International Standard: Third Edition 2022-08

INTERNATIONAL ISO
STANDARD 3951-1
Third edition
2022-08
Sampling procedures for inspection by

variables —
Part 1:
Specification for single sampling plans
indexed by acceptance quality limit
(AQL) for lot-by-lot inspection for a
single quality characteristic and a
single AQL
Règles d'échantillonnage pour les contrôles par mesures —
Partie 1: Spécification pour les plans d’échantillonnage simples

indexés d’après un niveau de qualité acceptable (NQA) pour un
contrôle lot par lot pour une caractéristique qualité unique et
un NQA unique
Reference number
ISO 3951-1:2022(E)
© ISO 2022
Licensed to Ameco medical industry user. ANSI store order # X_851262. Downloaded 11/29/2022. Single user license only. Copying and networking prohibited.
ISO 3951-1:2022(E)
COPYRIGHT PROTECTED DOCUMENT

© ISO 2022
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: [email protected]
Website: www.iso.org
Published in Switzerland
ii © ISO 2022 – All rights reserved

ISO 3951-1:2022(E)
Contents Page
Foreword..........................................................................................................................................................................................................................................v
Introduction .............................................................................................................................................................................................................................. vi
1 Scope ................................................................................................................................................................................................................................. 1
2 Normative references ..................................................................................................................................................................................... 1
3 Terms and definitions .................................................................................................................................................................................... 2
4 Symbols .......................................................................................................................................................................................................................... 6
5 Choice of a sampling plan ........................................................................................................................................................................... 7
5.1 Choice between variables and attributes ....................................................................................................................... 7
5.2 General ........................................................................................................................................................................................................... 8
5.3 Choice between the s-method and σ-method .............................................................................................................. 8
5.4 Choice of inspection level and AQL ....................................................................................................................................... 9
6 Standard procedures for the s-method..................................................................................................................................... 10
6.1 General ........................................................................................................................................................................................................ 10
6.2 Single specification limits.......................................................................................................................................................... 10
6.3 Double specification limits ....................................................................................................................................................... 11
7 Standard procedures for the σ-method.................................................................................................................................... 11
7.1 General ........................................................................................................................................................................................................ 11
7.2 Single specification limits.......................................................................................................................................................... 11
7.3 Double specification limits ....................................................................................................................................................... 12
8 The p*-method ..................................................................................................................................................................................................... 12
9 Switching between inspection severities ............................................................................................................................... 13
9.1 Rules for switching between inspection severities ............................................................................................ 14
9.2 Records for switching between inspection severities ..................................................................................... 15
10 Relation to ISO 2859-1 ................................................................................................................................................................................. 15
10.1 Similarities .............................................................................................................................................................................................. 15
10.2 Differences .............................................................................................................................................................................................. 16
11 Allowing for measurement uncertainty .................................................................................................................................. 17
12 Normality, data transformations and outliers ................................................................................................................. 17
12.1 Normality ................................................................................................................................................................................................. 17
12.2 Data transformations .................................................................................................................................................................... 17
12.3 Outliers ....................................................................................................................................................................................................... 18
13 Monitoring and recording of inspection results ............................................................................................................. 18
13.1 Monitoring of inspection results ......................................................................................................................................... 18
13.2 Process capability and performance assessment................................................................................................. 18
13.3 Monitoring of process parameters .................................................................................................................................... 18
14 Tables ............................................................................................................................................................................................................................ 19
14.1 Form k for single sampling plans: s-method.............................................................................................................. 19
14.2 Form k for single sampling plans: σ-method............................................................................................................. 26
14.3 Form p* single sampling plans .............................................................................................................................................. 33
14.4 Values of fσ for maximum process standard deviation (MPSD)............................................................... 40
14.5 Supplementary acceptance constants for qualifying towards reduced inspection .............. 40
15 Examples ................................................................................................................................................................................................................... 41
15.1 General ........................................................................................................................................................................................................ 41
15.2 Examples for the s-method ....................................................................................................................................................... 41
15.3 Examples for the σ-method ...................................................................................................................................................... 48
15.4 Examples for the p*-method .................................................................................................................................................... 52
Annex A (informative) Procedures for obtaining s and σ........................................................................................................... 55
© ISO 2022 – All rights reserved iii

ISO 3951-1:2022(E)
Annex B (informative) Accommodating measurement variability..................................................................................58

Annex C (informative) Sampling strategies ..............................................................................................................................................63
Annex D (informative) Operating characteristics for the σ–method.............................................................................65
Annex E (informative) Operating characteristic for the s-method – tabulated values for
single sampling plans, normal inspection.............................................................................................................................66
Annex F (informative) Consumer’s risk qualities ............................................................................................................................... 75
Annex G (informative) Producer’s risks........................................................................................................................................................82
Annex H (informative) Construction of acceptance diagrams for double specification limits..........90
Annex I (informative) Use of the underlying software.............................................................................................................. 101
Bibliography......................................................................................................................................................................................................................... 106
iv © ISO 2022 – All rights reserved

ISO 3951-1:2022(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 5, Acceptance sampling.
This third edition cancels and replaces the second edition (ISO 3591-1:2013), which has been technically
revised.
The main changes are as follows:
— procedures have been introduced to accommodate measurement uncertainty;
— many of the sampling plans have been adjusted to improve the match between their operating
characteristic curves and the operating characteristic curves of the corresponding plans for single
sampling by attributes in ISO 2859-1.
A list of all parts in the ISO 3951 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.
© ISO 2022 – All rights reserved v

ISO 3951-1:2022(E)
Introduction
This document specifies an acceptance sampling system of single sampling plans for inspection by
variables. It is indexed in terms of the acceptance quality limit (AQL). A more comprehensive and
technical treatment of the AQL scheme is given in ISO 3951-2. This document is complementary to
ISO 2859-1.
The objectives of the methods laid down in this document are to ensure that lots of acceptable quality
have a high probability of acceptance and that the probability of not accepting inferior lots is as high as
practicable. This is achieved by means of the switching rules, which provide the following:
a) an automatic protection to the consumer (by means of a switch to tightened inspection or
discontinuation of sampling inspection) should a deterioration in quality be detected; and
b) an incentive (at the discretion of the responsible authority) to reduce inspection costs (by means of
a switch to a smaller sample size) should consistently good quality be achieved.
In this document, the acceptance of a lot is implicitly determined from an estimate of the percentage of
nonconforming items in the process, based on a random sample of items from the lot.
This document is intended for application to a continuing series of lots of discrete products all supplied
by one producer using one production process. If there are different producers or production processes,
this document is applied to each one separately.
This document is intended for application to a single quality characteristic that is measurable on
a continuous scale and is normally distributed. For two or more such quality characteristics, see
ISO 3951-2. For information on normality and data transformations, see Clause 12.
It is assumed in the body of this document that measurement error is negligible (see ISO 10576-1:2003).
For information on allowing for measurement error, see Annex B.
For double specification limits, this document covers combined control. For other types of control, refer
to ISO 3951-2.
CAUTION — The procedures in this document are not suitable for application to lots that have
been screened for nonconforming items.
Inspection by variables for nonconforming items, as described in this document, includes several
possible modes, the combination of which leads to a presentation that may appear quite complex to the
user:
— unknown standard deviation, or originally unknown then estimated with fair precision, or known
since the start of inspection;
— a single specification limit, or combined control of double specification limits;
— normal inspection, tightened inspection, or reduced inspection.
The choice of the most suitable variables plan, if one exists, requires experience, judgement, and some
knowledge of both statistics and the product to be inspected. Clause 5 of this document is intended
to help those responsible for specifying sampling plans in making this choice. They suggest the
considerations that should be kept in mind when deciding whether a variables plan is suitable and the
choices to be made when selecting an appropriate standard plan.
The basic definitions and notations are provided in Clauses 3 and 4. The basic operational rules are
contained in Clauses 5 through 9. Clause 10 informs about the relations between this document and
the attributes sampling standard ISO 2859-1. Clauses 11, 12 and 13 provide background on accounting
for measurement uncertainty, the normality assumption, and monitoring of inspection results and
the underlying process. All tables needed for the sampling procedure can be found in Clause 14 and
examples for the s–method and the σ–method for both one and two specification limits can be found in
Clause 15.
vi © ISO 2022 – All rights reserved

ISO 3951-1:2022(E)
Nine annexes are provided. Annex A indicates how the sample standard deviation, s, and the presumed
known value of the process standard deviation, σ, should be determined. Annex B provides procedures
for accommodating measurement uncertainty. Annex C shows five different sampling strategies.
Annex D gives the general formula for the operating characteristic of the σ–method and provides tables
with values of the operating characteristics of single sampling plans with known σ. Annex E gives the
general formula for the operating characteristic of the s–method and provides tables with values of the
operating characteristics of single sampling plans with unknown σ. Annex F provides the statistical
theory underlying the calculation of the consumer’s risk qualities, together with tables showing these
quality levels for normal, tightened, and reduced inspection, as well as for the s–method and σ–method.
Annex G provides similar information for the producer’s risks. Annex H give details of how Acceptance
diagrams for double specification limits are constructed, Annex I shows the use of the underlying
software (R package to support implementation of this document).
© ISO 2022 – All rights reserved vii

INTERNATIONAL STANDARD ISO 3951-1:2022(E)
Sampling procedures for inspection by variables —

Part 1:
Specification for single sampling plans indexed by
acceptance quality limit (AQL) for lot-by-lot inspection for
a single quality characteristic and a single AQL
1 Scope
This document specifies single sampling plans for lot-by-lot inspection under the following conditions:
a) where the inspection procedure is applied to a continuing series of lots of discrete products, all
supplied by one producer using one production process;
b) where only a single quality characteristic, x, of these products is taken into consideration, which is
measurable on a continuous scale;
c) where production is under statistical control and the quality characteristic, x, is distributed
according to a normal distribution or a close approximation to the normal distribution;
d) where a contract or standard defines a lower specification limit, L, an upper specification limit,
U, or both. An item is qualified as conforming if its measured quality characteristic, x, satisfies as
appropriate one of the following inequalities:
1) x ≥ L (i.e. the lower specification limit is not violated);
2) x ≤ U (i.e. the upper specification limit is not violated);
3) x ≥ L and x ≤ U (i.e. neither the lower nor the upper specification limit is violated).
Inequalities 1) and 2) are cases with a single specification limit, and 3) is a case with double specification
limits.
Where double specification limits apply, it is assumed in this document that conformity to both
specification limits is equally important to the integrity of the product. In such cases, it is appropriate
to apply a single AQL to the combined percentage of a product outside the two specification limits. This
is referred to as combined control.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content
constitutes requirements of this document. For dated references, only the edition cited applies. For
undated references, the latest edition of the referenced document (including any amendments) applies.
ISO 2859-1, Sampling procedures for inspection by attributes — Part 1: Sampling schemes indexed by
acceptance quality limit (AQL) for lot-by-lot inspection
ISO 3534-1, Statistics — Vocabulary and symbols — Part 1: General statistical terms and terms used in
probability
ISO 3534-2, Statistics — Vocabulary and symbols — Part 2: Applied statistics
© ISO 2022 – All rights reserved 1

ISO 3951-1:2022(E)
ISO 3951-2, Sampling procedures for inspection by variables — Part 2: General specification for single
sampling plans indexed by acceptance quality limit (AQL) for lot-by-lot inspection of independent quality
characteristics
3 Terms and definitions

For the purposes of this document, the terms and definitions given in ISO 2859-1, ISO 3534-1 and
ISO 3534-2 and the following apply.
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 https://www.electropedia.org/
3.1
inspection by variables
inspection by measuring the magnitude(s) of a characteristic(s) of an item
[SOURCE: ISO 3534-2:2006, 4.1.4]
3.2
sampling inspection
inspection of selected items in the group under consideration
[SOURCE: ISO 3534-2:2006, 4.1.6]
3.3
acceptance sampling inspection
acceptance inspection where the acceptability is determined by means of sampling inspection (3.2)
[SOURCE: ISO 3534-2:2006, 4.1.8]
3.4
acceptance sampling inspection by variables
acceptance sampling inspection (3.3) in which the acceptance of the process is determined statistically
from measurements from inspection by variables (3.1)
[SOURCE: ISO 3534-2:2006, 4.2.11 modified — “on specified quality characteristics of each item in a
sample from a lot” has been replaced by “from inspection by variables”]
3.5
process fraction nonconforming
rate at which nonconforming items are generated by a process
Note 1 to entry: It is expressed as a proportion.
3.6
acceptance quality limit
AQL
(acceptance sampling) worst tolerable quality level (3.7)
Note 1 to entry: This concept only applies when an acceptance sampling scheme with rules for switching and for
discontinuation, such as ISO 2859-1 and ISO 3951, is used.
Note 2 to entry: See 5.4.
[SOURCE: ISO 3534-2:2006, 4.6.15 modified — Notes 2 to 4 to entry have been deleted and Note 1 to
entry has been added.]
2 © ISO 2022 – All rights reserved

ISO 3951-1:2022(E)
3.7
quality level
quality expressed as the fraction nonconforming
[SOURCE: ISO 3534-2:2006, 4.6.16 modified — “or rate of number of nonconformities” has been
removed”.]
3.8
limiting quality
LQ
quality level (3.7), when a lot is considered in isolation, which, for the purposes of acceptance sampling
inspection (3.3), is limited to a low probability of acceptance
Note 1 to entry: In this document: 10 %.
[SOURCE: ISO 3534-2:2006, 4.6.13, modified — Note 1 to entry has been added.]
3.9
nonconformity
non-fulfilment of a requirement
[SOURCE: ISO 3534-2:2006, 3.1.11]
3.10
nonconforming unit
unit with one or more nonconformities
[SOURCE: ISO 3534-2:2006, 1.2.15]
3.11
p*-method acceptance sampling plan
acceptance sampling (3.3) plan by variables where the the estimated fraction nonconforming p̂ is
compared to the maximum acceptable value p*.
Note 1 to entry: The method is applicable to both the s-method (3.12) and the σ-method (3.13) and gives equivalent
results. It has the advantage that it deals directly with the fraction nonconforming.
3.12
s–method acceptance sampling plan
acceptance sampling (3.3) plan by variables using the sample standard deviation
Note 1 to entry: See Clause 6.
[SOURCE: ISO 3534-2:2006, 4.3.10 modified — “s method” has been replaced by “s-method” and
“acceptance sampling plan” has been added]
3.13
σ–method acceptance sampling plan
acceptance sampling (3.3) plan by variables using the presumed value of the process standard deviation
[SOURCE: ISO 3534-2:2006, 4.3.9 modified — “sigma method” has been replaced by “σ-method” and
“acceptance sampling plan” has been added]
3.14
specification limit
conformance boundary specified for a characteristic
[SOURCE: ISO 3534-2:2006, 3.1.3, modified — "limiting value stated" has been replaced with
"conformance boundary specified".]

ISO 3951-1:2022(E)
3.15
lower specification limit
L
specification limit (3.14) that defines the lower conformance boundary
[SOURCE: ISO 3534 2:2006, 3.1.5, modified — "limiting value" has been replaced with "conformance
boundary".]
3.16
upper specification limit
U
specification limit (3.14) that defines the upper conformance boundary
[SOURCE: ISO 3534-2:2006, 3.1.4, modified — "limiting value" has been replaced with "conformance
boundary".]
3.17
combined control
requirement when both upper and lower limits are specified for the quality characteristic and an AQL
(3.6) that applies to the combined percent nonconforming beyond the two limits is given
Note 2 to entry: The use of combined control implies that nonconformity beyond either specification limit (3.14) is
believed to be of equal, or at least roughly equal, importance to the lack of integrity of the product.
3.18
form k acceptance constant
k
constant depending on the specified value of the acceptance quality limit (3.6) and the sample size, used
in the criteria for accepting the lot in an acceptance sampling (3.3) plan by variables
Note 1 to entry: See Clause 6 and Clause 7.
[SOURCE: ISO 3534-2:2006, 4.4.4, modified — "acceptability constant" has been replaced with
"acceptance constant". The original Note 1 to entry has been deleted, a new Note 1 to entry has been
added.]
3.19
form p* acceptance constant
p*
constant depending on the specified value of the acceptance quality limit (3.6) and the sample size, used
in the criteria for accepting the lot in an acceptance sampling (3.3) plan by variables
[SOURCE: ISO 3534-2:2006, 4.4.4, modified — "acceptability constant" has been replaced with
"acceptance constant". The original Note 1 to entry has been deleted, a new Note 1 to entry has been
added.]
3.20
quality statistic
Q
function of the specification limit (3.14), the sample mean and the sample or process standard deviation,
used in assessing the acceptance of a lot
Note 1 to entry: For the case of a single specification limit, the lot can be accepted or rejected on the result of
comparing Q with the form k acceptance constant (3.18).
Note 2 to entry: See Clause 6 and Clause 7.
[SOURCE: ISO 3534-2:2006, 4.4.9, modified — In the Note 1 to entry, "may be sentenced" has been
replaced with " can be accepted or rejected". Note 2 to entry has been added.]

ISO 3951-1:2022(E)
3.21
lower quality statistic
QL
function of the lower specification limit (3.15), the sample mean, and the sample or process standard
deviation
Note 1 to entry: For a single lower specification limit, the lot can be accepted or rejected on the result of comparing
QL with the form k acceptance constant (3.18).
Note 2 to entry: See Clause 4, Clause 6, and Clause 7.
3.22
upper quality statistic
QU
function of the upper specification limit (3.16), the sample mean, and the sample or process standard
deviation
Note 1 to entry: For a single upper specification limit, the lot can be accepted or rejected on the result of
comparing QU with the form k acceptance constant (3.18).
Note 2 to entry: See Clause 4, Clause 6, and Clause 7.
3.23
maximum process standard deviation
MPSD
σmax
largest process standard deviation for a given sample size code letter and AQL (3.6) for which it is
possible to satisfy the acceptance criterion for a double specification limit under all inspection severities
(i.e. normal, tightened and reduced) when the process variability is known
Note 1 to entry: The MPSD depends on whether the double specification limits are combined, separate or complex,
but does not depend on the inspection severity.
3.24
switching rule
instruction within an acceptance sampling (3.3) scheme for changing from one acceptance sampling
plan to another of greater or lesser severity based on demonstrated quality history
Note 1 to entry: Normal, tightened, or reduced inspection or discontinuation of inspection are examples of
‘severity of sampling’.
3.25
measurement
set of operations to determine the value of some quantity
[SOURCE: ISO 3534-2:2006, 3.2.1, modified — "having the object of determining a value of a quantity"
has been replaced with " to determine the value of some quantity". Notes 1 and 2 to entry have been
deleted.]

ISO 3951-1:2022(E)
4 Symbols
fσ factor that relates the maximum process standard deviation to the difference between
U and L (see 14.5)
FBETA(α , β ) ( x ) the distribution of the standard beta distribution with parameters α and β . In this
document α = β = n 2 − 1 throughout.
Ft (ν ,δ ) ( x ) the distribution function of the non-central t-distribution with ν degrees of freedom
and non-centrality parameter δ
Kp the upper p-quantile of the standardized normal distribution i.e. x such that
1 − Φ ( x ) = p , which corresponds to the process fraction nonconforming p
k form k acceptance constant for use with a single quality characteristic and a single
specification limit (see 14.1 for the s–method acceptance sampling plan or 14.2 for the
σ–method acceptance sampling plan)
L lower specification limit (as a subscript to a variable, it denotes its value at L)
M unknown process mean
N lot size (number of items in a lot)
n sample size (number of items in a sample)

p̂ estimate of the process fraction nonconforming
p̂L estimate of the process fraction nonconforming below the lower specification limit
p̂U estimate of the process fraction nonconforming above the upper specification limit
form p* acceptance constant i.e. the maximum acceptable value for the estimate of the
p*
process fraction nonconforming
Pa probability of acceptance
Φ(x ) the distribution function of the standardized normal distribution
Q quality statistic
QL lower quality statistic

NOTE QL is defined as ( x − L ) s when the process standard deviation is unknown, and
as ( x − L ) σ when it is presumed to be known.
QU upper quality statistic

NOTE QU is defined as (U − x ) s when the process standard deviation is unknown, and
as (U − x ) σ when it is presumed to be known.
S sample standard deviation of the measured values of the quality characteristic (also
an estimate of the standard deviation of the process), i.e.
n
∑( x j − x )2
j =1
s=
n−1
(See Annex A.)

ISO 3951-1:2022(E)
σ process standard deviation

NOTE σ 2 , the square of the process standard deviation, is known as the process
variance.
σ max maximum process standard deviation (MPSD)
σ root weighted root mean square of s
U upper specification limit (as a suffix to a variable, it denotes its value at U)
xj measured value of the quality characteristic for the jth item of the sample
x the arithmetic mean of the measured values of the quality characteristic in the sam-
ple, i.e.
n
∑xj
j =1
x=
n
5 Choice of a sampling plan
5.1 Choice between variables and attributes

The first question to consider is whether it is desirable to inspect variables rather than attributes. The
following points should be taken into account.
a) In terms of economics, it is necessary to compare the total cost of the relatively simple inspection
of a larger number of items by means of an attributes scheme with the generally more elaborate
procedure required by a variables scheme, which is usually more time-consuming and costly per
item.
b) In terms of the knowledge gained, the advantage lies with inspection by variables as the information
obtained indicates more precisely the quality of the product. Therefore, earlier warning can be
given if the quality is slipping.
c) An attributes scheme can be more readily understood and accepted. For example, two phenomena
of variables sampling are difficult to understand:
1) lots containing 100 % conforming items can be rejected, occasionally even with a high
probability (see example 3 in 15.1);
2) there are cases of larger AQL and smaller lot size where lots containing 100 % nonconforming
items are accepted.
d) From a comparison of the size of the samples required for the same AQL from standard plans
for inspection by attributes, such as from ISO 2859-1, and the standard plans in this document,
the smallest samples are generally required by the σ–method (used when the process standard
deviation is presumed to be known). The sample sizes for the s-method (used when the process
standard deviation is presumed to be unknown) are larger than for the σ-method but are, in
general, substantially smaller than for sampling by attributes.
e) Inspection by variables is particularly appropriate in conjunction with the use of control charts for
variables.
f) Variables sampling has a substantial advantage when the inspection process is expensive, for
example, in the case of destructive testing.
g) A variables scheme becomes relatively more complicated to operate as the number of measurements
to be taken on each item increases.

