an event should stop the study and a new study should be begun. A run chart is helpful to
identify such circumstances. The pattern on the run chart in Figure 1 might have been caused by such
an adjustment or something might have gone wrong with the machine itself or it is being used wrongly.
If a change such has occurred, it is necessary to take special measures according to the circumstances.
These might range between repeating the whole study to analysing the data in its separate parts or
eliminating certain results.
ISO 7870-1 contains guidance about the application of control charts and their associated statistical
tests that should be applied to plots such as that shown in Figure 1 to assist with the interpretation of
the plots.
7.2.2 Review the plot
Inspect the plot for evidence of instability. This can appear as a step change in the data. Other patterns
might appear such as a drift. It is possible to use control limits and control chart rules to assess, easily,
for any other assignable causes in the data. The data might be put into an individual and moving range
chart to check for potential outliers in the data. (See ISO 7870-2 for further information about such
limits and rules.)
There exist a number of software products that can replace the manual methods. These have become
popular because they produce the graphs mentioned above quickly and easily.
4 © ISO 2020 – All rights reserved

---------------------- Page: 9 ----------------------
ISO/FDIS 22514-3:2020(E)

Key
X observation number (N)
Y diameter, in mm
Figure 1 — Example of a run chart
7.3 Analyse the pattern of the data
7.3.1 Software approach
The data should be entered into a software tool and a histogram produced of the data. There exist a
number of suitable software products that carry out such analysis. Figure 2 shows the histogram of the
data from Figure 1.
Figure 2 — Example of a histogram for normally distributed data
7.3.2 Check the pattern of the data
Study the pattern of the data to see if it conforms to a known distribution. Investigate the cause if
the data appear to form a quite different pattern. If the data do not form a normal distribution, it can
become necessary to employ a different distribution model. An analysis carried out on non-normal
data using the normal distribution can produce inaccurate results. Non-normality can occur from
circumstances where the data are limited in some way, such as the results of measurements of stress
or of concentricity. There might be some anticipation of non-normal data if geometrical tolerances have
© ISO 2020 – All rights reserved 5

---------------------- Page: 10 ----------------------
ISO/FDIS 22514-3:2020(E)

been specified for a dimension or characteristic, for example. Consult the Bibliography for assistance
in determining if the data conform to a normal distribution (e.g. ISO 5479) as well as in using other
statistical procedures beyond the scope of this document.
Special cases, such as skewed distributions and bimodal data, are discussed in 5.6.
If similar studies have been conducted prior to the current one, there can be a certain expectation of
what the distribution might be. Scientific knowledge might also suggest what the pattern ought to be
and this can be an important reference should the pattern appear unusual. It can be that something has
happened to induce a non-random pattern and an investigation should be conducted.
Misleading results can occur if the computer program used does not check for normality.
7.3.3 Summarize the data
Report the sample mean X and the sample standard deviation (S). For the mean value, this is usually
()
one decimal place more than the resolution of the raw data, and for the standard deviation it is three
significant figures. If the distribution is non-normal, report the sample statistics corresponding to the
relevant parameters for the assumed distribution.
7.3.4 Manual approach
A simple manner to begin analysing the pattern the data form is to construct a tally chart.
The data are arranged into ‘classes’. The convention of counting the data into groups of five is often
used and an example of this can be seen in Figure 3. In this example, the data have been recorded to the
nearest 5 mm that is appropriate for the process from which the data are coming.
6 © ISO 2020 – All rights reserved

---------------------- Page: 11 ----------------------
ISO/FDIS 22514-3:2020(E)

Figure 3 — Example of a worksheet for normally distributed data
© ISO 2020 – All rights reserved 7

---------------------- Page: 12 ----------------------
ISO/FDIS 22514-3:2020(E)

7.4 Produce a probability plot
7.4.1 General
A probability plot should be produced of the data. This may be achieved by using either a software tool
or by using the manual method described in 7.3.4. An example of the output of one software package
can be seen in Figure 4.
Key
X diameter, in mm
Y percent
U upper specification limit
L lower specification limit
Figure 4 — Example of a probability plot for normal distribution data
7.4.2 Ana
...

Questions, Comments and Discussion

Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.