ISO 3951-1:2022(E)
NOTE For two or more quality characteristics, that need to be treated together using a single AQL, this
document does not apply. See ISO 3951-2 for details.
5.2 General
The sampling schemes found in this document, with their rules for switching and for discontinuation of
sampling inspection, are designed to encourage suppliers to keep the process fraction nonconforming
consistently better than the respective AQLs.
Otherwise, there is a high risk that the inspection severity be switched to tightened inspection, under
which the criteria for lot acceptance become more demanding. Once on tightened inspection, unless
action is taken to improve the process, it is very likely that the rule requiring discontinuation of
sampling inspection be invoked pending such improvement.
The following procedures shall be followed before starting an inspection by variables:
a) check that an AQL has been designated, and that it is one of the preferred AQLs for use with this
document (see Clause 5). If it is not, then the sampling tables of this document are not applicable;
NOTE If an AQL other than a preferred AQL is designated, then this document is only applicable under
the advice of a statistical expert.
b) check that the inspection level to be used has been designated. If none has been given, the inspection
level II shall be used (see 5.4);
c) determine the quality characteristic x and an upper limit U and/or a lower limit L for x;
d) for a quality characteristic with double specification limits, check that nonconformities beyond
each limit are of equal importance. If this is not the case, refer to ISO 3951-2;
NOTE If the lot size, the inspection level, U, L or the AQL change during the process, the user might need
to stop and begin again with 5.1.
e) check whether the s–method (Clause 6) is to be used initially or whether the standard deviation is
stable and known, in which case the σ–method (Clause 7) should be used (see 5.3);
f) for the σ–method and a quality characteristic with double specification limits, a process capability
study in the following sense should be done:
1) enter Table 11 in 14.4 with the AQL to determine the value of the factor fσ;
2) calculate the maximum allowable value of the process standard deviation using the formula
σ max = (U − L ) fσ ;
3) If σ exceeds σmax, the process is not capable and sampling inspection is pointless until it is
demonstrated that the process variability has been adequately reduced.
g) obtain the sample size code letter with the inspection limit (normally level II) and the lot size from
Table 1.
5.3 Choice between the s-method and σ-method

If the inspection by variables method as proposed in this document is chosen, the decision shall be
made whether to use the s–method or the σ–method. The σ–method is the more economical in terms of
sample size, but before this method can be applied, it is necessary to have a reliable value of σ, usually
obtained from previous process analyses.
In the case that no reliable assumptions on the value of σ can be made, it shall be necessary to begin
with the s–method, but subject to the agreement of the responsible authority and provided the quality
remains satisfactory, the operational rules of this document permit a change to the σ–method and
consequently the use of a smaller sample size.

ISO 3951-1:2022(E)
If the variability is under control, a reliable estimation for the process standard deviation can be made
from the observed data, in which case it may be economical to change to the σ–method. The size of the
sample is generally smaller and the acceptance criterion is simpler using the σ–method (see Clause 7).
On the other hand, it shall still be necessary to calculate the sample standard deviation, s, for record
and continued monitoring purposes of the product quality and to keep the pattern or control charts up
to date (see Clause 13). Methods of determining s and σ are given in Annex A.
In case of using the σ–method, it is recommended to perform, periodically and systematically,
comparisons of observed data, i.e. observed process standard deviations with the assumption made on
σ. In the framework of continued monitoring, this can be done in several ways:
— use of descriptive, graphical methods, e.g. pattern charts;
— use of statistical control charts;
— descriptive numerical analysis; or
— statistical tests, e.g. Bartlett’s test for the comparison of variances.
If based on these analyses there is doubt in the reliability of the assumptions made on σ, it can be
necessary to switch to the s-method.
5.4 Choice of inspection level and AQL

The AQL is the quality level that is the worst tolerable nonconforming process fraction when a
continuing series of lots is submitted for acceptance sampling. Although individual lots determined to
be of “bad” quality may be accepted by the AQL with fairly high probability, the designation of an AQL
does not suggest that this is a desirable quality level.
The AQL, together with the sample size code letter, is used to index the sampling plans in this document.
For a standard sampling plan, the inspection level, in conjunction with the size of the lots and the AQL,
determines the size of the sample to be taken and governs the severity of the inspection. The appropriate
operating characteristic from Annex E shows the extent of the risk that is involved in such a plan.
The 16 AQLs given in this document, ranging in value from 0,01 % to 10 % nonconforming, are
described as “preferred AQLs”. They are only preferred in the sense that they are the AQL values used
in the tabulations and charts. It follows that, if for any product or service, an AQL other than a preferred
AQL is designated, then this document is only applicable under the advice of a statistical expert.
The AQL to be used shall be designated in the product specification or in the contract, or by the
responsible authority. Where both upper and lower specification limits are given, this document
addresses only the case of an overall AQL applying to the combined percent nonconforming beyond the
two limits; this is known as “combined control.” (See ISO 3951-2 for “separate” and “complex” control of
double specification limits.)
From the definition of the AQL, it follows that the desired protection can only be ensured when a
continuing series of lots is provided for inspection.
The designation of an AQL shall not imply that the supplier has the right to supply knowingly any
nonconforming product.
The choice of inspection level and AQL is governed by a number of factors but is mainly a balance
between the total cost of inspection and the consequences of nonconforming items passing into service.
Three inspection levels, I, II, and III, are given in Table 1 for general use. Inspection level II shall be used
unless special circumstances indicate that another level is more appropriate. Level I may be used when
less discrimination is needed, and level III when greater discrimination is required. Four additional
special levels, S-1, S-2, S-3, and S-4, are also given in Table 1 and may be used when relatively small
sample sizes are necessary and larger sampling risks can be tolerated, or when very high process
capability has been demonstrated.

ISO 3951-1:2022(E)
Table 1 — Sample size code letters and inspection levels

Lot or batch size Special inspection levels General inspection levels
S-1 S-2 S-3 S-4 I II III
2 to 8 B B B B B B B
9 to 15 B B B B B B C
16 to 25 B B B B B C D
26 to 50 B B B C C D E
51 to 90 B B C C C E F
91 to 150 B B C D D F G
151 to 280 B C D E E G H
281 to 500 B C D E F H J
501 to 1 200 C C E F G J K
1 201 to 3 200 C D E G H K L
3 201 to 10 000 C D F G J L M
10 001 to 35 000 C D F H K M N
35 001 to 150 000 D E G J L N P
150 000 to 500 000 D E G J M P Q
500 000 and over D E H K N Q R
NOTE The sample size code letters and inspection levels in this document correspond to those given in ISO 2859-1 and
ISO 3951-2.
6 Standard procedures for the s-method
6.1 General
The s-method shall be used if information about the process standard deviation is missing or
unreliable. Using the s-method, the standard deviation is estimated directly from each sample. As soon
as the conditions for the use of the σ-method are warranted, one may switch from the s-method to the
σ-method (see 5.3).
6.2 Single specification limits

Before starting the inspection by variables, see Clause 5.
The procedure for a single specification limit is as follows.
a) Enter Table 2, Table 3, or Table 4, as appropriate, with the code letter and the AQL to obtain the
sample size, n, and the acceptance constant, k.
b) Take a random sample of size n, measure the characteristic x in each item, and then calculate x , the
sample mean, and s, the sample standard deviation (see Annex A).
NOTE Some sampling strategies are provided in Annex C.
c) Calculate the quality statistic QU = (U − x )/ s or QL = ( x − L )/ s and compare it with the acceptance

constant k. The lot is accepted if QU ≥ k or QL ≥ k ; or rejected if QU < k or QL < k .
d) Proceed with Clause 9.
For examples of single lower and upper specification limits using the s-method, see Example 1 and
Example 2 in 15.1.

ISO 3951-1:2022(E)
6.3 Double specification limits

The procedure for double specification limits for the s-method is as follows.
a) Enter Table 2, Table 3, or Table 4, as appropriate, with the code letter and the AQL to obtain the
b) Take a random sample of size, n, measure the characteristic x in each item, and then calculate x ,
the sample mean, and s, the sample standard deviation (see Annex A).
NOTE Some sampling strategies are provided in Annex C.
c) Determination of acceptance.
1) If x < L or x > U , reject the lot; or continue with the next step.
2) Plot ( s , x ) on the acceptance diagram, which can be obtained using the accCurve function in
the underlying software (see I.5).
s x −L
NOTE The standardized values ( s S , x S ) where sS = and x S = may be used in which case
U −L U −L
( sS , xS ) is plotted on the standardized acceptance diagram, which can be obtained using the accCurve
function in the underlying software without providing the standardized lower and upper specification
limits L = 0, U = 1 (see I.5).
3). If the plotted point is outside the acceptance region the lot is rejected; otherwise the lot is
accepted.
For examples of combined control of double specification limits using the s-method, see Example 3 to
Example 6 in 15.2.
7 Standard procedures for the σ-method
7.1 General
The σ–method shall only be used when there is valid evidence that the standard deviation σ of the
process can be considered constant with a known value.
7.2 Single specification limits

The procedure for a single specification limit is as follows.
a) Enter Table 5, Table 6, or Table 7, as appropriate, with the code letter and the AQL and obtain the
NOTE 1 Some sampling strategies are provided in Annex C.
NOTE 2 The sample standard deviation is needed for the records (see Clause 13).
c) Calculate the quality statistic QU = (U − x ) σ or QL = ( x − L ) σ and compare it with the acceptance

constant k. The lot is accepted if QU ≥ k or QL ≥ k ; or rejected if QU < k or QL < k .

ISO 3951-1:2022(E)
For examples of single lower and upper specification limits using the σ–method, see Example 1 and
Example 2 in 15.2.
7.3 Double specification limits

The procedure for double specification limits is as follows.
a) Enter Table 2, Table 3, or Table 4, as appropriate, with the code letter and the AQL and obtain the
NOTE 2 The sample standard deviation is needed for the records (see Clause 13).
c) Determination of acceptance.
1) If x < L or x > U , reject the lot; or continue with the next step.
2) Calculate the quality statistics QL = ( x − L ) / σ and QU = (U − x ) / σ . If QL < k or QU < k, reject the
lot; or continue with the next step.
3) If σ ≤ 0 , 75 σ max accept the lot; otherwise continue with the next step.
4) If neither QL or QU are close to k accept the lot; otherwise use the p*-method in Clause 8.
For an example of combined control of double specification limits using the σ–method, see Example 3 in
15.2.
8 The p*-method
The p*–method acceptance sampling plan is an alternative to the standard procedures in Clause 6 and
Clause 7. For the application of this method, distribution functions are calculated.
The procedure for the p*–method for a single limit and double limits is as follows.
a) Enter Tables 8, 9 or 10 in 14.3 with the code letter and the AQL, to obtain the sample size, where
n=ns is the sample size for the s-method, and n = nσ is the sample size for the σ–method, and to
obtain the acceptance constant, i.e. the maximum fraction nonconforming, p*.
NOTE 2 In the case of known standard deviation, σ, the sample standard deviation is needed for the
records (see Clause 13).
c) Estimated fraction nonconforming:
1) unknown standard deviation: For a lower specification limit, calculate p̂L , for an upper limit,
calculate p̂U , and for two specification limits, calculate both,

ISO 3951-1:2022(E)
  1 1 x − L n     1  n   
pˆL = F n n   max 0, −   = F n n   max 0, 1 − QL   
BETA −1, −1   
 2 2 s n − 1 
 BETA − − 
 2  n − 1   
2 
 1, 1  
2 2  2
  1 1 U − x n     1  n   
pÛ = F  max 0, −   = F  max 0, 1 − Q 
BETA −1, −1  
n n  n n  U
 2 2 s n − 1  
 BETA −1, −1    2 
 n − 1   
2 2  2 2 
2) known standard deviation: For a lower specification limit, calculate p̂L , for an upper limit,
calculate p̂U , and for two specification limits, calculate both,
L−x n   n 
p̂L = Φ   =Φ  −QL 
 σ n − 1  n − 1 
 
 x −U n   n 
p̂U = Φ   =Φ  −QU 
 σ n − 1  n − 1 
 
NOTE The fraction nonconforming can be calculated with the R package, see Annex I.
d) Acceptance criteria:
1) single limit: A lot shall be accepted if pˆL ≤ p * for a lower limit, and if pÛ ≤ p * for an upper
limit; if pˆL > p * or pÛ > p * respectively, the lot is rejected;
2) double limits: A lot shall be accepted if pˆL + pÛ ≤ p * ; if pˆL + pÛ > p * , the lot is rejected;
e) Proceed with Clause 9.
For examples of single specification limits see Example 1 and Example 3 in 15.3. For examples of
combined control of double specification limits see Example 2 and Example 4 in 15.3.
9 Switching between inspection severities

Switching rules discourage the producer from operating at a quality level that is worse than the AQL.
This document prescribes a switch to tightened inspection when inspection results indicate that the
AQL is being exceeded. It further prescribes a discontinuation of sampling inspection altogether if
tightened inspection fails to stimulate the producer into rapidly improving this production process.
Tightened inspection and the discontinuation rule are integral and, therefore, obligatory procedures of
this document if the protection implied by the AQL is to be maintained.
This document also provides the possibility of switching to reduced inspection when inspection results
indicate that the quality level is stable and reliable at a level better than the AQL. This practice is,
however, optional (at the discretion of the responsible authority).
When it has been necessary to discontinue acceptance sampling inspection, inspection as outlined by
this document shall not be resumed until action has been taken by the producer to improve the quality
of the submitted product.
The sampling system specified by this document can only work efficiently if the switching rules in 9.1
are obeyed (see also Figure 1), and therefore records of some values of the inspection plan need to be
kept (see 9.2).

ISO 3951-1:2022(E)
Figure 1 — Outline of the switching rules
9.1 Rules for switching between inspection severities

Normal inspection is used at the start of inspection (unless otherwise designated). It shall be used
during inspection until tightened inspection becomes necessary or reduced inspection is allowed. For
further information about the three inspection severities see 5.4.
This document prescribes a switch to tightened inspection when inspection results indicate that the
AQL is being exceeded, and also provides the possibility of switching to reduced inspection when
inspection results indicate that the quality level is stable and reliable at a level better than the AQL. The
switch to reduced inspection is, however, optional (at the discretion of the responsible authority).
The standard switching rules are as follows.
a) Within normal inspection:
1) when two lots are rejected within any five or fewer successive lots, tightened inspection shall
be instituted;
NOTE Tightened inspection is generally achieved by increasing the value of the acceptance constant k.
The values are tabulated in Table 3 for the s–method and in Table 6 for the σ–method.
2) after 10 successive lots have been accepted, reduced inspection may be instituted, provided
that
i) these lots would have been accepted if the AQL had been one step tighter (e.g. 0,65 %
instead of 1,0 %),
NOTE 1 To determine this, it is useful to keep a record of the values of QU or QL , respectively.
Next, follow the procedures in Clause 6 or Clause 7 respectively, with the sample size code letter, the
new AQL, the values of QU or QL and the last 10 values of x and s.
NOTE 2 If a value of k for the tighter AQL is not given in Table 2 (s–method) or Table 6 (σ–
method), refer to the supplementary acceptance constants provided in Table 13.
ii) production is in statistical control (see 5.3), and

iii) reduced inspection is considered desirable by the responsible authority.

ISO 3951-1:2022(E)
NOTE 1 When the previous 10 lots have been accepted under original inspection, reduced inspection can
be instituted without the condition that these lots would have been accepted if the AQL had been one step
tighter, subject to the approval of the responsible authority.
NOTE 2 Reduced inspection is conducted on a much smaller sample than normal inspection and the value
of the acceptance constant is also decreased. The values of n and k for reduced inspection are given in Table 4
for the s–method and Table 7 for the σ–method.
b) Within reduced inspection: Normal inspection shall be reinstated if at least one of the following
occur:
1) a lot is rejected;
2) production is no longer in statistical control or is delayed; or
3) reduced inspection is no longer considered desirable by the responsible authority.
c) Within tightened inspection:
1) when five successive lots have been accepted, normal inspection shall be reinstated;
2) if the cumulative number of lots rejected in a sequence of consecutive lots on original tightened
inspection reaches five, the acceptance procedures of this document shall be discontinued.
Inspection under the provisions of this document shall not be resumed until action has been taken by
the supplier to improve the quality of the submitted product or service and the responsible authority
has agreed that this action is likely to be effective. Tightened inspection shall then be used again.
9.2 Records for switching between inspection severities

As a variables sampling inspection plan can only operate efficiently if records are kept, this is an
important part of the inspection. For further information about the advantages of keeping records, and
control charts especially, see Clause 13. In particular, the following values should be kept:
a) the values of x and s , the lot size, the inspection level, the inspection severity, the code letter, U, L
and the AQL (especially when they changed since the last inspection);
When one of the prescribed values changes, apart from the inspection severity, one needs to start again
with Clause 5.
b) whether the lot was accepted;
Any lot rejected by the sampling plan shall not be resubmitted either in whole or in part without the
permission of the responsible authority.
c) if the lot has been accepted; the values of QU and QL , if they have been evaluated.
NOTE These values are important for switching to reduced inspection.
10 Relation to ISO 2859-1
10.1 Similarities
The similarities are as follows.
a) This document is complementary to ISO 2859-1; the two documents share a common principle and,
as far as possible, their procedures and vocabulary are the same.
b) Both use the AQL to index the sampling plans, and the preferred values used in this document are
identical to those given for percent nonconforming in ISO 2859-1 (i.e. from 0,01 % to 10 %).

ISO 3951-1:2022(E)
c) In both this document and in ISO 2859-1, lot size and inspection level (inspection level II) determine
a sample size code letter. General tables give the sample size to be taken and the acceptance
criterion, indexed by the sample size code letter and the AQL. Separate tables are given for the s–
method and σ–method, and for normal, tightened, and reduced inspection.
d) The switching rules are essentially equivalent.
10.2 Differences
a) Determination of acceptance. In ISO 2859-1, acceptance for an attributes sampling plan for
percent nonconforming is determined by the number of nonconforming items found in the sample.
Acceptance for a plan for inspection by variables is based on the distance of the estimated value
of the process mean from the specification limit(s) in terms of the estimated or presumed value
of the process standard deviation. In this document, two methods are considered: the s–method,
for use when the process standard deviation, σ , is unknown, and the σ–method, for use when σ is
presumed to be known.
b) Normality. In ISO 2859-1, there is no requirement relating to the distribution of the characteristics.
However, in this document, it is necessary for the efficient operation of the plans that the
measurements be distributed according to a normal distribution (see 12.1), or a close approximation
to a normal distribution, or that the data is appropriately transformed to a normal distribution (see
12.2).
c) Operating characteristic curves (OC curves). The OC curves of the variables plans in this
document are not identical to those of the corresponding attributes plans in ISO 2859-1. The curves
for s-method plans have been matched by minimizing the area between the curves representing the
squares of the OC values, which gives greater emphasis to the match at the top of the OC curves. In
most cases, the resulting match between the OC curves is so close that, for most practical purposes,
the attributes and variables of OC curves may be considered to be identical.
NOTE The σ-method plans are derived by minimizing the area between the squared OC functions
subject to keeping the same Form p* acceptance constant as for the corresponding s-method plan, i.e. only
the sample size is open to choice, so the match is, in general, less perfect.
d) Producer’s risk. For process quality precisely at the AQL, the producer’s risk that a lot not be
accepted tends to decrease with increments in sample size coupled with decrements in AQL, as
visible down the ‘right’ diagonals of the master tables, i.e., the tables in Clause 14, running from
top right to bottom left. The progressions of probabilities are similar, but not identical, to those in
ISO 2859-1.
NOTE The producer’s risks of the plans are given in Annex G.
e) Sample sizes. The variables sample sizes for given combinations of sample size code letter and AQL
are usually smaller than the corresponding attributes sample sizes. This is particularly true for the
σ-method. Moreover, due to the method by which the variables plans were derived, their sample
sizes vary over AQL for a given sample size code letter.
f) Double sampling plans. Double sampling plans by variables are presented separately in
ISO 3951-3.
g) Multiple sampling plans. There are no multiple sampling plans by variables in any part of ISO 3951
series.
h) Average outgoing quality limit (AOQL). The AOQL concept is mainly of value when 100 %
inspection and rectification is feasible for non-accepted lots. It follows that the AOQL concept
cannot be applied under destructive or expensive testing. As variables plans shall generally be
used under these circumstances, no tables of AOQL have been included in this document.

ISO 3951-1:2022(E)
11 Allowing for measurement uncertainty

The master tables, i.e. the tables in Clause 14, are based on the assumption that the quality
characteristic, x, of the items in the lots is normally distributed with unknown process mean, µ, and
either known or unknown process standard deviation, σ. The assumption is also made that x can be
measured without measurement error, i.e. that measurement of an item with the true value, xi, results
in the value xi. However, the tables in Clause 14 can also be used, with appropriate adjustments, in the
result of measurement error.
Ideally, as a prerequisite for the application of the proposed sampling schemes, the measurement
uncertainty, i.e. the measurement standard deviation, is determined by a structured measurement
analysis followed by a measurement system capability assessment. If the measurement standard
deviation is not higher than 10 % of the process standard deviation, the measurement system can be
considered as capable and no further adjustments on sample sizes are required.
For measurement standard deviation higher than 10 % of the process standard deviation, the sample
size needs to be increased, although the acceptance constant remains the same. Moreover, if neither
the measurement standard deviation nor the process standard deviation is known, more than one
measurement needs to be made on each sampled item and the total variability of the measurements
needs to be separated into the components.
Details are provided in Annex B.
12 Normality, data transformations and outliers
12.1 Normality
The assumption of a normal distribution for the inspected quality characteristic x is the fundamental
requirement for the application of the proposed variables sampling plans. Therefore, it is essential to
assess the data normality before applying the sampling plans, which can be done by several approaches:
a) assessment based on practical, engineering experience and expertise;
b) assessment based on graphical, descriptive methods, such as histograms, quantile/quantile or
normal probability plots (see ISO 5479); or
c) statistical tests for departure from normality, such as those given in ISO 5479 or in ISO 5725-2:2019,
Clause 2.
12.2 Data transformations

For non-normal distributed data, the first approach is usually to try to find a suitable transformation
f of the observed data x, so that Z = f(x) now follows an (approximate) normal distribution. The most
commonly used data transformations are (see ISO 16269):
— power transformations;
— Box-Cox transformations;
— Johnson transformations;
— transformations using Pearson functions.
For the transformed data, the goodness of fit to a normal distribution can be assessed using normal
probability plots and/or statistical tests for normality. For subsequent capability analyses and
assessments, the targets and specification limits of the quality characteristic x have to be transformed
in the same way as the observed process data.

ISO 3951-1:2022(E)
12.3 Outliers
An outlier (or an outlying observation) is one that appears to deviate markedly from other observations
in the sample in which it occurs. A single outlier, even when it lies within specification limits, increase
variability, changes the mean, and may consequently lead to non-acceptance of the lot (see ISO 16269-4).
Generally, outliers can be a result of
a) (unexpectedly high) natural process variability, and/or
b) presence of special causes not representing the controlled and capable process status.
For case b), it might be considered to exclude the outlying observations for the quality assessment,
whereas for case a), these values have to be considered in the analysis. Nevertheless, a sensitivity
analysis including and excluding the outliers can help to assess the impact of these values on the
decision on lot acceptance.
When outliers are detected, the disposition of the lot should be a matter for the responsible authority
and may involve negotiation between the vendor and the vendee.
13 Monitoring and recording of inspection results
13.1 Monitoring of inspection results

The results of the quality assessment using the inspection by variables sampling plans proposed in
the corresponding tables should be monitored continuously. The monitoring results should be analysed
to detect changes or trends in the quality level of the product, so that warning can be given before an
unacceptable quality level is reached. The adequate data recording of quality assessment results is a
necessary prerequisite for monitoring.
Independent of the performed inspection, continuous monitoring should serve to assess the inherent
product and process quality level of the production. Monitoring allows the user of this document to
obtain a long-term view of the production quality in addition to the information on lot-by-lot decisions.
13.2 Process capability and performance assessment

Before implementing this document, it is recommended to assess the process capability, i.e. the ability
of the production process to fulfil predefined quality requirements. These assessments shall be done
in a scientifically planned manner or performance study. Several methods for estimating process
capability or performance depending on the assumed process data distribution model are provided in
ISO 22514-2.
13.3 Monitoring of process parameters

For variable inspection procedures, monitoring of product and process quality mainly focuses on
process mean and process standard deviation. Whatever method used, s–method or σ–method, records
should be kept of the values of x and s, preferably in the form of pattern or control charts.
An introduction to the use of control charts is given in ISO 7870, and special control chart applications
for assessing process means and standard deviations can be found for standard Shewhart control
charts in ISO 7870-2 or ISO 8258 and, for more complex CUSUM (cumulated sum) charts, in ISO 7870-4.
The process standard deviation should be monitored particularly when using the σ–method in order
to verify that the values of s obtained from the samples fall within the limits of the prescribed value
of σ. With the s–method, the standard deviation shall be monitored to indicate whether a switch to
the σ-method is possible. In particular, the values of s or σ, respectively, and whether they exceed their
control limit should be kept.

ISO 3951-1:2022(E)
14 Tables
14.1 Form k for single sampling plans: s-method

Table 2 — Single sampling plans of form k for normal inspection: s–method
20
Code Acceptance quality limit (in percent nonconforming)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
n n n n n n n n n n n n n n n n
k k k k k k k k k k k k k k k k
B 3 4 4
0,950 0,735 0,586
ISO 3951-1:2022(E)
C 4 6 6 5
1,242 1,061 0,939 0,550
D 6 9 9 6 7
1,476 1,323 1,218 0,887 0,507
E 9 13 13 9 9 9
1,696 1,569 1,475 1,190 0,869 0,618
F 11 17 18 13 14 14 14
1,889 1,769 1,682 1,426 1,147 0,935 0,601
G 15 22 23 18 20 21 21 21
2,079 1,972 1,893 1,659 1,411 1,227 0,945 0,724
H 18 28 30 24 27 30 32 33 33
2,254 2,153 2,079 1,862 1,636 1,471 1,225 1,036 0,806
J 23 36 38 31 37 41 46 49 52 53
2,425 2,331 2,263 2,061 1,853 1,702 1,482 1,316 1,120 0,911
K 28 44 47 40 48 54 63 69 75 79 82
2,580 2,493 2,428 2,237 2,043 1,904 1,702 1,552 1,377 1,195 0,946
L 34 54 58 50 61 71 84 94 105 115 124
2,737 2,653 2,592 2,412 2,230 2,101 1,914 1,777 1,619 1,456 1,239
M 40 64 69 60 76 89 108 124 143 159 178
2,882 2,802 2,722 2,573 2,400 2,279 2,104 1,977 1,832 1,683 1,488
N 47 75 82 71 93 110 137 159 186 213 247
3,023 2,948 2,892 2,728 2,564 2,449 2,285 2,166 2,031 1,894 1,716
P 55 88 96 86 112 134 171 202 239 277 332
3,161 3,089 3,036 2,879 2,723 2,614 2,459 2,347 2,220 2,092 1,928
Q 63 101 110 102 132 159 207 244 293 348 424
3,288 3,219 3,167 3,016 2,867 2,762 2,615 2,508 2,388 2,268 2,114
NOTE The sample size code letters in this document correspond to those given in ISO 2859-1 and ISO 3951-2 in all cases where a correspondence can reasonably be considered.
© ISO 2022 – All rights reserved

Table 2 (continued)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
R 116 127 120 155 189 247 298 362 438 541
3,351 3,301 3,156 3,012 2,912 2,771 2,670 2,556 2,443 2,298
Key There is no suitable plan in this area; use the first sampling plan in the appropriate column.
If the sample size equals or exceeds the lot size, carry out 100 % inspection.

There is no suitable plan in this area; use the last sampling plan in the appropriate column.
21
ISO 3951-1:2022(E)
Table 3 — Single sampling plans of form k for tightened inspection: s–method
22
Acceptance quality limit (in percent nonconforming)
Code 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
letter n n n n n n n n n n n n n n n n
B 3 4
0,950 0,735
ISO 3951-1:2022(E)
C 4 6 6
1,242 1,061 0,939
D 6 9 9 6
1,476 1,323 1,218 0,887
E 9 13 13 9 9
1,696 1,569 1,475 1,190 0,869
F 11 17 18 13 14 14
1,889 1,769 1,682 1,426 1,147 0,935
G 15 22 23 18 20 21 21
2,079 1,972 1,893 1,659 1,411 1,227 0,945
H 18 28 30 24 27 30 32 33
2,254 2,153 2,079 1,862 1,636 1,471 1,225 0,954
J 23 36 38 31 37 41 46 50 53
2,425 2,331 2,263 2,061 1,853 1,702 1,482 1,245 1,010
K 28 44 47 40 48 54 63 71 78 82
2,580 2,493 2,493 2,237 2,043 1,904 1,702 1,489 1,281 1,045
L 34 54 58 50 61 71 84 99 111 122
2,737 2,653 2,592 2,412 2,230 2,101 1,914 1,720 1,533 1,325
M 40 64 69 60 76 89 108 131 150 170
2,882 2,802 2,744 2,573 2,400 2,279 2,104 1,924 1,752 1,564
N 47 75 82 73 93 110 137 169 201 233
3,023 2,948 2,892 2,728 2,564 2,449 2,285 2,117 1,958 1,785
P 55 88 96 86 112 134 171 214 260 233
3,161 3,089 3,036 2,879 2,723 2,614 2,459 2,300 2,152 1,992
Q 63 101 110 102 132 159 207 262 323 395
3,288 3,219 3,167 3,016 2,867 2,762 2,615 2,464 2,324 2,174

Table 3 (continued)
Code 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
R 90 116 127 120 155 189 247 320 398 498
3,408 3,351 3,301 3,156 3,012 2,912 2,771 2,628 2,495 2,354

23
ISO 3951-1:2022(E)
Table 4 — Single sampling plans of form k for reduced inspection: s–method
24
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 3 4 4 4 7
0,950 0,850 0,735 0,586 0,218
ISO 3951-1:2022(E)
E 4 6 6 6 5 9
1,242 1,155 1,061 0,939 0,550 0,162
F 6 8 9 9 6 7 8
1,476 1,406 1,323 1,218 0,887 0,507 0,231
G 9 11 13 13 9 9 9 12
1,696 1,642 1,569 1,475 1,190 0,869 0,628 0,237
H 11 15 17 18 13 14 14 14 13
1,889 1,835 1,769 1,682 1,426 1,147 0,935 0,601 0,454
J 15 19 22 23 18 20 21 21 21 21
2,079 2,033 1,972 1,893 1,659 1,411 1,227 0,945 0,830 0,626
K 18 24 28 30 24 27 30 32 33 33 33
1,254 2,209 2,153 2,079 1,862 1,636 1,471 1,225 1,126 0,954 0,806
L 23 30 36 38 31 37 41 46 48 50 52
2,425 2,385 2,331 2,263 2,061 1,853 1,702 1,482 1,394 1,245 1,120
M 28 37 44 47 40 48 54 63 66 71 75
2,580 2,543 2,493 2,428 2,237 2,043 1,904 1,702 1,622 1,489 1,377
N 34 44 54 58 50 61 71 84 90 99 105
2,737 2,701 2,653 2,592 2,412 2,230 2,101 1,914 1,642 1,720 1,619
P 40 52 64 69 60 76 89 108 117 131 143
2,882 2,848 2,802 2,744 2,573 2,400 2,279 2,104 2,037 1,924 1,832
Q 47 61 75 82 73 93 110 137 149 169 186
3,032 2,991 2,948 2,892 2,728 2,564 2,449 2,285 2,222 2,117 2,031
R 71 88 96 86 112 134 171 187 214 239
3,131 3,089 3,036 2,879 2,723 2,614 2,459 2,399 2,300 2,220

Table 4 (continued)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0

25
ISO 3951-1:2022(E)
ISO 3951-1:2022(E)
14.2 Form k for single sampling plans: σ-method

Table 5 — Single sampling plans of form k for normal inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 3 4 3
0,709 0,571 0,417
C 3 5 5 4

1,115 0,945 0,821 0,436
D 4 6 6 5 5
1,406 1,240 1,128 0,770 0,431
E 4 7 8 7 7 7
1,595 1,506 1,419 1,115 0,792 0,555
F 5 8 9 8 10 9 11
1,845 1,720 1,635 1,366 1,094 0,877 0,564
G 5 9 10 9 12 13 13 15
2,006 1,934 1,856 1,610 1,370 1,186 0,906 0,694
H 6 10 11 10 13 16 16 19 23
2,218 2,122 2,046 1,820 1,599 1,439 1,191 1,009 0,786
J 7 11 12 11 15 19 21 24 29 34
2,401 2,302 2,234 2,025 1,823 1,677 1,456 1,293 1,102 0,897
K 7 12 13 13 17 21 27 29 35 42 53
2,541 2,468 2,401 2,210 2,018 1,882 1,683 1,533 1,361 1,182 0,937
L 8 13 15 14 19 24 32 34 42 52 66
2,710 2,629 2,573 2,387 2,209 2,083 1,900 1,761 1,606 1,446 1,231
M 8 14 16 15 21 27 36 39 50 61 79
2,844 2,780 2,726 2,550 2,382 2,264 2,094 1,963 1,821 1,674 1,481
N 9 15 17 17 24 30 40 45 57 72 94
2,996 2,929 2,874 2,709 2,550 2,437 2,274 2,155 2,022 1,887 1,710
P 10 17 19 19 26 33 45 51 65 82 110
3,141 3,069 3,023 2,865 2,711 2,603 2,450 2,337 2,212 2,086 1,923
Q 11 18 20 20 28 35 49 57 72 92 125
3,275 3,207 3,155 3,002 2,856 2,752 2,607 2,500 2,381 2,262 2,110
27
ISO 3951-1:2022(E)
Table 5 (continued)
28
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
R 19 21 22 30 38 54 64 81 105 142
3,339 3,289 3,145 3,002 2,903 2,764 2,663 2,550 2,438 2,294
ISO 3951-1:2022(E)

Table 6 — Single sampling plans of form k for tightened inspection: σ–method
Code 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 3 4
0,709 0,571
C 3 5 5

1,115 0,945 0,821
D 4 6 6 5
1,406 1,240 1,128 0,770
E 4 7 8 7 7
1,595 1,506 1,419 1,115 0,792
F 5 8 9 8 10 9
1,845 1,720 1,635 1,366 1,094 0,877
G 5 9 10 9 12 13 13
2,006 1,934 1,856 1,610 1,370 1,186 0,906
H 6 10 11 10 13 16 16 20
2,218 2,122 2,046 1,820 1,599 1,439 1,191 0,929
J 7 11 12 11 15 19 21 25 32
2,401 2,302 2,234 2,025 1,823 1,677 1,456 1,223 0,994
K 7 12 13 13 17 21 27 31 39 49
2,541 2,468 2,401 2,210 2,018 1,882 1,683 1,471 1,267 1,035
L 8 13 15 14 19 24 32 37 47 61
2,710 2,629 2,573 2,387 2,209 2,083 1,900 1,705 1,521 1,316
M 8 14 16 15 21 27 36 43 55 72
2,844 2,780 2,726 2,550 2,382 2,264 2,094 1,912 1,742 1,556
N 9 15 17 17 24 30 40 49 65 85
2,996 2,929 2,874 2,709 2,550 2,437 2,274 2,106 1,950 1,779
P 10 17 19 19 26 33 45 55 74 99
3,142 3,076 3,023 2,865 2,711 2,603 2,450 2,291 2,145 1,987
Q 11 18 20 20 28 35 49 61 83 112
3,275 3,207 3,155 3,002 2,856 2,752 2,607 2,456 2,318 2,169
29
ISO 3951-1:2022(E)
Table 6 (continued)
30
Code 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
R 14 19 21 22 30 38 54 68 92 126
3,391 3,391 3,289 3,145 3,002 2,903 2,764 2,621 2,490 2,350
ISO 3951-1:2022(E)

Table 7 — Single sampling plans of form k for reduced inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 3 4 4 3 6
0,700 0,670 0,571 0,417 0,187
E 3 5 5 5 4 8

1,115 1,047 0,945 0,821 0,436 0,145
F 4 5 6 6 5 5 7
1,406 1,314 1,240 1,128 0,770 0,431 0,204
G 4 6 7 8 7 7 7 11
1,595 1,581 1,506 1,419 1,115 0,792 0,555 0,220
H 5 7 8 9 8 10 9 11 11
1,845 1,788 1,720 1,635 1,366 1,094 0,877 0,564 0,424
J 5 7 9 10 9 12 13 13 14 16
2,006 1,982 1,934 1,856 1,610 1,370 1,186 0,906 0,796 0,601
K 6 8 10 11 10 13 16 16 18 20 23
2,218 2,171 2,122 2,046 1,820 1,599 1,439 1,191 1,096 0,929 0,786
L 7 9 11 12 11 15 19 21 22 25 29
2,401 2,355 2,302 2,234 2,025 1,823 1,677 1,456 1,369 1,223 1,101
M 7 10 12 13 13 17 21 27 26 31 35
2,541 2,518 2,468 2,401 2,210 2,018 1,882 1,683 1,601 1,471 1,361
N 8 10 13 15 14 19 24 32 31 37 42
2,710 2,669 2,573 2,573 2,387 2,209 2,084 1,900 1,825 1,705 1,606
P 8 11 14 16 15 21 27 36 38 43 50
2,844 2,822 2,726 2,726 2,550 2,382 2,264 2,092 2,024 1,912 1,821
Q 9 12 15 17 17 24 30 40 45 49 57
2,996 2,969 2,929 2,874 2,709 2,550 2,437 2,274 2,212 2,106 2,022
R 13 17 19 19 26 33 45 50 55 65
3,113 3,076 3,023 2,865 2,711 2,603 2,459 2,390 2,291 2,212
31
ISO 3951-1:2022(E)
Table 7 (continued)
32
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
ISO 3951-1:2022(E)

ISO 3951-1:2022(E)
14.3 Form p* single sampling plans

34
Table 8 — Single sampling plans of form p* for normal inspection (ns sample size for s-method, nσ sample size for σ–method)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ ns nσ
p* p* p* p* p* p* p* p* p* p* p* p* p* p* p* p*
B 3 3 4 4 4 3
0,192 5 0,255 0 0,304 7
ISO 3951-1:2022(E)
C 4 3 6 5 6 5 5 4
0,086 0 0,145 3 0,179 3 0,307 4
D 6 4 9 6 9 6 6 5 7 5
0,052 2 0,087 2 0,108 2 0,194 6 0,314 9
E 9 4 13 7 13 8 9 7 9 7 9 7
0,032 8 0,051 9 0,064 7 0,114 3 0,196 1 0,274 3
F 11 5 17 8 18 9 13 8 14 10 14 9 14 11
0,019 6 0,033 0 0,041 4 0,072 0 0,124 5 0,176 1 0,277 1
G 15 5 22 9 23 10 18 9 20 12 21 13 21 13 21 15
0,012 4 0,020 1 0,025 2 0,043 8 0,076 3 0,108 5 0,172 9 0,236 2
H 18 6 28 10 30 11 24 10 27 13 30 16 32 16 33 19 33 23
0,007 5 0,012 7 0,015 9 0,027 5 0,048 0 0,068 6 0,109 4 0,150 0 0,210 9
J 23 7 36 11 38 12 31 11 37 15 41 19 46 21 49 24 52 29 53 34
0,004 8 0,007 9 0,009 8 0,016 8 0,029 6 0,042 4 0,067 8 0,093 2 0,131 1 0,181 4
K 28 7 44 12 47 13 40 13 48 17 54 21 63 27 69 29 75 35 79 42 82 53
0,003 0 0,005 0 0,006 2 0,010 7 0,018 8 0,026 9 0,043 1 0,059 4 0,083 6 0,115 7 0,172 2
L 34 8 54 13 58 15 50 14 61 19 71 24 84 32 94 34 105 42 115 52 124 66
0,001 9 0,003 1 0,003 9 0,006 6 0,011 6 0,016 7 0,026 8 0,036 9 0,052 0 0,072 2 0,107 4
M 40 8 64 14 69 16 60 15 76 21 89 27 108 36 124 39 143 50 159 61 178 79
0,001 2 0,002 0 0,002 4 0,004 1 0,007 3 0,010 5 0,016 9 0,023 4 0,032 9 0,045 7 0,068 0
N 47 9 75 15 82 17 71 17 93 24 110 30 137 40 159 45 186 57 213 72 247 94
0,000 7 0,001 2 0,001 5 0,002 6 0,004 6 0,006 6 0,010 6 0,014 7 0,020 7 0,028 7 0,042 8
P 55 10 88 17 96 19 86 19 112 26 134 33 171 45 202 51 239 65 277 82 332 110
0,000 5 0,000 8 0,000 9 0,001 6 0,002 9 0,004 1 0,006 6 0,009 1 0,012 9 0,017 9 0,026 7
Q 63 11 101 18 110 20 102 20 132 28 159 35 207 49 244 57 293 72 348 92 424 125
0,000 3 0,000 5 0,000 6 0,001 0 0,001 8 0,002 6 0,004 2 0,005 8 0,008 2 0,011 5 0,017 1

Table 8 (continued)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
R 116 19 127 21 120 22 155 30 189 38 247 54 298 64 362 81 438 105 541 142
0,000 3 0,000 4 0,000 6 0,001 1 0,001 6 0,002 6 0,003 6 0,005 1 0,007 1 0,010 7

35
ISO 3951-1:2022(E)
36
Table 9 — Single sampling plans of form p* for tightened inspection (ns sample size for s-method, nσ sample size for σ–method)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 3 3 4 4
0,192 5 0,255 0
ISO 3951-1:2022(E)
C 4 3 6 5 6 5
0,086 0 0,145 3 0,179 3
D 6 4 9 6 9 6 6 5
0,052 2 0,087 2 0,108 2 0,194 6
E 9 4 13 7 13 8 9 7 9 7
0,032 8 0,051 9 0,064 7 0,114 3 0,196 1
F 11 5 17 8 18 9 13 8 14 10 14 9
0,019 6 0,033 0 0,041 4 0,072 0 0,124 5 0,176 1
G 15 5 22 9 23 10 18 9 20 12 21 13 21 13
0,012 4 0,020 1 0,025 2 0,043 8 0,076 3 0,108 5 0,172 9
H 18 6 28 10 30 11 24 10 27 13 30 16 32 16 33 20
0,007 5 0,012 7 0,015 9 0,027 5 0,048 0 0,068 6 0,109 4 0,170 3
J 23 7 36 11 38 12 31 11 37 15 41 19 46 21 50 25 53 32
0,004 8 0,007 9 0,009 8 0,016 8 0,029 6 0,042 4 0,067 8 0,105 9 0,156 3
K 28 7 44 12 47 13 40 13 48 17 54 21 63 27 71 31 78 39 82 49
0,003 0 0,005 0 0,006 2 0,010 7 0,018 8 0,026 9 0,043 1 0,067 4 0,099 6 0,148 0
L 34 8 54 13 58 15 50 14 61 19 71 24 84 32 99 37 111 47 122 61
0,001 9 0,003 1 0,003 9 0,006 6 0,011 6 0,016 7 0,026 8 0,041 9 0,062 1 0,092 2
M 40 8 64 14 69 16 60 15 76 21 89 27 108 36 131 43 150 55 170 72
0,001 2 0,002 0 0,002 4 0,004 1 0,007 3 0,010 5 0,016 9 0,026 5 0,039 4 0,058 5
N 47 9 75 15 82 17 73 17 93 24 110 30 137 40 169 49 201 65 233 85
0,000 7 0,001 2 0,001 5 0,002 6 0,004 6 0,006 6 0,010 6 0,016 7 0,024 7 0,036 8
P 55 10 88 17 96 19 86 19 112 26 134 33 171 45 214 55 260 74 312 99
0,000 5 0,000 8 0,000 9 0,001 6 0,002 9 0,004 1 0,006 6 0,010 4 0,015 4 0,022 9
Q 63 11 101 18 110 20 102 20 132 28 159 35 207 49 262 61 323 83 395 112
0,000 3 0,000 5 0,000 6 0,001 0 0,001 8 0,002 6 0,004 2 0,006 6 0,009 8 0,014 7

Table 9 (continued)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
R 90 14 116 19 127 21 120 22 155 30 189 38 247 54 320 68 398 92 498 126
0,000 2 0,000 3 0,000 4 0,000 6 0,001 1 0,001 6 0,002 6 0,004 1 0,006 2 0,009 2

37
ISO 3951-1:2022(E)
38
Table 10 — Single sampling plans of form p* for reduced inspection (ns sample size for s-method, nσ sample size for σ–method)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 3 3 4 4 4 4 4 3 7 6
0,192 5 0,216 7 0,255 0 0,304 7 0,418 8
ISO 3951-1:2022(E)
E 4 3 6 5 6 5 6 5 5 4 9 8
0,086 0 0,120 9 0,145 3 0,179 3 0,307 4 0,438 3
F 6 4 8 5 9 6 9 6 6 5 7 5 8 7
0,052 2 0,070 9 0,087 2 0,108 2 0,194 6 0,314 9 0,413 0
G 9 4 11 6 13 7 13 8 9 7 9 7 9 7 12 11
0,032 8 0,041 6 0,051 9 0,064 7 0,114 3 0,196 1 0,274 3 0,408 8
H 11 5 15 7 17 8 18 9 13 8 14 10 14 9 14 11 13 11
0,019 6 0,026 7 0,033 0 0,041 4 0,072 0 0,124 5 0,176 1 0,277 1 0,328 4
J 15 5 19 7 22 9 23 10 18 9 20 12 21 13 21 13 21 14 21 16
0,012 4 0,016 1 0,020 1 0,025 2 0,043 8 0,076 3 0,108 5 0,172 9 0,204 5 0,267 5
K 18 6 24 8 28 10 30 11 24 10 27 13 30 16 32 16 33 18 33 20 33 23
0,007 5 0,010 2 0,012 7 0,015 9 0,027 5 0,048 0 0,068 6 0,109 4 0,129 6 0,170 3 0,210 9
L 23 7 30 9 36 11 38 12 31 11 37 15 41 19 46 21 48 22 50 25 52 29
0,004 8 0,006 2 0,007 9 0,009 8 0,016 8 0,029 6 0,042 4 0,067 8 0,080 6 0,105 9 0,131 1
M 28 7 37 10 44 12 47 13 40 13 48 17 54 21 63 27 66 26 71 31 75 35
0,003 0 0,004 0 0,005 0 0,006 2 0,010 7 0,018 8 0,026 9 0,043 1 0,051 3 0,067 4 0,083 6
N 34 8 44 10 54 13 58 15 50 14 61 19 71 24 84 32 90 31 99 37 105 42
0,001 9 0,002 5 0,003 1 0,003 9 0,006 6 0,011 6 0,016 7 0,026 8 0,031 8 0,041 9 0,052 0
P 40 8 52 11 64 14 69 16 60 15 76 21 89 27 108 36 117 38 131 43 143 50
0,001 2 0,001 5 0,002 0 0,002 4 0,004 1 0,007 3 0,010 5 0,016 9 0,020 1 0,026 5 0,032 9
Q 47 9 61 12 75 15 82 17 73 17 93 24 110 30 137 40 149 45 169 49 186 57
0,000 7 0,001 0 0,001 2 0,001 5 0,002 6 0,004 6 0,006 6 0,010 6 0,012 6 0,016 7 0,020 7
R 71 13 88 17 96 19 86 19 112 26 134 33 171 45 187 50 214 55 239 65
0,000 6 0,000 8 0,000 9 0,001 6 0,003 2 0,004 1 0,006 6 0,007 9 0,010 4 0,012 9

Table 10 (continued)
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0

39
ISO 3951-1:2022(E)
ISO 3951-1:2022(E)
14.4 Values of fσ for maximum process standard deviation (MPSD)
Table 11 — Values of fσ for maximum process standard deviation for combined control of
double specification limits: σ–method
Acceptance quality limit fσ
(in percent nonconforming)
0,010 0,128 5
0,015 0,131 9
0,025 0,136 5
0,040 0,141 2
0,065 0,146 6
0,10 0,152 0
0,15 0,157 5
0,25 0,165 4
0,40 0,173 7
0,65 0,183 7
1,0 0,194 1
1,5 0,205 6
2,5 0,223 1
4,0 0,243 5
6,5 0,271 0
10 0,304 0
NOTE The MPSD is obtained by multiplying the standardized MPSD, fσ, by the difference between the upper
specification limit, U, and the lower specification limit, L, i.e. MPSD = (U − L)fσ.
The MPSD indicates the greatest allowable magnitude of the process standard deviation when using
plans for combined control of double specification limits when the process variability is known. If the
process standard deviation is less than the MPSD, then there is a possibility, but not a certainty, that the
lot shall be accepted.
14.5 Supplementary acceptance constants for qualifying towards reduced inspection
Table 12 — Supplementary acceptance constants for qualifying towards reduced inspection

Sample size code AQL Acceptance constant for AQL that is one step tighter
letter s–method σ–method
%
k p* k
%
B 4,0 1,114 8,502 0,918
C 2,5 1,409 3,041 1,325
D 1,5 1,601 n/a 1,562
E 1,0 1,825 n/a 1,752
F 0,65 2,029 n/a 2,013
G 0,40 2,209 n/a 2,161
H 0,25 2,390 n/a 2,379
J 0,15 2,530 n/a 2,523
K 0,10 2,689 n/a 2,667
NOTE For this document, n/a is “not applicable.”

ISO 3951-1:2022(E)
Table 12 (continued)
Sample size code AQL Acceptance constant for AQL that is one step tighter
letter s–method σ–method
%
k p* k
%
L 0,065 2,857 n/a 2,847
M 0,040 2,995 n/a 2,972
N 0,025 3,143 n/a 3,131
P 0,015 3,254 n/a 3,246
Q 0,010 3,385 n/a 3,382
R 0,010 3,449 n/a 3,446
NOTE For this document, n/a is “not applicable.”
NOTE These constants were calculated so that the probability of acceptance at the next lower AQL is the
same as the probability of acceptance at the given AQL. For example, the normal inspection s–method sampling
plan for sample size code letter B and AQL of 4,0 % is found from Table 2 to be n = 3 and k = 0,950. This can be
shown to have a probability of acceptance of 96,440 4 % at process quality level 4,0 %. The next smaller AQL is
2,5 %. The s-method form k acceptance constant that provides the same probability of acceptance 96,440 4 %
with the same sample size n = 3 can be shown to be k = 1,114. Thus, in order to be acceptable at a one level tighter
AQL, the sample mean needs to be at least 1,114 times the sample standard deviation within specification, rather
than the 0,950 times the sample standard deviation needed to be considered merely acceptable.
15 Examples
15.1 General
Examples are included for the s-method (15.2), the σ-method (15.3) and the p*-method (15.4) with
details of the calculations and acceptance decisions given. The examples used for the p*-method are
a subset of those used for the s-method and the σ-method and this illustrates the equivalence of the
methods. The software that supports the use of this standard, described in Annex I, can carry out all
the calculations and acceptance decisions and produce the information given here.
15.2 Examples for the s-method

EXAMPLE 1 Single upper specification limit using the s–method.
The maximum temperature of operation for a certain device is specified as 60 °C and the operating
temperature is known from previous experience to be normally distributed. Production is
inspected in lots of 100 items and the process standard deviation is unknown. Inspection level
II, normal inspection with AQL of 2,5 % is to be used. From Table 1, the sample size code letter
is found to be F; from Table 2, it is seen that the sample size, n, is 13 and the form k acceptance
constant is 1,426. Suppose that the measurements are as follows: 53 °C; 57 °C; 49 °C; 58 °C; 59 °C;
54 °C; 58 °C; 56 °C; 50 °C; 50 °C; 55 °C; 54 °C; 57 °C. Conformity with the acceptance criterion is to
be determined.
Information needed Values obtained
Sample size: n 13
Form k acceptance constant: k 1,426
Sample mean: x = ∑ x / n 54,62 °C

ISO 3951-1:2022(E)
3,330 °C
Sample standard deviation: s = ∑( x j − x )2 /(n − 1)
j
Upper specification limit: U 60 °C

Upper quality statistic: QU = (U − x ) / s 1,617
Acceptance criterion: Is QU ≥ k ? Yes (1,617 ≥ 1,426)
The quality statistic is greater than or equal to the acceptance constant; therefore, the lot is accepted.
Note that the acceptance test, QU ≥ k , is equivalent to checking that ( s , x ) is below the acceptance
region boundary. This is shown in the following Figure 2.
Key
acceptance region boundary 1 reject
sample 2 accept
Figure 2 — Example of the use of an acceptance diagram, s–method
EXAMPLE 2 Single, lower specification limit using the s–method.
A certain pyrotechnic delay mechanism has a specified minimum delay time of 4,0 s. The process
standard deviation is unknown. Production is inspected in lots of 1 000 items and inspection level
II, normal inspection, is to be used with an AQL of 0,1 % applied to the lower limit. From Table 1,
it is seen that the sample size code letter is J. However, on entering Table 2 with sample size code
letter J and AQL of 0,1 %, it is found that there is an arrow pointing to the cell below. This means
that an entirely suitable plan is unavailable, and the next best plan is given by sample size code
letter K, with sample size, n = 28 and form k acceptance constant 2,580. A random sample of size 28
is drawn. Suppose the sample delay times, in seconds, are as follows:

ISO 3951-1:2022(E)
6,95 6,04 6,68 6,63 6,65 6,52 6,59 6,40 6,44 6,34 6,04 6,15 6,29 6,63
6,44 7,15 6,70 6,59 6,51 6,80 5,94 6,35 7,17 6,83 6,25 6,96 7,00 6,38
Conformity with the acceptance criterion is to be determined.
Sample size: n 28
Sample mean: x = ∑ x / n 6,551 s
0,325 1 s
j
Lower specification limit: L 4,0 s

Lower quality statistic: QL = ( x − L ) / s 7,847
Acceptance criterion: Is QL ≥ k ? Yes (7,847 ≥ 2,580)
The quality statistic is greater than or equal to the acceptance constant; therefore, the lot is accepted.
Note that the acceptance test, QL ≥ k , is equivalent to checking that ( s , x ) is above the acceptance
Key
sample 2 accept

ISO 3951-1:2022(E)
EXAMPLE 3 Combined control of double specification limits using the s–method.
Torpedoes supplied in batches of 100 are to be inspected for accuracy in the horizontal plane.
Positive or negative angular errors are equally unacceptable, so combined control of the double
specification limits is appropriate. The specification limits are set at 10 m either side of the point of
aim at a distance of 1 km, with an AQL of 4 %. Because testing is destructive and very costly, it has
been agreed between the producer and the responsible authority that special inspection level S-2 is
to be used. From Table 1, the sample size code letter is found to be B; from Table 2, it is seen that the
sample size, n, is 3 and the form k acceptance constant is 0,950. Three torpedoes are tested, yielding
errors –5,0 m, 6,7 m, and 8,8 m. Conformity with the acceptance criterion is to be determined.
Information needed Value obtained
Sample size: n 3
Acceptance constant: k 0,950

n 3,5 m
∑
1
Sample mean: x = xj
nj =1
n 7,436 m
Sample standard deviation: s = ∑ ( x j − x )2 /( n − 1 )
j =1
Upper specification limit: U 10,0 m
Lower specification limit: L –10,0 m

Determination of acceptance. Neither x < L or x > U so ( s , x ) can be plotted on the acceptance
diagram, as shown in Figure 4. The point is outside the acceptance region; therefore the lot is
rejected.
NOTE This lot is rejected even though all inspected items in the sample are within the specification limits.

ISO 3951-1:2022(E)
Key
sample 2 accept
Items are being manufactured in lots of size 25. The lower and upper specification limits on their
diameters are 82 mm to 84 mm. Items with diameters that are too large are equally unsatisfactory
as those with diameters that are too small, and it has been decided to control the total fraction
nonconforming using an AQL of 10 % at inspection level II. Normal inspection is to be instituted at
the beginning of inspection operations. From Table 1, the sample size code letter is found to be C;
from Table 2, it is seen that the sample size, n, is 5 and the form k acceptance constant is 0,550. The
diameters of five items from the first lot are measured, yielding diameters of 82,41 mm, 82,20 mm,
83,10 mm, 82,09 mm and 82,70 mm. Conformity with the acceptance criterion is to be determined.
Sample size: n 5
n 82,50 mm
∑ xj
1
Sample mean: x =
nj =1
n 0,408 2 mm
Sample standard deviation: s = ∑ ( x j − x )2 / ( n − 1 )
j =1
Upper specification limit: U 84,0 mm
Lower specification limit: L 82,0 mm

ISO 3951-1:2022(E)
s x −L 
Determination of acceptance. Neither x < L or x > U so  sS = , xS =  can be plotted on
 U − L U −L
the standardized acceptance diagram, as shown in Figure 5. The point is not outside the acceptance
region; therefore the lot is accepted.
Key
sample 2 accept
Figure 5 — Example of the use of a standardized acceptance diagram, s–method
The minimum temperature of operation for a certain device is specified as 60 °C and the maximum
temperature as 70 °C. Production is in inspection lots of 80 items. Inspection level II, normal
inspection, with AQL of 2,5 %, is to be used. From Table 1, the sample size code letter is found to
be E; from Table 2, it is seen that the sample size, n, is 13 and the form k acceptance constant is
1,475. Suppose the measurements obtained are as follows: 63,5 °C; 61,9 °C; 65,2 °C; 61,7 °C; 68,4 °C;
67,1 °C; 60,0 °C; 66,4 °C; 62,8 °C; 68,0 °C; 63,4 °C; 60,7 °C; 65,8 °C. Conformity with the acceptance
criterion is to be determined.
Sample size: n 13
2,789 9 °C
j

ISO 3951-1:2022(E)
Upper specification limit: U 70,0 °C
Lower specification limit: L 60,0 °C

Determination of acceptance. Neither x < L or x > U so ( s , x ) can be plotted on the acceptance
diagram, as shown in Figure 6. The point is outside the acceptance region; therefore, the lot is
rejected.
Key
sample 2 accept
For a medical inhalation device, the cap removal force has a specification interval of [5 N, 10 N].
No reliable information on process standard deviation is available, so that the s-method has to be
used. The usual lot size is ~100 000 units and with inspection level II, AQL = 1 %, from Table 1, the
sample size code letter is found to be N; from Table 2, it is seen that the sample size, n, is 186 and
the form k acceptance constant is 2,031. The observed measurement values were within a range of
7,03 N to 7,42 N and data normality can be assumed. Conformity with the acceptance criterion is to
be determined.
Sample size: n 186
Sample mean: x = ∑ x n 7,21 °C
0,053 8 °C
j
Lower specification limit: L 5 °N
Upper specification limit: U 10 °N

ISO 3951-1:2022(E)
s x −L 
Determination of acceptance. Neither x < L or x > U so  sS = , xS =  can be plotted on
 U − L U −L
the standardized acceptance diagram, as shown in Figure 7. The point is not outside the acceptance
region; therefore, the lot is accepted.
Key
sample 2 accept
Figure 7 — Example of the use of a standardized acceptance diagram, s–method
15.3 Examples for the σ-method

EXAMPLE 1 Single, lower specification limit using the σ–method.
The specified minimum yield point for certain steel castings is 400 N/mm2. The next lot of
500 items is submitted for inspection. Inspection level II, normal inspection, with AQL of 0,65 %,
is to be used. The value of σ is considered to be 21 N/mm2. From Table 1, it is seen that the sample
size code letter is H. From Table 5 for an AQL of 0,65 %, it is seen that the sample size, n, is 11 and
the form k acceptance constant is 2,046. Suppose the yield points of the sample specimens in N/
mm2 are: 431; 417; 469; 407; 450; 452; 427; 411; 429; 420; 400. Conformity with the acceptance
Sample size: n 11
Sample mean: x = ∑ x / n 428,5 N/mm2
Known σ 21 N/mm2
Lower specification limit: L 400 N/mm2

Lower quality statistic: QL = ( x − L ) σ 1,355
Acceptance criterion: Is QL x ≥ k ? No (1,355 < 2,046)

ISO 3951-1:2022(E)
The sample does not meet the acceptance criterion; therefore, the lot is rejected.
Note that the acceptance test, QL ≥ k , is equivalent to checking that ( s , x ) is above the acceptance
Key
sample 2 accept
Figure 8 — Example of the use of an acceptance diagram, σ–method
EXAMPLE 2 Single, upper specification limit using the σ–method.
The actuation force of a medical inhalation device has to be below the upper specification limit of
20 N in order to guarantee usability. Based on an extensive process validation study, the process
standard deviation can be assumed to be known, so that in this case the σ method can be applied.
With an average lot size of ~200 000 units, with normal inspection level II with an AQL = 0,4 %,
from Table 1, the sample size code letter is found to be P; from Table 5, it is seen that the sample
size, n, is 51 and the form k acceptance constant is 2,337. The observed measurement values were
within a range of 12,41 N to 18,98 N and data normality can be assumed. Conformity with the
acceptance criterion is to be determined.
Sample size: n 51

Sample mean: x 15,8 N
Known σ 1,42 N

ISO 3951-1:2022(E)
Specification limit: U 20 N
Upper quality statistic: QU = (U − x ) σ 2,958 N
Acceptance criterion: Is QU ≥ k ? Yes (2,958 ≥ 2,337)
The sample meets the acceptance criterion; therefore, the lot is accepted.
Note that the acceptance test, QU ≥ k , is equivalent to checking that ( s , x ) is below the acceptance
Key
sample 2 accept
EXAMPLE 3 Combined control of double specification limits using the σ–method.
The specification for electrical resistance of a certain electrical component is (520 ± 50) Ω. Production
is at a rate of 1 000 items per inspection lot. Inspection level II, normal inspection, with a single AQL of
1,5 %, is to be used for the two specification limits (470 Ω and 570 Ω). σ is known to be 18,5 Ω.
Factor from Table 11: fσ 0,205 6
Upper specification limit: U 570 Ω
Lower specification limit: L 470 Ω
Known σ 18,5 Ω

ISO 3951-1:2022(E)
It is noted that the known σ of 18,5 Ω is less than (U – L) × MPSD = 20,56 Ω so there is evidence that
there is a possibility that lots will be accepted so acceptance sampling can be undertaken. Entering
Table 1 with the lot size and inspection level, it is found that the sample size code letter is J; from
Table 2, it is seen that the sample size, n, is 19 and the form k acceptance constant is 1,677. Suppose
that the 19 sample values of the resistance, in Ω, are as follows: 515; 491; 479; 513; 521; 536; 483;
509; 514; 507; 484; 526; 532; 499; 530; 512; 492; 522; 488. Conformity with the acceptance criterion
is to be determined.
Further information needed Value obtained
Sample size: n 19
Acceptance constant (from Table 5) 1,677

Sample mean: x 508,1 Ω
Lower quality statistic QL = ( x − L ) σ 2,059
Upper quality statistic QU = (U − x ) σ 3,346
L ≤ x ≤ U ; k < QL and k < QU with neither QL or QU close to k, even though σ > 0 , 75 σ max = 15 , 417 ;
therefore the lot is accepted.
Note that this acceptance test is equivalent to checking that ( s , x ) is inside the acceptance region
boundary. This is shown in the following Figure 10.
Key
sample 2 accept

ISO 3951-1:2022(E)
15.4 Examples for the p*-method

EXAMPLE 1 Single upper specification limit using the s–method. Example 1 from 15.1.
The maximum temperature of operation for a certain device is specified as 60 °C and the operating
temperature is known from previous experience to be normally distributed. Production is
inspected in lots of 100 items and the process standard deviation is unknown. Inspection level II,
normal inspection with AQL of 2,5 % is to be used. From Table 1, the sample size code letter is
found to be F. From Table 8, it is seen that the sample size, n, is 13 and that the form p* acceptance
constant, is 0,072 0. Suppose that the measurements are as follows: 53 °C; 57 °C; 49 °C; 58 °C; 59 °C;
54 °C; 58 °C; 56 °C; 50 °C; 50 °C; 55 °C; 54 °C; 57 °C. Conformity with the acceptance criterion is to
be determined.
Sample size: n 13
Form p* acceptance constant: p* 0,072 0
3,330 °C
j
Upper specification limit: U 60 °C
Estimated fraction nonconforming

  1 1 U − x n   0,046 3
pÛ = F n n   max 0, −  

BETA  −1, −1 
 2 2 
  2 2 s n − 1  
Acceptance criterion: Is pÛ ≤ p * Yes (0,046 3 ≤ 0,072 0)
EXAMPLE 2 Combined control of double specification limits using the s–method. Example 5 from 15.1.
The minimum temperature of operation for a certain device is specified as 60 °C and the maximum
temperature as 70 °C. Production is in inspection lots of 80 items. Inspection level II, normal
inspection, with AQL of 2,5 %, is to be used. From Table 1, the sample size code letter is found to
be E. From Table 8, it is seen that the sample size, n, is 13 and that the form p* acceptance constant
is 0,064 7. Suppose the measurements obtained are as follows: 63,5 °C; 61,9 °C; 65,2 °C; 61,7 °C;
68,4 °C; 67,1 °C; 60,0 °C; 66,4 °C; 62,8 °C; 68,0 °C; 63,4 °C; 60,7 °C; 65,8 °C. Conformity with the
acceptance criterion is to be determined.
Sample size: n 13
2,789 9 °C
j
Upper specification limit: U 70,0 °C
Lower specification limit: L 60,0 °C

ISO 3951-1:2022(E)
Estimated fraction nonconforming 0,059 2

  1 1 x − L n  
pˆL = F n n   max 0, − 
BETA  −1, −1    2 2 s n − 1  
2 2 
  1 1 U − x n   0,011 6
pÛ = F n n   max 0, −  

BETA  −1, −1 
2 2    2 2 s n − 1  
pˆ = pˆL + pÛ 0,070 8
Acceptance criterion: Is pˆ ≤ p * No (0,070 8 > 0,064 7)

The sample does not meet the acceptance criterion; therefore the lot is rejected.
EXAMPLE 3 Single, lower specification limit using the σ–method. Example 1 from 15.2.
The specified minimum yield point for certain steel castings is 400 N/mm2. The next lot of
500 items is submitted for inspection. Inspection level II, normal inspection, with AQL of 0,65 %, is
to be used. The value of σ is considered to be 21 N/mm2. From Table 1, it is seen that the sample size
code letter is H. From Table 8, it is seen that for an AQL of 0,65 %, the sample size, n, is 11 and that
the form p* acceptance constant is 0,015 9. Suppose the yield points of the sample specimens in N/
mm2 are: 431; 417; 469; 407; 450; 452; 427; 411; 429; 420; 400. Conformity with the acceptance
Sample size: n 11

Sample mean: x 428,5 N/mm2
Known σ 21 N/mm2
Lower specification limit: L 400 N/mm2

L−x n 
p̂L = Φ  
 σ n − 1 

Acceptance criterion: Is pˆL ≤ p * ? No (0,082 4 > 0,072 0)
The sample does not meet the acceptance criterion; therefore, the lot is rejected.
EXAMPLE 4 Combined control of double specification limits using the σ–method. Example 3 from 15.2.
The specification for electrical resistance of a certain electrical component is (520 ± 50) Ω. Production
is at a rate of 1 000 items per inspection lot. Inspection level II, normal inspection, with a single AQL of
1,5 %, is to be used for the two specification limits (470 Ω and 570 Ω). σ is known to be 18,5 Ω.
Factor from Table 11: fσ 0,205 6
Upper specification limit: U 570 Ω
Known σ 18,5 Ω

ISO 3951-1:2022(E)
It is noted that the known σ of 18,5 Ω is less than (U – L) × MPSD = 20,56 Ω so there is evidence that
there is a possibility that lots will be accepted so acceptance sampling can be undertaken. Entering
Table 1 with the lot size and inspection level, it is found that the sample size code letter is J; from
Table 8, it is seen that the sample size, n, is 19 and that the form p* acceptance constant is 0,042 4.
Suppose that the 19 sample values of the resistance, in Ω, are as follows: 515; 491; 479; 513; 521;
536; 483; 509; 514; 507; 484; 526; 532; 499; 530; 512; 492; 522; 488. Sample acceptability is to be
determined.
Further information needed Value obtained
Sample size: n 19

Sample mean: x 508,0 Ω
Upper specification limit; U 570 Ω

L−x n 
p̂L = Φ  
 σ n − 1 

 x −U n  0,000 3
p̂U = Φ  
 σ n − 1 

pˆ = pˆL + pÛ 0,017 7
Acceptance criterion: Is pˆ ≤ p * Yes (0,017 7 ≤ 0,042 4)


ISO 3951-1:2022(E)
Annex A
(informative)
Procedures for obtaining s and σ
A.1 Procedure for obtaining s

A.1.1 Definition
The observed value of a sample standard deviation is generally denoted by the symbol s. It is defined in
Formula (A.1).
n
∑ ( x j − x )2
j =1
s= (A.1)
n −1
where
xj is the value of the quality characteristic of the jth item in a sample of n articles, expressed as a
decimal fraction;
x is the mean value of the xj, i.e.
n
1
x=
n ∑xj (A.2)
j =1
A.1.2 One-pass formula

Formula (A.1) for s is accurate but requires two passes through the sample data. A number of alternative
methods exist and these are analysed in Reference [17] and are described here. An algebraically
equivalent formula, which only requires a single pass through the sample data, is:
n n
n ∑ x 2j −( ∑ x j )2
j =1 j =1
s= (A.3)
n( n − 1 )
A.1.3 Potential numerical inaccuracy

Although Formula (A.3) is well suited for manual calculations it can produce poor results for difficult
problems where the variability is very small relative to the mean, i.e. s is very small in comparison with

ISO 3951-1:2022(E)
x . Although Formula (A.3) can be improved by subtracting a suitable arbitrary constant, a, from all the
values before computing s, i.e.
2
n  n 
∑
n ( x j − a ) −  x j − a
2
 j =1
∑

j =1
s= (A.4)
n( n − 1 )
it should be noted that there is no generally accepted guidance on how to choose this arbitrary constant
and if either of these formulae is used the result may be suspect should s / x turn out to be much less
than 1.
A.1.4 An updating algorithm for sequential data

Both Formula (A.1) and (A.4) require complete recalculation in situations where additional sample data
is to be added to the initial sample, for example in double sampling or sequential sampling. In these
situations an updating algorithm due to West, which makes the sample mean and sample standard
deviation at any stage in the updating process and which is almost as accurate as that for Formula (A.1)
is preferable:
M1 = x j ,T1 = 0
 x j − M j −1 
M j = M j −1 +   j = 2 , , n
 j 
 x j − M j −1 
T j = T j −1 + ( j − 1) ( x j − M j −1 )   j = 2 , , n (A.5)
 j 
xj =Mj
s j = T j ( j − 1)
A.1.5 Spreadsheets and pocket calculators
Many spreadsheets and pocket calculators have a standard deviation function. Unfortunately,
sometimes the sample size, n, is used by the machine in the denominator of Formula (A.1) instead of
n − 1. If it is planned to use a spreadsheet or calculator, it is important to check that the formula used
is equivalent to Formula (A.1). A simple check is to find the standard deviation of the three numbers 0,
1, and 2. The sample size n is 3, the sample mean is 1, the deviations from the mean are −1, 0, and 1, the
squares of the deviations are 1, 0, and 1, the sum of squares of the deviations is 2. So from Formula (A.1),
one obtains:
2
s= = 1 =1 (A.6)
2
If the spreadsheet or calculator is erroneously using n instead of n − 1 in the denominator, then the
result of the calculation is:
2
s= = 0 , 816 5
3
Use of n in the denominator shall be avoided, for otherwise the acceptance criterion is weakened and
any protection to the consumer is lost. The result obtained from the spreadsheet or calculator shall be
scaled by (n – 1)/n to obtain the correct result.
NOTE It is instructive to work through the use of Formula (A.3) for this example. It is found that

ISO 3951-1:2022(E)
3 × ( 02 + 12 + 22 ) − ( 0 + 1 + 2 )2 3 × ( 0 + 1 + 4 ) − 32 3× 5 − 9 6
s= = = = =1
3 × (3 − 1 ) 3× 2 6 6
as before.
A.2 Procedure for obtaining σroot

A.2.1 Definition
If it appears from the control chart that the value of s is in control, σroot can be presumed to be the
weighted root mean square of s given by the following Formula (A.7):
m
∑(ni − 1)si2
i =1
σ root = (A.7)
m
∑(ni − 1)
i =1
where
m is the number of lots;
ni is the sample size from the ith lot;
si is the sample standard deviation from the ith lot.
A.2.2 Simplication for equal sample sizes

If the sample sizes from each of the lots are equal, then Formula (A.7) simplifies to
m
∑ si2
i =1
σ root = (A.8)
m

ISO 3951-1:2022(E)
Annex B
(informative)
Accommodating measurement variability
B.1 General
The master tables, i.e. the tables in Clause 14, are based on the assumption that the true values of the
quality characteristic, x, of the items in the lots are normally distributed with unknown process mean,
μ, and either known or unknown process standard deviation, σ; the assumption is also made that x can
be measured without measurement error, i.e. that the measurement of an item with the true value,
xi, results in the value xi. This annex explains how the tables in Clause 14 can be used in the result of
measurement error. For further technical background, see Reference [19].
In the case of measurement error, the measured value of an item with true value, xi, differs from xi. It is
assumed that
— the measurement method is unbiased, i.e. the expected value of the measurement error is zero;
— measurement error inflates the perceived process variation and is independent of the actual process
standard deviation;
— measurement error is normally distributed with known or unknown measurement standard
deviation, σm.
It follows that the distribution of the measured values is a normal distribution with mean μ, and
standard deviation, given by Formula (B.1):
σ total = σ 2 + σ m
2
(B.1)
NOTE σtotal is always larger than σ if measurement error exists.
If it is known that σm < σ/10, i.e. the ratio γ = σ m / σ of measurement standard deviation to process
standard deviation is less than 10 %, the total standard deviation is given by Formula (B.2):
σ total < σ 2 + ( 0 , 1σ )2 = σ 1 + 0 , 01 = 1 , 005σ (B.2)
i.e. the standard deviation is increased by less than 0,5 %, which is negligible and hence, the sampling
plans do not need to be adjusted for measurement error.
In cases where σm ≥ 0,1σ, the sampling plans of this document shall be used with the following
adjustments.
a) Increase the sample size, n, in order to compensate for the perceived inflated variability, but do not
alter the acceptance constant, k or p*.
b) When the process standard deviation, σ, is known, use σ in calculating the test statistic, x ± kσ or
p̂ ; otherwise, use an estimate s of σ in calculating the test statistic x ± ks or p̂ .
Further details are given in B.2 to B.4 for three distinct cases.

ISO 3951-1:2022(E)
B.2 Process standard deviation σ and measurement standard deviation σm are

both known
a) Increase the sample size, n, of the sampling plan to
(
n* = n 1 + γ 2 ) (B.3)
b) Use the process standard deviation, σ, in calculating the test statistic x ± kσ or p̂ .
B.3 Process standard deviation σ is unknown but measurement standard

deviation σm is known
a) Increase the sample size, n, of the sampling plan as given by Formula (B.4):
(
n* = n 1 + γ 2 ) (B.4)
where γ is an estimated upper bound of γ = σ m / σ .

NOTE As γ increases, the operating characteristic curve of the sampling plan turns clockwise around the
indifference quality point (p50 %, 0,5), i.e. the point where the probability of acceptance of the lot is 50 %. If γ is
overestimated ( γ larger than γ ), the sampling plan is better than required, i.e. its probabilities of acceptance
are larger than required for P > p50 % and smaller than required for P > p50 %. Hence, overestimation of γ
ensures a sampling plan that is better than required.
b) Use the estimate given by Formula (B.5):
s * = s2 − σ m
2
(B.5)
of the process standard deviation instead of s in calculating the test statistic x ± ks or p̂ .
If s 2 − σ m
2
< 0 , use s* = 0.
B.4 Process standard deviation σ and measurement standard deviation σm are

both unknown
Increase the sample size, n, in accordance with Formula (B.4), perform duplicate (or multiple)
measurements on each sampled item, and use the measurement results to estimate the process
standard deviation separately from the measurement standard deviation, as shown below. Use this
estimate instead of s in calculating the test statistic x ± ks or p̂ .
a) Estimation of the process and measurement standard deviations.
To denote the jth measurement on the ith item by xij, the mean for the ith item by x , and the overall
mean by x .. . The number of measurements for the ith item shall be denoted by ni. The total sum of

ISO 3951-1:2022(E)
squares of the measurements about their overall mean can be partitioned as given in the following
Formula (B.6):
n ni n ni
∑∑ ( x ij − x .. )2 = ∑ ∑ ( xij − xi . + xi . − x.. )2
i =1 j =1 i =1 j =1
n ni
= ∑ ∑ ( xij − xi . )2 + ( xi . − x.. )2 + 2( xij − xi . )( xi . − x.. )
i =1 j =1
n ni n n ni
= ∑ ∑ ( xij − xi . ) + ∑ ni ( xi . − x.. )
2 2
+2 ∑( xi . − x.. )∑ ( xij − xi . ) (B.6)
i =1 j =1 i =1 i =1 j =1
n ni n
= ∑ ∑ ( xij − xi . ) + ∑ ni ( xi . − x.. )2 + 0
2
i =1 j =1 i =1
n ni n
= ∑ ∑ ( xij − xi . ) + ∑ ni ( xi . − x.. )2
2
i =1 j =1 i =1
=W+B
where
W is the within-items sum of squares;
B is the between-items sum of squares.

The expectations of these sums of squares are given in Formula (B.7):
n
E (W ) = σ m
2
∑(ni − 1) = σ m2 ( N − n) (B.7)
i =1
n
where N = ∑ ni is the total number of observations, and following Formula (B.8):
i =1
E(B ) = σ m
2
( n − 1 ) + ( N − n )σ 2 (B.8)
Hence, σ m
2
can be estimated by Formula (B.9)
σm
2
= W /( N − n ) (B.9)
and σ2 can be estimated by Formula (B.10)
s2 = σˆ2 =  B − (n − 1)σˆm
2
/ ( N − n) (B.10)
 
EXAMPLE
A manufactured component has a dimension with an upper specification limit of 13,05 cm. The process
standard deviation, σ, and measurement standard deviation, σm, are unknown, but from previous
experience, it is known that the ratio σm/σ is greater than 0,1 but less than 0,2. Lots of size 1 000 of
these components are to be inspected. Normal inspection is to be instituted with an AQL of 0,15 %.
From Table 1, it is found that the sample size code letter is J. As only one specification limit is being
controlled, form k can be used; from Table 2, the sampling plan for an AQL of 0,15 % in the absence of
sampling error is n = 23, k = 2,425.

ISO 3951-1:2022(E)
As σm / σ exceeds 0,1, it is necessary to adjust the sample size to allow for measurement uncertainty.
In the presence of the worst conceivable measurement error, the appropriate sample size (from
Formula B.3) is given by
( )
n* = n(1 + γ 2 ) = 23 1 + ( 0 , 2)2 = 23 × 1 , 04 = 23 , 92 .
The sample size should be an integer so, in order to provide at least the required AQL protection, n* is
rounded up to n* = 24. A random sample of 24 of the components is taken from the next lot, and, in order
to be able to assess the measurement uncertainty, each component is measured twice. The results for
the sample from the first lot are as shown in Table B.1.
Table B.1 — Sample results for the example of B.4

Item, Item, Item, Item, Item,
xi1 xi2 xi1 xi2 xi1 xi2 xi1 xi2 xi1 xi2
i i i i i
1 12,997 2 12,999 7 6 13,023 1 13,021 9 11 12,956 2 12,962 1 16 12,957 8 12,952 7 21 13,000 9 12,999 3
2 12,984 8 12,973 1 7 12,993 0 12,993 7 12 12,988 6 12,986 7 17 12,976 5 12,967 4 22 13,003 4 12,994 5
3 12,964 6 12,963 0 8 12,958 9 12,943 9 13 13,007 1 13,008 3 18 12,999 1 13,001 0 23 12,965 1 12,962 5
4 12,954 3 12,953 9 9 12,958 9 12,952 4 14 12,978 7 12,973 8 19 13,002 9 13,006 7 24 12,986 5 12,985 2
5 12,976 3 12,980 2 10 13,015 0 13,016 4 15 12,927 4 12,927 7 20 12,968 8 12,976 2
The accuracy of subsequent calculations can be improved by subtracting an arbitrary constant that
reduces the number of significant figures. Denote the constant by c and set c = 12,9. The resulting values
of yij = xij − 12,9 are as given in Table B.2.
Table B.2 — Sample results with subtraction of 12,9 for the example of B.4
Item, Item, Item, Item, Item,
yi1 yi2 yi1 yi2 yi1 yi2 yi1 yi2 yi1 yi2
i i i i i
1 0,097 2 0,099 7 6 0,123 1 0,121 9 11 0,056 2 0,062 1 16 0,057 8 0,052 7 21 0,100 9 0,099 3
2 0,084 8 0,071 1 7 0,093 0 0,093 7 12 0,088 6 0,086 7 17 0,076 5 0,067 4 22 0,103 4 0,094 5
3 0,064 6 0,063 0 8 0,058 9 0,043 9 13 0,107 1 0,108 3 18 0,099 1 0,101 0 23 0,065 1 0,062 5
4 0,054 3 0,053 9 9 0,058 9 0,052 4 14 0,078 7 0,073 8 19 0,102 9 0,099 2 24 0,086 5 0,085 2
5 0,076 3 0,080 2 10 0,115 0 0,116 4 15 0,027 4 0,027 7 20 0,068 8 0,076 2
24 2
The sum of the yij is ∑ ∑ yii = 3, 839 9 .
i =1 j =1
The sample mean value of y is y = 3,839 9/48 = 0,079 998.
Hence, the sample mean value of x is x = c + y = 12,9 + 0,079 998 = 12,979 998.
24 2
The total sum of squares of y is T = ∑ ∑ yij2 = 0,332 791 15 .
i =1 j =1
The total sum of squares, T, about the overall sample mean is given by Formula (B.11):
24 2  224 
2 
= ∑∑ − yij2

 ∑ ∑ yij  / 2
 
(B.11)
i =1 j =1 i =1  j =1  

= 0,332 791 15 – 0,307 184 00
= 0,025 607 15

ISO 3951-1:2022(E)
The within-items sum of squares, W, is given by Formula (B.12):

24 2
W= ∑ ∑ ( yij − yi . )2
i =1 j =1
2
24 2 24  2 
= ∑∑ yij2 ∑∑
− 

yij  / 2

(B.12)
i =1 j =1 i =1  j =1 
= 0 , 332 791 15 − 0 , 332 407 52
= 0 , 000 383 63
By subtraction, the between-item sum of squares, B, is given by Formula (B.13):
B = T −W
= 0 , 025 607 15 − 0 , 000 383 63 (B.13)
= 0 , 025 223 52
The measurement error variance is estimated as
σˆm
2
= W / ( N − n) = 0, 000 383 63 / (48 − 24) = 0, 000 015 984 6
The process variance is estimated as
s2 = σˆ2 =  B − (n − 1)σ m
2
/ ( N − n)
 
= 0, 025 223 52 − 23 × 0, 000 015 984 6  / (48 − 24)
= 0, 024 855 87 / 24
= 0, 001 035 66
so the process standard deviation is estimated as
s = σ = 0 , 001 035 66 = 0 , 032 182

U − 2 , 425 s = 13 , 05 − 2 , 425 × 0 , 032 182 = 12 , 972
As x = 12 , 980 > 12 , 972 , the lot is rejected.

ISO 3951-1:2022(E)
Annex C
(informative)
Sampling strategies
The sampling strategy is an important part of statistical sampling. Figure C.1 illustrates how the n
sample units are obtained from the lot.
For industrial applications, the following sampling strategies can be differentiated.
a) (Simple) Random sampling (SRS) – Each item in the lot has the same probability to be selected for
the sample.
b) Convenience sampling (CS) – Items that are most convenient and easy to select are taken.
c) Systematic sampling (SyS) – The items in the sample are systematically (by time, number, etc.)
spread over the lot.
d) Stratified sampling (StS) – The lot is divided into sub lots (strata) from which (random) samples are
taken. Usually sub lots and sub lot samples have equal size, but also different sizes of sub lots and/
or samples are feasible.
e) Cluster sampling (ClS) – The lot is divided into sub lots (clusters) from which sub lots are (randomly)
selected. For these selected sub lots all items are sampled (100 % inspection).
The drawings in Figure C.1 illustrate the different sampling strategies for a two-dimensional population,
e.g. tablets spread open for a drying process.
a) SRS b) CS c) SyS
d) StS e) ClS
Figure C.1 — Examples of different sampling strategies with a two-dimensional population
Generally, there is no statistical guidance on which sampling strategy is superior to the others. Some
statistical comparisons of different sampling strategies has been carried out (see Reference [3]).
However, this standard and associated probabilities does assume that random sampling is being used
no matter which sampling strategy is selected. There is risk for the validity of the results if convenience
(biased) sampling is used.

ISO 3951-1:2022(E)
ISO 2859-4:2020 specifies: “The items selected for the sample shall be drawn from the lot by simple
random sampling. However, when the lot consists of sub-lots or strata, identified by some rational
criterion, stratified sampling shall be used in such way that the size of the subsample from each sublot
or stratum is proportional to the size of that sublot or stratum”.
Under general conditions, i.e. if there are no technically or practically founded assumptions on the
existence of specific nonconformity patterns, a sampling strategy, including a systematic as well as a
random component, yields successful sampling in most practical applications.

ISO 3951-1:2022(E)
Annex D
(informative)
Operating characteristics for the σ–method
D.1 Probability of acceptance for the σ-method

The exact probability of lot acceptance for a single specification limit at process fraction nonconforming,
p, when the process standard deviation is known is given by Formula (D.1):
Pa = 1 − Φ ( k − K p ) n  (D.1)
where
Φ(.) is the distribution function of the standardized normal distribution;
n is the sample size;
Kp is the upper p-quantile of the standardized normal distribution;
k is the σ–method acceptance constant.
D.2 Example
Consider the calculation of the probability of acceptance at a process quality of 2,5 % nonconforming for
a σ–method plan with an AQL of 1,0 % and sample size code letter M under normal inspection. Entering
Table 5 with sample size code letter M and AQL of 1,0 %, it is found that the sample size, n, is 39 and the
acceptance constant, k, is 1,963. The process fraction nonconforming under consideration is p = 0,025 0,
and Kp = 1,960. Hence,
Pa = 1 − Φ (1 , 963 − 1 , 960 ) 39  = 1 − Φ ( −0 , 018 7 ) = 1 − 0 , 5075 = 0 , 492 5 .
D.3 Comparison with tabulated value for the s–method

It is instructive to observe that this probability of acceptance for the σ–method is very roughly in
agreement with the corresponding probability of acceptance for the s–method. From the column of the
table with code letter M in Annex E for AQL of 1,0 %, it is seen that a process quality level of 2,43 %, i.e.
P = 0,024 3, corresponds to a probability of acceptance of 50 %, i.e. to Pa = 0,500.

ISO 3951-1:2022(E)
Annex E
(informative)
Operating characteristic for the s-method – tabulated values for

single sampling plans, normal inspection
E.1 Probability of acceptance for the s-method

The exact probability of acceptance for a single specification limit at process fraction nonconforming
p, when the process standard deviation is unknown is given by Formula (E.1). In addition, Figure E.1
shows the operating characteristic curves for single sampling plans for normal inspection, Code Letter
B.
Pa = 1 − Ft
(n−1,K p n ) ( k n ) (E.1)
where
Ft (ν ,δ ) ( . ) is the distribution function of the noncentral t-distribution with ν degrees of freedom

and noncentrality δ;
n is the sample size;
Kp is the upper p-quantile of the standardized normal distribution;
k is the σ–method acceptance constant.
Key
X process quality (in percent nonconforming)
Y percent of lots expected to be accepted (Pa)
Figure E.1 — Operating characteristic curves for single sampling plans, normal inspection,
Code Letter B

ISO 3951-1:2022(E)
Table E.1 — Tabulated values for operating characteristic curves for single sampling plans –
sample size code letter B: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter B Pa
% 4,0 — 6,5 10,0 — %
99,0 0,458 1,34 2,06 3,35 13,65 99,0
95,0 1,94 3,73 5,11 7,37 20,19 95,0
90,0 3,73 5,98 7,80 10,64 24,33 90,0
75,0 9,32 11,85 14,40 18,18 32,15 75,0
50,0 20,49 21,92 25,10 29,57 41,87 50,0
25,0 36,55 35,40 38,75 43,34 52,11 25,0
10,0 53,01 49,17 52,27 56,44 61,22 10,0
5,0 62,60 57,47 60,26 64,00 66,43 5,0
1,0 78,03 71,75 73,82 76,56 75,33 1,0
6,5 — 10,0 — —
Acceptance quality limit (tightened inspection) in percent – sample size code letter B
1,5 2,5 4,0 6,5 10,0
Acceptance quality limit (reduced inspection) in percent – sample size code letter D
sample size code letter C: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter C Pa
% 2,5 — 4,0 6,5 10,0 — %
99,0 0,224 0,827 1,20 1,87 4,91 15,80 99,0
95,0 1,02 2,25 2,99 4,22 9,38 22,38 95,0
90,0 2,03 3,60 4,61 6,20 12,72 26,44 90,0
75,0 5,45 7,23 8,72 10,98 19,96 33,97 75,0
50,0 13,04 13,79 15,82 18,73 30,33 43,14 50,0
25,0 25,31 23,30 25,69 29,01 42,51 52,70 25,0
10,0 39,48 34,03 36,51 39,88 54,11 61,19 10,0
5,0 48,59 41,10 43,52 46,77 60,93 66,05 5,0
1,0 65,06 54,73 56,84 59,65 72,67 74,43 1,0
4,0 — 6,5 10,0 — —
Acceptance quality limit (tightened inspection) in percent – sample size code letter C
1,0 1,5 2,5 4,0 6,5 10,0
Acceptance quality limit (reduced inspection) in percent – sample size code letter E

ISO 3951-1:2022(E)
sample size code letter D: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter D Pa
% 1,5 — 2,5 4,0 6,5 10,0 — %
99,0 0,194 0,488 0,828 1,23 2,23 6,34 14,51 99,0
95,0 0,746 1,35 1,97 2,69 4,84 11,05 20,81 95,0
90,0 1,40 2,18 2,98 3,93 6,99 14,39 24,74 90,0
75,0 3,53 4,49 5,56 6,95 12,05 21,33 32,10 75,0
50,0 8,24 8,89 10,11 12,01 20,07 30,92 41,18 50,0
25,0 16,26 15,70 16,75 19,08 30,50 41,97 50,76 25,0
10,0 26,37 23,94 24,51 27,08 41,37 52,48 59,36 10,0
5,0 33,48 29,70 29,85 32,49 48,20 58,72 64,32 5,0
1,0 47,88 41,62 40,84 43,42 60,87 69,71 72,95 1,0
2,5 — 4,0 6,5 10,0 — —
Acceptance quality limit (tightened inspection) in percent – sample size code letter D
0,65 1,0 1,5 2,5 4,0 6,5 10,0
Acceptance quality limit (reduced inspection) in percent – sample size code letter F
sample size code letter E: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter E Pa
% 1,0 — 1,5 2,5 4,0 6,5 10,0 — %
99,0 0,168 0,313 0,549 0,785 1,36 3,79 7,26 18,19 99,0
95,0 0,552 0,839 1,25 1,68 2,92 6,70 11,44 23,93 95,0
90,0 0,975 1,35 1,87 2,44 4,22 8,83 14,25 27,36 90,0
75,0 2,28 2,77 3,46 4,31 7,36 13,40 19,89 33,57 75,0
50,0 5,13 5,56 6,30 7,51 12,55 20,06 27,50 41,03 50,0
25,0 10,08 10,05 10,58 12,16 19,74 28,31 36,27 48,83 25,0
10,0 16,68 15,80 15,84 17,70 27,80 36,81 44,82 55,90 10,0
5,0 21,59 20,02 19,62 21,62 33,22 42,23 50,06 60,05 5,0
1,0 32,44 29,31 27,87 30,01 44,13 52,64 59,79 67,51 1,0
1,5 — 2,5 4,0 6,5 10,0 — —
Acceptance quality limit (tightened inspection) in percent – sample size code letter E
0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
Acceptance quality limit (reduced inspection) in percent – sample size code letter G

ISO 3951-1:2022(E)
sample size code letter F: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter F Pa
% 0,65 — 1,0 1,5 2,5 4,0 6,5 10,0 — %
99,0 0,102 0,231 0,365 0,552 0,940 2,57 4,67 10,19 13,02 99,0
95,0 0,339 0,584 0,821 1,14 1,95 4,43 7,33 14,24 17,73 95,0
90,0 0,605 0,918 1,22 1,63 2,79 5,79 9,13 16,76 20,63 90,0
75,0 1,45 1,84 2,25 2,84 4,81 8,72 12,84 21,60 26,06 75,0
50,0 3,35 3,63 4,13 4,92 8,21 13,09 18,00 27,80 32,84 50,0
25,0 6,82 6,57 7,04 8,01 13,05 18,71 24,25 34,73 40,23 25,0
10,0 11,70 10,45 10,74 11,79 18,74 24,78 30,70 41,44 47,19 10,0
5,0 15,50 13,39 13,48 14,54 22,71 28,83 34,86 45,58 51,41 5,0
1,0 24,31 20,15 19,69 20,67 31,17 37,07 43,08 53,43 59,23 1,0
1,0 — 1,5 2,5 4,0 6,5 10,0 — —
Acceptance quality limit (tightened inspection) in percent – sample size code letter F
0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
Acceptance quality limit (reduced inspection) in percent – sample size code letter H
sample size code letter G: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter G Pa
% 0,40 — 0,65 1,0 1,5 2,5 4,0 6,5 — 10,0 — %
99,0 0,077 2 0,144 0,231 0,335 0,601 1,58 2,88 6,02 7,85 9,85 11,96 99,0
95,0 0,236 0,362 0,514 0,697 1,23 2,73 4,51 8,54 10,76 13,12 15,56 95,0
90,0 0,406 0,568 0,762 1,00 1,75 3,57 5,63 10,16 12,59 15,13 17,74 90,0
75,0 0,932 1,14 1,41 1,76 3,00 5,41 7,97 13,34 16,09 18,93 21,79 75,0
50,0 2,10 2,28 2,60 3,10 5,16 8,24 11,32 17,57 20,65 23,75 26,84 50,0
25,0 4,25 4,21 4,49 5,14 8,32 11,99 15,53 22,53 25,86 29,16 32,40 25,0
10,0 7,37 6,85 6,97 7,73 12,17 16,21 20,04 27,56 31,05 34,46 37,76 10,0
5,0 9,86 8,90 8,86 9,67 14,96 19,12 23,06 30,81 34,34 37,78 41,09 5,0
1,0 15,94 13,82 13,30 14,13 21,13 25,29 29,30 37,26 40,81 44,22 47,47 1,0
0,65 — 1,0 1,5 2,5 4,0 6,5 — 10,0 — —
Acceptance quality limit (tightened inspection) in percent – sample size code letter G
0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 — 10,0

ISO 3951-1:2022(E)
sample size code letter H: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter H Pa
% 0,25 — 0,40 0,65 1,0 1,5 2,5 4,0 — 6,5 — 10,0 %
99,0 0,047 8 0,096 6 0,153 0,225 0,398 1,01 1,85 3,85 5,03 6,22 7,49 10,21 99,0
95,0 0,146 0,237 0,334 0,457 0,797 1,73 2,88 5,43 6,85 8,29 9,77 12,90 95,0
90,0 0,253 0,368 0,492 0,650 1,13 2,26 3,60 6,46 8,01 9,57 11,18 14,51 90,0
75,0 0,586 0,734 0,901 1,13 1,92 3,45 5,09 8,49 10,25 12,03 13,82 17,49 75,0
50,0 1,35 1,47 1,67 1,98 3,31 5,29 7,27 11,24 13,22 15,22 17,21 21,21 50,0
25,0 2,79 2,72 2,90 3,30 5,38 7,80 10,06 14,55 16,71 18,91 21,07 25,34 25,0
10,0 4,96 4,47 4,54 5,01 7,96 10,71 13,14 18,01 20,28 22,64 24,92 29,37 10,0
5,0 6,75 5,87 5,82 6,30 9,87 12,76 15,24 20,30 22,61 25,04 27,38 31,91 5,0
1,0 11,27 9,32 8,90 9,36 14,25 17,25 19,72 24,99 27,34 29,87 32,28 36,89 1,0
0,40 — 0,65 1,0 1,5 2,5 4,0 6,5 — — 10,0 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter H
0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 — 6,5 10,0
Acceptance quality limit (reduced inspection) in percent – sample size code letter K
sample size code letter J: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter J Pa
% 0,15 — 0,25 0,40 0,65 1,0 1,5 2,5 — 4,0 — 6,5 — 10,0 %
99,0 0,033 4 0,0625 0,102 0,144 0,251 0,645 1,15 2,39 3,11 3,84 4,62 6,30 8,06 9,86 99,0
95,0 0,097 1 0,150 0,216 0,288 0,498 1,09 1,79 3,37 4,24 5,12 6,04 7,96 9,95 11,97 95,0
90,0 0,164 0,232 0,315 0,408 0,701 1,42 2,23 4,01 4,97 5,93 6,92 8,97 11,08 13,21 90,0
75,0 0,372 0,459 0,569 0,707 1,20 2,16 3,17 5,29 6,39 7,48 8,60 10,86 13,16 15,46 75,0
50,0 0,841 0,915 1,04 1,24 2,07 3,31 4,56 7,05 8,30 9,54 10,79 13,27 15,75 18,24 50,0
25,0 1,74 1,71 1,81 2,08 3,39 4,90 6,37 9,20 10,59 11,97 13,34 16,00 18,65 21,30 25,0
10,0 3,11 2,84 2,86 3,18 5,09 6,78 8,41 11,51 12,99 14,48 15,94 18,74 21,52 24,29 10,0
5,0 4,27 3,76 3,68 4,03 6,37 8,13 9,84 13,06 14,59 16,14 17,64 20,50 23,34 26,17 5,0
1,0 7,27 6,08 5,70 6,09 9,38 11,16 12,93 16,32 17,91 19,53 21,10 24,04 26,96 29,89 1,0
0,25 — 0,40 0,65 1,0 1,5 2,5 4,0 — — 6,5 — 10,0 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter J
0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 — 4,0 6,5 — —
Acceptance quality limit (reduced inspection) in percent – sample size code letter L

ISO 3951-1:2022(E)
sample size code letter K: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter K Pa
% 0,10 — 0,15 0,25 0,40 0,65 1,0 1,5 — 2,5 — 4,0 — 6,5 — 10,0 %
99,0 0,022 7 0,042 2 0,066 4 0,094 4 0,169 0,420 0,741 1,53 1,98 2,45 2,94 3,99 5,09 6,22 8,64 10,49 99,0
95,0 0,064 0 0,098 8 0,139 0,187 0,328 0,703 1,14 2,15 2,70 3,27 3,84 5,06 6,30 7,57 10,23 12,25 95,0
90,0 0,107 0,151 0,202 0,263 0,457 0,912 1,43 2,56 3,16 3,78 4,40 5,71 7,03 8,37 11,16 13,27 90,0
75,0 0,239 0,295 0,364 0,453 0,772 1,38 2,02 3,38 4,08 4,78 5,48 6,93 8,38 9,84 12,83 15,09 75,0
50,0 0,539 0,585 0,667 0,794 1,32 2,12 2,92 4,52 5,32 6,11 6,91 8,51 10,09 11,69 14,88 17,29 50,0
25,0 1,12 1,09 1,16 1,33 2,17 3,15 4,10 5,93 6,83 7,71 8,59 10,33 12,04 13,76 17,14 19,68 25,0
10,0 2,01 1,82 1,85 2,05 3,27 4,39 5,45 7,46 8,45 9,39 10,33 12,18 13,98 15,81 19,33 21,98 10,0
5,0 2,78 2,43 2,39 2,61 4,10 5,29 6,41 8,50 9,53 10,51 11,49 13,39 15,24 17,12 20,72 23,43 5,0
1,0 4,83 3,98 3,75 3,99 6,11 7,35 8,53 10,73 11,82 12,84 13,88 15,87 17,79 19,76 23,47 26,27 1,0
0,15 — 0,25 0,40 0,65 1,0 1,5 2,5 — — 4,0 — 6,5 — 10,0 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter K
0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 — 2,5 4,0 — — — —
Acceptance quality limit (reduced inspection) in percent – sample size code letter M
sample size code letter L: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter L Pa
% 0,065 — 0,10 0,15 0,25 0,40 0,65 1,0 — 1,5 — 2,5 — 4,0 — 6,5 %
99,0 0,014 9 0,026 6 0,043 2 0,061 0 0,108 0,264 0,470 0,959 1,24 1,53 1,84 2,48 3,15 3,86 5,34 6,47 99,0
95,0 0,041 0 0,061 8 0,089 0 0,119 0,207 0,440 0,720 1,34 1,68 2,03 2,40 3,14 3,91 4,70 6,33 7,58 95,0
90,0 0,068 0 0,094 4 0,128 0,166 0,287 0,570 0,894 1,60 1,97 2,35 2,75 3,55 4,37 5,21 6,92 8,23 90,0
75,0 0,150 0,184 0,229 0,285 0,483 0,862 1,27 2,11 2,54 2,98 3,42 4,32 5,22 6,14 7,99 9,39 75,0
50,0 0,336 0,367 0,418 0,497 0,827 1,33 1,82 2,83 3,32 3,83 4,32 5,32 6,31 7,32 9,31 10,82 50,0
25,0 0,697 0,690 0,729 0,835 1,36 1,98 2,57 3,73 4,28 4,85 5,39 6,50 7,57 8,66 10,78 12,39 25,0
10,0 1,27 1,16 1,16 1,29 2,06 2,78 3,43 4,72 5,31 5,94 6,52 7,71 8,85 10,00 12,24 13,93 10,0
5,0 1,76 1,56 1,51 1,65 2,60 3,37 4,04 5,40 6,02 6,67 7,27 8,50 9,68 10,87 13,16 14,91 5,0
1,0 3,11 2,60 2,39 2,54 3,92 4,72 5,42 6,87 7,52 8,22 8,84 10,16 11,39 12,64 15,03 16,85 1,0
0,10 — 0,15 0,25 0,40 0,65 1,0 1,5 — — 2,5 — 4,0 — 6,5 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter L
0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 — 1,5 2,5 — — — —
Acceptance quality limit (reduced inspection) in percent – sample size code letter N

ISO 3951-1:2022(E)
sample size code letter M: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter M Pa
% 0,04 — 0,065 0,10 0,15 0,25 0,40 0,65 — 1,0 — 1,5 — 2,5 — 4,0 %
99,0 0,009 7 0,017 2 0,027 8 0,039 1 0,068 5 0,170 0,299 0,610 0,787 0,973 1,17 1,58 2,00 2,44 3,36 4,09 99,0
95,0 0,026 3 0,039 5 0,056 8 0,075 5 0,131 0,282 0,457 0,85 1,07 1,29 1,52 1,99 2,48 2,98 4,00 4,79 95,0
90,0 0,043 4 0,060 1 0,081 6 0,106 0,181 0,364 0,567 1,01 1,25 1,49 1,74 2,25 2,77 3,30 4,38 5,20 90,0
75,0 0,095 2 0,117 0,145 0,180 0,305 0,548 0,803 1,34 1,61 1,89 2,17 2,74 3,31 3,89 5,06 5,95 75,0
50,0 0,213 0,233 0,265 0,315 0,524 0,842 1,16 1,80 2,11 2,43 2,75 3,38 4,02 4,65 5,92 6,87 50,0
25,0 0,444 0,439 0,465 0,532 0,868 1,26 1,64 2,37 2,73 3,09 3,44 4,13 4,84 5,52 6,89 7,89 25,0
10,0 0,813 0,746 0,743 0,826 1,33 1,77 2,19 3,02 3,40 3,79 4,17 4,91 5,68 6,39 7,85 8,91 10,0
5,0 1,14 1,00 0,97 1,06 1,68 2,16 2,60 3,46 3,86 4,27 4,66 5,43 6,23 6,97 8,47 9,55 5,0
1,0 2,04 1,69 1,55 1,65 2,57 3,05 3,51 4,44 4,85 5,29 5,70 6,51 7,37 8,14 9,73 10,86 1,0
0,065 — 0,10 0,15 0,25 0,40 0,65 1,0 — — 1,5 — 2,5 — 4,0 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter M
0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 — 1,0 1,5 — — — —
Acceptance quality limit (reduced inspection) in percent – sample size code letter P
sample size code letter N: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter N Pa
% 0,025 — 0,04 0,065 0,10 0,15 0,25 0,40 — 0,65 — 1,0 — 1,5 — 2,5 %
99,0 0,006 4 0,011 1 0,017 6 0,025 0 0,0445 0,109 0,190 0,387 0,497 0,614 0,735 0,991 1,26 1,54 2,12 2,57 99,0
95,0 0,016 9 0,025 1 0,035 8 0,048 0 0,083 7 0,178 0,289 0,538 0,673 0,813 0,956 1,25 1,56 1,87 2,51 3,01 95,0
90,0 0,027 6 0,038 0 0,051 3 0,066 8 0,115 0,230 0,358 0,639 0,787 0,940 1,10 1,41 1,74 2,07 2,75 3,27 90,0
75,0 0,060 2 0,073 7 0,091 2 0,114 0,193 0,345 0,506 0,842 1,01 1,19 1,37 1,72 2,08 2,45 3,19 3,74 75,0
50,0 0,134 0,146 0,167 0,198 0,330 0,531 0,730 1,13 1,33 1,53 1,73 2,13 2,53 2,93 3,73 4,33 50,0
25,0 0,280 0,277 0,293 0,335 0,547 0,796 1,03 1,50 1,72 1,95 2,17 2,61 3,05 3,48 4,35 4,98 25,0
10,0 0,515 0,473 0,471 0,521 0,836 1,12 1,39 1,91 2,15 2,40 2,64 3,12 3,58 4,05 4,97 5,64 10,0
5,0 0,725 0,640 0,618 0,671 1,06 1,37 1,65 2,20 2,45 2,71 2,96 3,45 3,94 4,42 5,37 6,06 5,0
1,0 1,32 1,09 1,00 1,05 1,63 1,95 2,25 2,83 3,10 3,38 3,64 4,16 4,67 5,18 6,19 6,91 1,0
0,04 — 0,065 0,10 0,15 0,25 0,40 0,65 — — 1,0 — 1,5 — 2,5 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter N
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 — 0,65 1,0 — — — —
Acceptance quality limit (reduced inspection) in percent – sample size code letter Q

ISO 3951-1:2022(E)
sample size code letter P: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter P Pa
% 0,015 — 0,025 0,04 0,065 0,10 0,15 0,25 — 0,40 — 0,65 — 1,0 — 1,5 %
99,0 0,0041 0,0070 ,0113 ,0158 ,0286 ,0682 0,119 0,243 0,312 0,385 0,461 0,620 0,787 0,960 1,32 1,60 99,0
95,0 0,0108 0,0158 ,0227 ,0301 ,0531 0,112 0,180 0,337 0,421 0,509 0,598 0,783 0,973 1,17 1,57 1,88 95,0
90,0 0,0175 0,0239 0,0323 ,0419 ,0728 0,143 0,223 0,399 0,492 0,588 0,685 0,883 1,09 1,29 1,72 2,04 90,0
75,0 0,0378 0,0461 0,0573 ,0710 0,121 0,215 0,316 0,526 0,634 0,743 0,854 1,08 1,30 1,53 1,99 2,34 75,0
50,0 0,0838 0,0914 0,104 ,124 0,206 0,331 0,456 0,705 0,831 0,956 1,08 1,33 1,58 1,83 2,33 2,70 50,0
25,0 0,175 0,173 0,183 ,210 0,341 0,498 0,647 0,936 1,08 1,22 1,36 1,63 1,91 2,18 2,72 3,12 25,0
10,0 0,323 0,297 0,296 ,328 0,522 0,705 0,873 1,19 1,35 1,50 1,66 1,95 2,25 2,54 3,11 3,53 10,0
5,0 0,456 0,404 0,389 ,423 0,665 0,862 1,04 1,38 1,54 1,70 1,86 2,17 2,47 2,78 3,36 3,80 5,0
1,0 0,836 0,694 0,632 ,668 1,03 1,24 1,42 1,78 1,95 2,12 2,30 2,62 2,94 3,27 3,88 4,35 1,0
0,025 — 0,04 0,065 0,10 0,15 0,25 0,40 — — 0,65 — 1,0 — 1,5 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter P
— 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 — 0,40 0,65 — — — —
Acceptance quality limit (reduced inspection) in percent – sample size code letter R
sample size code letter Q: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter Q Pa
% 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 — 0,40 — 0,65 — 1,0 %
99,0 0,002 7 0,007 3 0,010 2 0,018 4 0,044 0 0,076 7 0,156 0,246 0,295 0,396 0,504 0,614 0,845 1,02 99,0
95,0 0,007 0 0,014 6 0,019 3 0,034 1 0,071 5 0,116 0,216 0,325 0,383 0,500 0,622 0,747 1,00 1,20 95,0
90,0 0,011 3 0,020 7 0,026 9 0,046 7 0,091 9 0,143 0,256 0,376 0,438 0,565 0,695 0,827 1,10 1,30 90,0
75,0 0,024 2 0,036 6 0,045 5 0,077 5 0,138 0,202 0,336 0,476 0,546 0,689 0,833 0,978 1,27 1,49 75,0
50,0 0,053 6 0,066 7 0,079 5 0,132 0,212 0,292 0,451 0,613 0,693 0,853 1,01 1,17 1,49 1,73 50,0
25,0 0,112 0,117 0,135 0,219 0,319 0,415 0,599 0,782 0,872 1,05 1,22 1,40 1,74 2,00 25,0
10,0 0,208 0,190 0,212 0,336 0,453 0,562 0,766 0,968 1,06 1,26 1,44 1,63 2,00 2,27 10,0
5,0 0,294 0,250 0,274 0,430 0,555 0,670 0,883 1,10 1,20 1,40 1,59 1,79 2,16 2,44 5,0
1,0 0,544 0,410 0,435 0,669 0,799 0,920 1,14 1,37 1,48 1,70 1,90 2,11 2,50 2,80 1,0
0,015 0,025 0,04 0,065 0,10 0,15 0,25 — 0,40 — 0,65 — 1,0 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter Q

ISO 3951-1:2022(E)
sample size code letter R: s-method
Pa Acceptance quality limit (normal inspection) in percent – sample size code letter R Pa
% — 0,01 0,015 0,025 0,04 0,065 0,10 0,15 — 0,25 — 0,40 — 0,65 %
99,0 0,007 0 0,004 6 0,006 5 0,011 5 0,027 7 0,048 2 0,097 6 0,312 0,155 0,461 0,249 0,315 0,384 0,528 99,0
95,0 0,015 8 0,009 2 0,012 2 0,021 2 0,044 9 0,072 6 0,135 0,421 0,204 0,598 0,313 0,389 0,467 0,627 95,0
90,0 0,023 9 0,013 0 0,016 8 0,029 0 0,057 6 0,089 7 0,160 0,492 0,235 0,685 0,353 0,434 0,517 0,686 90,0
75,0 0,046 1 0,022 9 0,028 5 0,048 2 0,086 2 0,126 0,210 0,634 0,297 0,854 0,430 0,521 0,611 0,795 75,0
50,0 0,091 4 0,041 6 0,049 6 0,082 4 0,132 0,182 0,282 0,831 0,382 1,08 0,533 0,633 0,732 0,932 50,0
25,0 0,173 0,073 0,084 0,137 0,200 0,259 0,376 1,08 0,489 1,36 0,656 0,766 0,873 1,09 25,0
10,0 0,297 0,119 0,132 0,212 0,284 0,352 0,481 1,35 0,605 1,66 0,786 0,906 1,02 1,25 10,0
5,0 0,404 0,157 0,172 0,272 0,349 0,420 0,556 1,54 0,685 1,86 0,874 0,999 1,12 1,35 5,0
1,0 0,694 0,259 0,274 0,426 0,505 0,579 0,723 1,95 0,861 2,30 1,06 1,20 1,32 1,57 1,0
0,010 0,015 0,025 0,04 0,065 0,10 0,15 — 0,25 — 0,40 — 0,65 —
Acceptance quality limit (tightened inspection) in percent – sample size code letter R

ISO 3951-1:2022(E)
Annex F
(informative)
Consumer’s risk qualities
F.1 The consumer’s risk quality for the s-method plans

For a given sampling plan, the consumer’s risk quality is the process quality for which the probability of
accepting a given lot is 10 %.
For the s–method, the consumer’s risk quality is the solution in p such that the probability of acceptance
is 0,10 i.e.
1 − Ft
(n−1 ,K p n ) ( k n ) = 0 , 10
where
n is the sample size,
k is the s–method acceptance constant,
Kp is the upper p-quantile of the standardized normal distribution

Ft is the distribution function of the non-central t-distribution with n−1 degrees of
(n−1,K p n ) (.)
freedom and non-centrality parameter K p n .
F.2 Tables for the consumer’s risk qualities for the s–method plans
Consumer’s risk qualities for the s–method plans of this document are given in Table F.1, Table F.3, and
Table F.5 for normal, tightened, and reduced inspection, respectively.
F.3 The consumer’s risk quality for the σ–method plans

For the σ–method, the consumer’s risk quality is the solution in p such that the probability of acceptance
is 0,10 and is given by the formula:
1 − Φ k − K 0 ,10 / n  ≡ 1 − Φ k − 1 , 281 6 / n 
   
where
k is the σ–method acceptance constant

Φ (.) is the distribution function of the standardized normal distribution.
F.4 Tables for the consumer’s risk qualities for the σ–method plans
Consumer’s risk qualities for the σ–method plans of this document are given in Table F.2, Table F.4, and
Table F.6 for normal, tightened, and reduced inspection, respectively.

Table F.1 — Consumer’s risk quality (in percent) for normal inspection: s–method
76
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 53,0 52,3 56,4
C 39,5 36,5 39,9 54,1
D 26,4 24,5 27,1 41,4 51,2
E 16,7 15,8 17,7 27,8 36,8 44,8
ISO 3951-1:2022(E)
F 11,7 10,7 11,8 18,7 24,8 30,7 41,4

G 7,37 6,97 7,73 12,2 16,2 20,0 27,6 34,5
H 4,96 4,54 5,01 7,96 10,7 13,1 18,0 22,6 29,4
J 3,11 2,86 3,18 5,09 6,78 8,41 11,5 14,5 18,7 24,3
K 2,01 1,85 2,05 3,27 4,39 5,45 7,46 9,39 12,2 15,8 22,0
L 1,26 1,16 1,29 2,06 2,78 3,43 4,72 5,94 7,71 10,0 13,9
M 0,812 0,743 0,826 1,33 1,77 2,19 3,02 3,79 4,91 6,39 8,91
N 0,515 0,471 0,521 0,830 1,12 1,39 1,91 2,40 3,12 4,05 5,64
P 0,323 0,296 0,328 0,521 0,705 0,873 1,19 1,50 1,95 2,54 3,53
Q 0,207 0,190 0,211 0,336 0,453 0,562 0,766 0,968 1,26 1,63 2,27
R 0,119 0,132 0,209 0,284 0,352 0,481 0,605 0,768 1,02 1,42
If the sample size equals or exceeds the lot size, carry out 100 % inspection. When the sample size exceeds the lot size the consumers’s risk will be greater than that given in the
table.
NOTE The consumer’s risk quality is the process fraction nonconforming at which 10 % of lots are expected to be accepted.

Table F.2 — Consumer’s risk quality (in percent) for normal inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 51,2 52,8 62,7
C 35,4 35,5 40,2 58,1
D 22,2 23,7 27,3 42,2 55,3
E 17,0 15,4 16,7 26,4 37,9 47,2
F 10,2 10,3 11,4 18,1 24,6 32,6 43,0
G 7,59 6,59 7,34 11,8 15,9 20,3 29,1 35,8

H 4,50 4,30 4,85 7,85 10,7 13,2 19,2 23,7 30,2
J 2,76 2,77 3,12 5,07 6,79 8,33 12,0 15,1 19,4 24,9
K 1,98 1,80 2,04 3,18 4,39 5,45 7,54 9,76 12,6 16,2 22,3
L 1,20 1,15 1,25 2,05 2,78 3,43 4,72 6,16 7,95 20,2 14,2
M 0,840 0,738 0,807 1,32 1,78 2,18 3,02 3,94 5,05 6,55 9,06
N 0,510 0,469 0,518 0,821 1,11 1,38 1,91 2,48 3,20 4,13 5,73
P 0,311 0,284 0,317 0,523 0,696 0,865 1,19 1,55 2,00 2,59 3,59
Q 0,193 0,184 0,206 0,330 0,448 0,562 0,768 0,990 1,29 1,66 2,30
R 0,116 0,131 0,204 0,282 0,352 0,480 0,616 0,803 1,04 1,45
table.
77
ISO 3951-1:2022(E)
Table F.3 — Consumer’s risk quality (in percent) for tightened inspection: s–method
78
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 53,0 52,3
C 39,5 36,5 39,9
D 26,4 24,5 27,1 41,4
E 16,7 15,8 17,7 27,8 36,8
ISO 3951-1:2022(E)
F 11,7 10,7 11,8 18,7 24,8 30,7

G 7,37 6,97 7,73 12,2 16,2 20,0 27,6
H 4,96 4,54 5,01 7,96 10,7 13,1 18,0 24,9
J 3,11 2,86 3,18 5,09 6,78 8,41 11,5 15,9 21,5
K 2,01 1,85 2,05 3,27 4,39 5,45 7,46 10,3 14,0 19,3
L 1,26 1,16 1,29 2,06 2,78 3,43 4,72 6,52 8,85 12,2
M 0,812 0,743 0,826 1,33 1,77 2,19 3,02 4,17 5,68 7,85
N 0,515 0,471 0,521 0,830 1,12 1,39 1,91 2,64 3,58 4,97
P 0,323 0,296 0,328 0,521 0,705 0,873 1,19 1,66 2,25 3,11
Q 0,207 0,190 0,211 0,336 0,453 0,562 0,766 1,06 1,44 2,00
R 0,116 0,119 0,132 0,209 0,284 0,352 0,481 0,666 0,906 1,25
table.

Table F.4 — Consumer’s risk quality (in percent) for tightened inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 51,2 52,8
C 35,4 35,5 40,2
D 22,2 23,7 27,3 42,2
E 17,0 15,4 16,7 26,4 37,9
F 10,2 10,3 11,4 18,1 24,6 32,6
G 7,59 6,59 7,34 11,8 15,9 20,3 29,1

H 4,50 4,30 4,85 7,85 10,7 13,2 19,2 26,0
J 2,76 2,77 3,12 5,07 6,79 8,33 12,0 16,7 22,1
K 1,98 1,80 2,04 3,18 4,39 5,45 7,54 10,7 14,4 19,7
L 1,20 1,15 1,25 2,05 2,78 3,43 4,72 6,75 9,10 12,5
M 0,840 0,738 0,807 1,32 1,78 2,18 3,02 4,31 5,83 7,99
N 0,510 0,469 0,518 0,821 1,11 1,38 1,91 2,72 3,67 5,05
P 0,311 0,284 0,317 0,507 0,696 0,865 1,19 1,71 2,30 3,16
Q 0,193 0,184 0,206 0,330 0,448 0,562 0,768 1,10 1,47 2,03
R 0,115 0,116 0,131 0,204 0,282 0,352 0,480 0,684 0,924 1,27
table.
79
ISO 3951-1:2022(E)
Table F.5 — Consumer’s risk quality (in percent) for reduced inspection: s–method
80
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 53,0 49,2 52,3 56,4 61,1
E 39,5 34,1 36,5 39,9 54,1 60,8
F 26,4 23,9 24,5 27,1 41,4 51,2 59,4
G 16,7 15,8 15,8 17,7 27,8 36,8 44,8 55,7
ISO 3951-1:2022(E)
H 11,7 10,5 10,7 11,8 18,7 24,8 30,7 41,4 47,2

J 7,37 6,85 6,97 7,73 12,2 16,2 20,0 27,6 31,0 37,8
K 4,96 4,48 4,54 5,01 7,96 10,7 13,1 18,0 20,3 24,9 29,4
L 3,11 2,84 2,86 3,18 5,09 6,78 8,41 11,5 13,0 15,9 18,7
M 2,01 1,82 1,85 2,05 3,27 4,39 5,45 7,46 8,45 10,3 12,2
N 1,26 1,16 1,16 1,29 2,06 2,78 3,43 4,72 5,31 6,52 7,71
P 0,812 0,745 0,743 0,826 1,33 1,77 2,19 3,02 3,40 4,17 4,91
Q 0,515 0,473 0,471 0,521 0,830 1,12 1,39 1,91 2,15 2,64 3,12
R 0,297 0,296 0,328 0,521 0,705 0,873 1,19 1,35 1,66 1,95
table.

Table F.6 — Consumer’s risk quality (in percent) for reduced inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 51,2 48,5 52,8 62,7 63,3
E 35,4 31,8 35,5 40,2 58,1 62,8
F 22,2 22,9 23,7 27,3 42,2 55,3 61,1
G 17,0 14,5 15,4 16,7 26,4 37,9 47,2 56,8
H 10,2 9,61 10,3 11,4 18,1 24,6 32,6 43,0 48,5
J 7,59 6,71 6,59 7,34 11,8 15,9 20,3 29,1 32,5 39,0

K 4,50 4,30 4,30 4,85 7,85 10,7 13,2 19,2 21,4 26,0 30,2
L 2,76 2,69 2,77 3,12 5,07 6,79 8,33 12,0 13,7 16,7 19,4
M 1,98 1,73 1,80 2,04 3,18 4,39 5,45 7,54 8,86 10,7 12,6
N 1,20 1,18 1,15 1,25 2,05 2,78 3,43 4,72 5,54 6,75 7,95
P 0,840 0,743 0,738 0,807 1,32 1,78 2,18 3,02 3,47 4,31 5,05
Q 0,510 0,467 0,469 0,518 0,821 1,11 1,38 1,91 2,16 2,72 3,20
R 0,292 0,284 0,317 0,523 0,696 0,865 1,19 1,36 1,71 2,00
table.
81
ISO 3951-1:2022(E)
ISO 3951-1:2022(E)
Annex G
(informative)
Producer’s risks
G.1 The producer’s risk for the s-method plans

The producer’s risk is the probability of not accepting a given lot when the process fraction
nonconforming is equal to the AQL, i.e. 1 minus the probability of accepting a given lot when the process
fraction nonconforming is equal to the AQL.
For the s–method, the producer’s risk is given by the following Formula (G.1):
Ft
(n−1,K p n ) ( k n) (G.1)
where
p is the AQL expressed as a fraction nonconforming,
k is the s-method acceptance constant,
Kp is the upper p-quantile of the standard normal distribution

Ft
(n−1,K p n ) (.) is the distribution function of the non-central t-distribution with degrees of freedom
n − 1 and non-centrality parameter nK p .
G.2 Tables for the producer’s risks for the s–method plans
Producer’s risks for the s–method plans of this document are given in Table G.1, Table G.3, and Table G.5
for normal, tightened, and reduced inspection, respectively.
G.3 The producer’s risk for the σ–method plans

For the σ–method, the producer’s risk is given by by the following Formula (G.2):
Φ  n ( k − K p ) (G.2)
where
p is the AQL expressed as a fraction nonconforming,
k is the σ-method acceptance constant,
Kp is the upper p-quantile of the standard normal distribution,
Φ(.) is the distribution function of the standard normal distribution.

ISO 3951-1:2022(E)
G.4 Tables for the producer’s risk for the σ–method plans
Producer’s risks for the σ–method plans of this document are given in Table G.2, Table G.4, and Table G.6
for normal, tightened, and reduced inspection, respectively.

Table G.1 — Producer’s risk (in percent) for normal inspection: s–method
84
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 10,8 7,46 8,93
C 12,2 8,00 10,8 5,82
D 10,8 7,52 10,3 8,74 2,50
E 10,3 6,88 10,4 9,07 4,62 3,18
ISO 3951-1:2022(E)
F 10,8 7,12 8,54 8,14 3,77 3,34 0,908

G 9,81 7,62 9,99 7,49 3,94 3,35 1,45 1,10
H 9,88 6,98 9,99 7,95 3,37 3,07 1,21 1,30 0,853
J 8,91 6,61 9,63 8,64 3,91 2,71 1,26 1,28 1,27 1,13
K 9,16 5,79 9,08 7,54 3,99 3,14 0,891 1,12 1,01 1,48 0,568
L 9,45 6,29 8,16 7,54 3,78 3,51 1,24 0,891 1,08 1,37 1,05
M 9,01 6,54 8,99 6,77 3,51 3,12 1,39 1,19 0,685 1,23 0,787
N 8,76 6,26 9,48 7,30 2,97 2,98 1,20 1,43 1,07 0,803 0,741
P 8,09 6,12 9,15 7,88 3,60 2,55 1,18 1,27 1,42 1,44 0,462
Q 8,47 5,32 8,68 7,20 3,74 2,93 0,806 1,10 1,07 1,66 0,759
R 6,00 7,90 7,07 3,52 3,35 1,14 0,821 1,05 1,42 1,18
If the sample size equals or exceeds the lot size, carry out 100 % inspection. When the sample size exceeds the lot size the producers’s risk will be greater than that given in the
table.
NOTE The producer’s risk is the probability of not accepting a given lot when the process fraction nonconforming is equal to the AQL.

Table G.2 — Producer’s risk (in percent) for normal inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 3,57 2,96 6,72
C 7,17 3,59 6,06 4,54
D 6,33 3,89 6,37 4,81 2,86
E 7,17 3,94 6,29 4,62 2,81 2,74
F 7,65 4,32 5,42 4,66 1,89 2,80 0,865
G 7,44 4,96 6,87 4,66 2,04 2,09 1,41 1,15

H 7,47 4,68 7,35 5,48 1,98 1,86 1,26 1,38 0,871
J 6,69 4,70 7,38 6,40 2,56 1,59 1,05 1,25 1,32 1,24
K 7,32 4,16 7,17 5,56 2,74 2,10 0,572 1,08 1,06 1,58 0,602
L 7,64 4,82 6,30 5,80 2,67 2,48 0,788 0,854 1,09 1,39 1,07
M 7,52 5,16 7,26 5,29 2,56 2,19 0,933 1,17 0,682 1,28 0,829
N 7,30 5,02 7,95 5,82 2,04 2,12 0,844 1,36 1,07 0,808 0,774
P 6,70 4,77 7,55 6,30 2,64 1,82 0,832 1,23 1,42 1,46 0,481
Q 7,06 4,16 7,25 5,85 2,84 2,26 0,578 1,02 1,07 1,69 0,776
R 4,89 6,71 5,76 2,73 2,68 0,830 0,738 1,04 1,43 1,20
table.
85
ISO 3951-1:2022(E)
Table G.3 — Producer’s risk (in percent) for tightened inspection: s–method
86
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 17,6 14,7
C 19,0 16,7 21,7
D 18,1 15,7 22,7 18,6
E 16,2 15,8 22,4 20,7 13,4
ISO 3951-1:2022(E)
F 17,3 13,9 20,6 18,7 13,2 13,0

G 17,2 15,4 19,7 18,8 13,1 15,0 9,41
H 16,9 15,7 20,8 16,8 12,7 13,6 10,2 5,66
J 16,4 14,9 22,1 18,8 11,4 13,8 9,92 7,33 5,18
K 15,1 14,3 20,7 19,0 12,5 11,6 9,13 6,21 6,13 4,11
L 16,1 13,1 20,5 18,4 13,7 13,8 7,85 6,26 5,82 6,19
M 16,5 14,2 18,7 18,1 12,6 14,8 9,52 4,65 5,34 5,00
N 15,9 15,0 20,5 16,3 12,3 13,8 10,7 6,36 3,84 4,76
P 15,7 14,5 21,8 18,2 11,2 13,9 10,1 7,75 6,00 3,38
Q 14,5 13,9 20,5 18,7 12,4 11,5 9,18 6,35 6,64 4,86
R 12,4 12,9 20,4 18,1 13,5 13,8 7,80 6,33 6,01 6,68
table.

Table G.4 — Producer’s risk (in percent) for tightened inspection: σ–method
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B 8,17 7,75
C 13,6 10,2 15,2
D 13,4 10,5 17,2 12,6
E 12,5 11,5 17,4 14,5 9,78
F 14,1 10,2 16,5 13,9 9,18 11,3
G 14,3 12,0 16,1 14,7 9,35 11,8 8,77

H 14,4 12,6 17,7 13,4 9,68 10,6 9,80 5,73
J 14,1 12,3 19,3 15,9 8,94 10,9 8,86 7,29 5,19
K 13,0 12,0 18,3 16,2 10,2 9,37 7,54 5,99 6,15 4,19
L 14,1 11,1 18,2 16,1 11,5 11,7 6,29 6,07 5,79 6,10
M 14,7 12,3 16,7 16,0 10,8 12,7 7,96 4,50 5,31 4,97
N 14,2 13,3 18,7 14,3 10,4 11,9 9,26 6,18 3,79 4,73
P 14,2 12,7 20,1 16,3 9,49 12,1 8,77 7,61 5,94 3,40
Q 13,0 12,3 18,8 17,0 10,7 10,1 8,07 6,26 6,54 4,84
R 11,0 11,4 18,9 16,5 12,0 12,4 6,74 6,24 5,96 6,67
table.
87
ISO 3951-1:2022(E)
Table G.5 — Producer’s risk (in percent) for reduced inspection: s–method
88
letter
0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 3,77 2,70 3,29 3,91 0,257
E 4,92 2,65 3,69 4,52 2,05 0,041
F 4,28 3,18 3,08 4,33 3,41 0,571 0,159
G 3,30 3,37 3,30 3,98 3,67 1,18 0,655 0,017
ISO 3951-1:2022(E)
H 3,38 2,67 3,22 3,80 2,87 0,915 0,544 0,082 0,214

J 2,67 2,70 3,10 4,34 3,23 0,838 0,573 0,112 0,338 0,290
K 2,97 2,28 2,84 3,77 3,20 0,978 0,421 0,092 0,256 0,379 0,853
L 2,81 2,45 2,38 3,69 3,08 1,03 0,557 0,072 0,273 0,370 1,27
M 2,49 2,35 2,53 3,09 2,68 0,849 0,581 0,087 0,184 0,325 1,01
N 2,36 2,26 2,57 3,40 2,34 0,822 0,503 0,107 0,272 0,235 1,08
P 2,09 2,14 2,36 3,56 2,66 0,631 0,457 0,084 0,309 0,333 0,685
Q 2,19 1,89 2,30 3,30 2,65 0,743 0,349 0,074 0,256 0,419 1,07
R 2,09 2,01 3,25 2,49 0,838 0,459 0,052 0,244 0,363 1,42
table.

Table G.6 — Producer’s risk (in percent) for reduced inspection: σ–method
letter 0,01 0,015 0,025 0,04 0,065 0,10 0,15 0,25 0,40 0,65 1,0 1,5 2,5 4,0 6,5 10,0
B-D 0,570 0,519 0,913 2,87 0,367
E 1,80 0,601 1,16 1,88 1,55 0,065
F 1,56 1,18 1,14 2,08 1,42 0,772 0,217
G 1,72 1,35 1,49 1,68 1,27 0,561 0,560 0,021
H 1,57 1,12 1,54 1,90 1,15 0,307 0,439 0,081 0,222

J 1,58 1,46 1,56 2,36 1,59 0,278 2,63 0,116 0,359 0,323
K 1,64 1,21 1,51 2,23 1,80 0,438 0,173 0,105 0,275 0,443 0,871
L 1,55 1,37 1,36 2,36 1,88 0,524 0,234 0,054 0,278 0,418 1,32
M 1,59 1,36 1,55 2,06 1,57 0,447 0,293 0,042 0,185 0,326 1,06
N 1,47 1,53 1,72 2,25 1,49 0,456 0,266 0,047 0,261 0,235 1,09
P 1,46 1,44 1,61 2,50 1,82 0,361 0,240 0,039 0,230 0,327 0,682
Q 1,51 1,26 1,62 2,43 1,84 0,407 0,181 0,038 0,158 0,412 1,07
R 1,44 1,31 2,31 1,67 0,498 0,258 0,026 0,161 0,368 1,42
If the sample size equals or exceeds the lot size, carry out 100 % inspection. When the sample size exceeds the lot size the producers’s risk will
be greater than that given in the table.
89
ISO 3951-1:2022(E)
ISO 3951-1:2022(E)
Annex H
(informative)
Construction of acceptance diagrams for double specification

limits
H.1 General
Earlier issues of this document contained standardized acceptance diagrams for combined control of
double specification limits. A realization that most users would want to have the particular acceptance
curves they need to use in an unstandardized form with the specification limits relevant to their
application, perhaps presented in a document to be agreed with the responsible authority and perhaps
supported by a spreadsheet to calculate results and plot results has led to a decision not to include these
standardized curves. It was decided that a software package to support use of the standard generally
(see Annex I) and guidance on how acceptance diagrams are drawn in this Annex would be provided.
H.2 s-method acceptance diagrams for double specification limits under separate
control
It is instructive to start with an acceptance diagram for double specification limits under separate
control. If the same AQL is applied to both limits then the sample size n and acceptance constant k will
be the same for both limits. The acceptance region is the union of the acceptance regions for the lower
and upper specification limits and can be constructed by drawing the lines
x = L + ks and x = U − ks
from s = 0 to their intersection at s = (U − L ) ( 2k ) . Using the same sampling plan as 15.1 Example 4 i.e.
U = 84 mm, L = 82 mm, n = 5 and k = 0,550 we obtain the diagram below in figure H.1.

ISO 3951-1:2022(E)
Key
x = L + ks 1 reject
x = U − ks 2 accept
Figure H.1 — acceptance region, s-method, separate control
This standard presents sampling plans in two ways: the k method and the p* method with equivalent
acceptance criteria
x −L U−x
Q= ≥ k or pˆL ≤ p * and Q = ≥ k or pÛ ≤ p *
s s
Note that Q is the slope of a line from the specification limits though x , s and k is the slope of the
acceptance boundary lines so the acceptance criteria can be thought of in terms of slopes; if Q is steeper
than k accept otherwise reject.
H.3 s-method acceptance diagrams for double specification limits under

combined control
The equivalence between the k and p* methods is provided by formulae in 8 d) 1)
  1 1 x − L n  
pˆL = F  max 0, −  (H.1)
BETA −1, −1  
n n 
 2 2 s n − 1  
2 2 
  1 1 U − x n  
pÛ = F n n   max 0, −   (H.2)
BETA −1, −1 
 2 2 
  2 2 s n − 1  
If ( x − L ) s is set to k in Formula (H.1) then the equation will yield pˆL = p * similarly if (U − x ) s in
Formula (H.2) is set to k then the equation will yield pÛ = p * . This means that for separate control all
along the lower acceptance boundary pˆL = p * and all along the upper acceptance boundary pÛ = p * .

ISO 3951-1:2022(E)
At the apex pˆL + pÛ = 2 p * but this is alright because each specification limit is under separate control.
The relationship between k and p̂ is obtained by inverting these cases:
n−1 −1  n n  
k ( n, pˆ ) = 1 − 2 FBETA  − 1, − 1  ( pˆ )  (H.3)
n  2 2  
Suppose now it is decided to put both specification limits under combined control and the acceptance
criterion becomes pˆL + pÛ ≤ p * . It is clear that the triangular acceptance region we had under separate
control has to be adjusted to exclude a region where the combined control acceptance criterion is not
met. This is done by using the relationship in Formula (H.3) to construct pairs of lines representing p̂L
and pÛ = p* − pˆL
(
x = L + k ( n, pˆL ) s ≡ L + kL s and x = U − k n, p* − pˆL s ≡ U − kU s )
The intersections of these pairs of lines are points ( s , x ) on the acceptance region boundary with
pˆ = pˆL + pÛ = p* where
U −L U k L + L kU
s= and x =
kU + k L kU + k L
This is illustrated in Figure H.2 where three lines for pˆL = 0, p * /4, p * /2 corresponding to
pÛ = p*, 3 p * /4, p * /2 have been added. Note that the value of s where p* is shared equally between
p̂L and p̂U defines the maximum sample standard deviation (MSSD) that was used in previous issues of
this document in preliminary screening for unacceptable samples.
Key
x = L + ks
x = U − ks
Figure H.2 — acceptance region, s-method, combined control – construction lines

pˆL = 0, p * / 4, p * / 2 and pÛ = p*, 3 p * / 4, p * / 2
The acceptance region for double specification limits under combined control is the union over
0 ≤ pˆL ≤ p * of the triangular regions enclosed by the construction lines and the x axis. The boundary

ISO 3951-1:2022(E)
of this region is the locus over 0 ≤ pˆL ≤ p * of the points ( s , x ) . This is illustrated in Figure H.3 where
the acceptance region boundary has been added. Note that the horizontal scale has been adjusted.
Key
acceptance region
1 reject
boundary
2 accept
Figure H.3 — acceptance region, s-method, combined control with construction lines
pˆL = 0, p * / 4, p * / 2 and pÛ = p*, 3 p * / 4, p * / 2
The acceptance diagram can be completed as shown in Figure H.4 by removing the construction lines
and the sample mean and sample standard deviation for the data in the example can be plotted. The
sample mean and sample standard deviation are within the accept region agreeing with the result of
the numerical procedure in 15.1.
Key
sample 2 accept
Figure H.4 — Acceptance region, s-method, combined control with sample mean and standard
deviation

ISO 3951-1:2022(E)
H.4 σ-method – determination of combined specification limits acceptance

criteria
The first step in setting up any σ–method sampling scheme is to establish a value of the process
standard deviation that can be considered to be known. See 5.3. This is best done by collecting data
over time, maybe whilst using the s-method, and using Statistical Process Control (see References [13]
and [14]) to assess whether this assumption can be justified. If the standard deviation is under control,
i.e. there is no more variation in it than can be explained by sampling variation, then an estimate of the
process standard deviation can be made (see Annex A).
Next, for the use of combined specification limits the question of process capability arises i.e. is the
estimated process standard deviation sufficiently small to allow for the possibility, but not the certainty
of lots drawn from the process being acceptable. This uncertainty arises from the possibility that the
means of lots may be too high or too low to allow acceptance. Table 11 gives values fσ for maximum
process standard deviation (MPSD) to help make this determination e.g. in paragraph 15.2 Example 3
the AQL is 1,5 % and the specification limits are 520±50 Ω . Entering Table 11 with AQL 1,5 % gives fσ
= 0,205 6 and multiplying this by U − L=100 gives 20,56 Ω as a MPSD. Since the known value of sigma
given in the example is 18,5 Ω it is concluded that there is the possibility of lots being acceptable.
Acceptance sampling using the σ–method can therefore begin.
The calculation of the MSSD relevant to the s-method is described earlier and defines the maximum
point on the acceptance region for an s-method combined double specification limit acceptance region.
It should be noted that the calculation of the MPSD is quite different and it does not define the maximum
point on an acceptance diagram in the same way. It will instead usually truncate the acceptance
diagram that could be drawn for the σ-method. The MPSD is calculated on the basis that since the value
of σ is known and the minimum fraction defective that can possibly be obtained from such a process
will be when the mean is midway between the specification limits the process capability can be
assessed by noting whether the area under a standardized normal distribution beyond z = ± (U − L ) / σ
is greater than the AQL. If it is the process may be rejected without sampling. This criterion can be
rewritten as:
 AQL  U − L
reject the process if −Φ −1  ≥
 2  2σ
where Φ −1 is the inverse of the distribution function of the standardized normal distribution. So for
AQL = 1,5 % = 0,015 as a fraction and with U = 1 and L = 0 we get:
1 1
2 , 434 ≥ ⇒ fσ = = 0 , 2056
2σ 2 × 2 , 434
which is the value from Table 11 used above.
It is not necessary to draw an acceptance diagram to determine σ–method acceptance criteria but
may be useful to determine actions to adjust a process where the sample is close to being rejected.
The acceptance region is constructed in the same way as that for the s-method except the equivalence
between the k-method and p*-method is provided by formulae in 8 d) 2):
L−x n 
p̂L = Φ   (H.4)
 σ n − 1 


ISO 3951-1:2022(E)
 x −U n 
p̂U = Φ   (H.5)
 σ n − 1 

and the relationship between k and p* is obtained by inverting these:
n−1 n−1
k ( n, pˆ ) = −Φ −1 ( pˆ ) = K pˆ (H.6)
n n
The triangular acceptance region for separate control, together with three pairs of construction lines
for pˆL = 0, p * /4, p * /2 corresponding to pÛ = p*, 3 p * /4, p * /2 is shown Figure H.5.
Key
x = L + ks
x = U − ks
Figure H.5 — acceptance region, σ -method, combined control – construction lines

pˆL = 0, p * / 4, p * / 2 and pÛ = p*, 3 p * / 4, p * / 2
Although in principle the normal distribution has an infinite interval (-∞,∞) in practice it should be
limited in computer arithmetic to approximately (-8,29,8,29) where 8,29 is the point at which the
distribution function of the standardized normal distribution becomes 1 and this should be used when
the formulae use p̂ = 0 .
As for the s-method the acceptance region is the union over 0 ≤ pˆL ≤ p * of the triangular regions
enclosed by the construction lines and the x axis. The acceptance region boundary is the locus over
0 ≤ pˆL ≤ p * of the points (σ , x ) . For both the acceptance region and its boundary points that are
beyond the MPSD are excluded. This is illustrated in Figure H.6 where the acceptance region boundary
is shown with the points outside and in Figure H.7 where the acceptance region boundary is shown
with these points excluded.

ISO 3951-1:2022(E)
Key
2 accept
Figure H.6 — Acceptance region MPSD included, σ -method, combined control with construction
lines pˆL = 0, p * / 4, p * / 2 and pÛ = p*, 3 p * / 4, p * / 2
Key
acceptance region boundary x = L + kL s 1 reject
x = U + kU s 2 accept
Figure H.7 — acceptance region MPSD excluded, σ -method, combined control with construction
lines pˆL = 0, p * / 4, p * / 2 and pÛ = p*, 3 p * / 4, p * / 2
The acceptance diagram can be completed as shown in Figure H.8 by removing the construction lines
and plotting the sample mean and known standard deviation for the data in the example. The sample
mean and known standard deviation are within the accept region agreeing with the result of the
numerical procedure in 15.2. However, there is value in a graphical approach in that is readily seen that
the sample mean of the lot under consideration is low and the known value of sigma is not far short of

ISO 3951-1:2022(E)
the process capability limit MPSD. This shows that an effort to adjust the mean and reduce the process
standard deviation might be rewarded.
Key
sample 2 accept
Figure H.8 — acceptance region, σ-method, combined control with sample mean and standard
deviation
With this particular sampling plan we can also illustrate the accuracy of the approximate method, in
7 c) 3), that uses separate limits. The known value of σ and the separate limits are shown in Figure H.9.
It can be seen that this is a reasonable assumption. There some sampling plans in the tables where
truncation does not occur e.g. reduced inspection and small AQLs. If σ > 0 , 75σ max and either QL or Qu
is close to k the sample values (σ , x ) have to plotted on the acceptance region to make the correct
acceptance decision.

ISO 3951-1:2022(E)
Key
acceptance region boundary U-kσ 1 reject
sample L-kσ 2 accept
σ
Figure H.9 — acceptance region, σ-method, combined control with sample mean and standard
deviation and separate limits
H.5 standardized acceptance diagrams for double specification limits under

separate control
Standardized acceptance diagrams for the s-method are constructed by following the process in H.3
s x −L
using the standardized variables ( s S , x S ) where sS = and xS = . For the example above this
U −L U −L
produces the standardized acceptance region shown in Figure H.10.

ISO 3951-1:2022(E)
Key
sample 2 accept
Figure H.10 — standardized acceptance region, s-method, combined control with sample mean
and standard deviation
Standardized acceptance diagrams for the s-method are constructed by following the process in H.4
σ x −L
using the standardized variables (σ S , x S ) where σ S = and xS = . For the example above
U −L U −L
this produces the standardized acceptance region shown in Figure H.11.
Key
acceptance region
U-kσ 1 reject
boundary
sample L-kσ 2 accept
σ
Figure H.11 — standardized acceptance region, σ-method, combined control with sample mean
and standard deviation and separate limits σS = 0,185

ISO 3951-1:2022(E)
H.6 Spreadsheet implementation

Whilst the software that supports the use of this standard is an R package (see Annex I), most
spreadsheets have functions that enable acceptance diagrams to be constructed. To illustrate this we
give the Microsoft Excel formulae for (H.3) and (H.6).
(H.3) corresponds to (1-2*BETA.INV(phat;n/2-1;n/2-1)*(n-1)/SQRT(n)
where phat and n are cells containing these numbers. With n = 5 and p̂ = 0,307 4/2 i.e. p*/2 we obtain
k = 1,034 to 3 significant figures.
The MSSD for the sample in 15.1 Example 4 can be calculated by dividing twice this value into (U − L),
which for the standardized limits (0,1) gives 0,484 to 3 significant figures.
(H.3) corresponds to -NORM.INV(phat;0;1)*SQRT((n-1)/n))
Where phat and n are cells containing these numbers. With n = 19 and p̂ = 0,021 2 i.e. we obtain k =
1,975 to 3 significant figures.
The maximum intersection point on the construction lines for combined control for the sample in 15.2
Example 3 can be calculated by dividing twice this value into (U − L), which for the standardized limits
(0,1) equals 25,31 Ω to 3 significant figures matching the value in Figure H.5 but as noted the acceptance
region is truncated at MPSD = 20,56 Ω.

ISO 3951-1:2022(E)
Annex I
(informative)
Use of the underlying software
This annex describes the use of the software that supports the use of this standard. The software is
informative and therefore not a requirement for the use of this standard. The software is available on
https://standards.iso.org/iso/3951/-1/ed-3/en/.
The user should execute the steps from the document itself, get confidence intervals for the lot fraction
nonconforming, plot and export OC-functions. Furthermore, the consumer’s risk quality (CRQ) and the
producer’s risk (PR) can be calculated for all the sampling plans presented in this standard. For the case
with two specification limits and unknown σ, the software can calculate the acceptance curve for a
given sampling plan. All examples given in this annex refer to the Examples in Clause 15.
Therefore, an R (a language and environment for statistical computing) package to implement this
document is presented.
I.1 Execution of the standard

The function ISO3951_1 provides the decision on whether a lot with given measured values x i can be
accepted or not.
Accordingly, the function calculates at first the sample mean and the sample standard deviation from
the measured values and then the acceptance curve for the case with two specification limits and
unknown σ and the quality statistic QL and/or QU for the other three cases. Afterwards the acceptance
curve and the corresponding point ( x , s ) are plotted or a result of the following kind is given:
"The quality statistic Q = 1,616 9 is greater or equal then the acceptance constant k = 1,426, therefore
the lot is accepted."
Table I.1 — Function ISO3951_1

ISO3951_1(n, k, L = NA, U = NA, sMethod, x, s=NA, sigma = NA, f_sigma = NA)
Arguments
n Sample size.
k Acceptance constant.
L Lower specification limit. Default value: NA for no lower limit.
U Upper specification limit. Default value: NA for no upper limit.
sMethod Logical value indicating whether the standard deviation is unknown (True) or
stable and known (False).
x Vector with all measured values. If the sample mean x and the sample standard
deviation s have been calculated already, simply set x = x and s = s (see below).
s Sample standard deviation, if already calculated. Default value: NA to calculate s
from the measurements x.
sigma Known standard deviation (only for the method with known standard devia-
tion).
f_sigma Value for the maximum process standard deviation (only for the method with
known standard deviation).
Output
Text prompt Whether the lot can be accepted and the relevant calculated values

ISO 3951-1:2022(E)
Table I.1 (continued)

Values List with:
the sample size, n;
the acceptance constant, k;
the quality statistics (for all cases except the cases with unknown σ and two
spec. limits);
the estimated and the maximum fraction nonconforming; and
the calculated value of 𝜎max (for the case with known σmax and two spec. limits).
Examples
s-Method, lower specification ISO3951_1(n=28, k=2.580, L=4, sMethod=T, x=c(6.95,6.04,6.68,6.63,6.65,6.52,6
limit .59,6.40, 6.44, 6.34, 6.04, 6.15,6.29, 6.63, 6.44, 7.15,6.7,6.59,6.51, 6.8, 5.94, 6.35,
7.17, 6.83, 6.25, 6.96, 7.0, 6.38))
s-Method, upper specifica- ISO3951_1(n=13, k=1.426, U=60, sMethod=T, x=c(53, 57, 49, 58, 59, 54, 58, 56, 50,
tion limit 50, 55, 54, 57))
s-Method, two specification ISO3951_1(n=5, k=0.550, L=82, U=84, sMethod=T, x=c(82.4,82.2,83.1,82.3))
limits
σ-Method, lower specifica- ISO3951_1(n=11, k=2.046, L=400, sMethod=F, x=c(431, 417, 469, 407, 450, 452,
tion limit 427, 411, 429, 420, 400), sigma=21)
σ-Method, upper specifica- ISO3951_1(n=51, k=2.337, U=20, sMethod=F, x=c(15,8), sigma=1,42)
tion limit
σ-Method, two specification ISO3951_1(n=19, k=1.677, L=470, U=570, sMethod=F, x=c(515, 491, 479, 513, 521,
limits 536, 483, 509, 514, 507, 484, 526, 532, 499, 530, 512, 492, 522, 488), sigma=18,5,
f_sigma=0.205 6)
I.2 Operating characteristic functions

Annex E gives some tabulated values of the OC function for single sampling plans with unknown σ.
Since the amount of space in this standard is limited and there are four different OC functions (known
and unknown ó , one and two specification limits), it is not possible to present all OC charts and values.
Since this standard uses only the sampling plans for one specification limit, one needs to set only
one specification limit in both cases, since the OC functions for two spec. limits are not used in this
document.
The function OC calculates values of the corresponding OC for given process qualities p . Furthermore,
the OC chart is plotted and can be exported to the working directory in form of a picture or a pdf-file.
Table I.2 — Function OC

OC(n, k, sMethod, L=NA, U=NA, sigma=NA, p = c(0.99, 0.95, 0.9, 0.75, 0.5, 0.25,
0.1, 0.05, 0.01), output = "no_output", plot=T)
Arguments
n Sample size.
sigma Known process standard deviation (for the sigma-method).
p Vector of probabilities to calculate the OC-values. Default value: c(0.99, 0.95, 0.9,
0.75, 0.5, 0.25, 0.1, 0.05, 0.01).

ISO 3951-1:2022(E)

Output Specify a filetype to save a plot of the OC in your working directory, possibili-
ties: "PDF", "JPEG", "PNG". Default value: "no_output".
Output
Values Matrix with the two columns p and OC(p), calculated from the given values of p.
Files Plot of the OC as a PDF, JPEG or PNG, if demanded.
Examples
s-Method, lower specification OC(n=28, k=2.580, L=4, sMethod=T, output="PDF")
limit
s-Method, upper specification OC(n=13, k=1.426, U=60, sMethod=T, output="JPEG")
limit
s-Method, two specification OC(n=5, k=0.550, L=82, U=84, sMethod=T, output="PNG")
limits
σ-Method, lower specification OC(n=11, k=2.046, L=400, sMethod=F, sigma=21, p=seq(0,1, by=0,1))
limit
σ-Method, upper specification OC(n=51, k=2.337, U=20, sMethod=F, sigma=1,42)
limit
σ-Method, two specification OC(n=19, k=1.677, L=470, U=570, sMethod=F, sigma=18.5)
limits
I.3 Confidence intervals

The functions confint calculates one-sided and two-sided confidence interval for the fraction
nonconforming based on the Lieberman-Resnikoff OC functions, which means that the fraction
nonconforming pL, pU, or pL,U are overlapped by the given confidence interval with the probability γ. The
confidence intervals are only informative and have no normative meaning.
Since the intervals depend on the sample mean and the specification limits, it is not feasible to present
the boundaries of the intervals in tables in this document.
Table I.3 — Function confint

confint(n, k, x_bar, s, L=NA, U=NA, sMethod, gamma=0.95)
Arguments
n Sample size.
x_bar Sample mean.
s Sample standard deviation or the known process standard deviation.
gamma Confidence level. Default value: 95 %.
Output List with the three intervals with a lower boundary, an upper boundary and
two (non-trivial) boundaries.
Examples
s-Method, upper specification confint(n=13, k=1.426, x_bar=54.615 38, s=3.330 127, U=60, sMethod=T)
limit
s-Method, two specification confint(n=3, k=0.95, x_bar=3.5, s=7.435 725, L=-10, U=10, sMethod=T)
limits
σ-Method, lower specification confint(n=11, k=2.046, x_bar=428.454 5, s=21, L=400, sMethod=F)
limit

ISO 3951-1:2022(E)

σ-Method, two specification confint (n=19, k=1.677, x_bar=508.052 600, s=18.5, L=470, U=570, sMethod=F)
limits
I.4 Consumer’s risk quality and the producer’s risk

The following definitions of the CRQ and the PR, along with some tabulated values are given in the
Annexes E and F:
“For a given sampling plan, the consumer’s risk quality is the process quality for which the probability
of accepting a given lot is 10 %.”
“The producer’s risk is the probability of not accepting a given lot when the process fraction
nonconforming is equal to the AQL, i.e. 1 minus the probability of accepting a given lot when the process
fraction nonconforming is equal to the AQL.”
The function risk calculates the CRQ and the PR based on the corresponding operating characteristic
Function for a given sampling plan.
Since this document uses only the sampling plans for one specification limit, one needs to set only one
specification limit in both cases, since the sampling processes with one and two specification limits
have the same CRQ and PR.
Table I.4 — Function risk

risk(n, k, sMethod, sigma=NA, L = NA, U = NA, aql, lq = 0.1)
Arguments
n Sample size.
sigma Known process standard deviation (for the case with known sigma and two
specification limits).
aql Acceptance quality limit (in fraction nonconforming).
lq Limiting quality. Default value: 0.1 (for all sampling plans in this document).
Output
Values List with the values of the Consumer’s Risk Quality and the Producer’s Risk.
Examples
s-method, normal insp. with risk(n=4, k=0.586, sMethod=T, L=0, aql=0.1)
AQL = 10 % and code letter B
σ-method, normal insp. with risk(n=3, k=1.115, sMethod=F, L=0, aql=0.025)
AQL = 2,5 % and code letter C
σ -method, normal inspection risk(n=19, k=1.677, sMethod=F, U=570, aql=0.015, sigma=18.5)
with AQL = 1,5 % and code
letter J
I.5 Constructing the acceptance curve for the case with two specification limits
and the s-method
For the case with two specification limits and unknown σ one can only give the boundary of acceptance
region to check whether a lot can be accepted. A lot with sample mean x and sample standard deviation
s is therefore accepted if the point ( x , s ) lies under the acceptance curve.

ISO 3951-1:2022(E)
The function accCurve calculates the acceptance region for two specification limits and unknown
standard deviation for a given sampling plan (n, k).
Table I.5 — Function accCurve

accCurve(n, k, L = 0, U = 1, compl = T, onlyC = F, nPoints = 10^4, plot = T)
Arguments
n Sample size.
L Lower specification limit. Default value: 0 to get a curve that can be easily
transformed to fit any specification limit.
U Upper specification limit. Default value: 1 to get a curve that can be easily
transformed to fit any specification limit.
Compl Logical value indicating whether the whole or the half curve (with L<=
x<=(L+U)/2) should be constructed.
onlyC Logical value indicating whether only the non-linear part of the curve
(L+delta<=x<=U-delta) should be constructed. Can be combined with compl
(set compl=F, onlyC=T) to get only the half of the nonlinear part (L+delta<=x-
<=(L+U)/2).
nPoints Number of support points for the curve and the lines.
Plot Logical value indicating whether the curve should be plotted.
Output
Values Matrix with the x- and y-values of the acceptance curve.
Examples
Curve for the code letter E, aql m <- accCurve(n=13, k=1.475, compl=F)
= 2.5 % and normal inspec-
tion:
The curve with x_bar ranging m <- accCurve(n=186, k=2.031, L=5, U=10)
from L to U for the code letter
N, aql = 1 % and normal in-
spection:

ISO 3951-1:2022(E)
Bibliography
[1] Automotive Industry Action Group. Measurement Systems Analysis — MSA. AIAG, Michigan,
Fourth Edition, 2010
[2] BOTHE D.R., 1997. Measuring process capability, New York: Irwin – McGraw-Hill.
[3] COCHRAN W.G., 2007. Sampling Techniques. 3rd Edition. New York: Wiley.
[4] DUNCAN A. J., 1986. Quality Control and Industrial Statistics, New York: Irwin – McGraw-Hill.
[5] GÖB R. 2001. Methodological Foundations of Statistical Lot Inspection, pp. 3-24, In: Lenz, H.J.
and Wilrich, P.-Th. [Editors], Frontiers in Statistical Quality Control 6, Physica-Verlag. Heidelberg:
New York.
[6] ISO 80000-2:2019, Quantities and units — Part 2: Mathematics
[7] ISO 2854, Statistical interpretation of data — Techniques of estimation and tests relating to means
and variances
[8] ISO 28590:2017, Sampling procedures for inspection by attributes — Introduction to the ISO 2859
series of standards for sampling for inspection by attributes
[9] ISO 5479:1997, Statistical interpretation of data — Tests for departure from the normal distribution
[10] ISO 16269-3, Guide to statistical interpretation of data — Part 3: Tests for departure from the
normal distribution
[11] ISO 16269-4:2010, Statistical interpretation of data — Part 4: Detection and treatment of outliers
[12] ISO 5725-2, Accuracy (trueness and precision) of measurement methods and results — Part 2:
Basic method for the determination of repeatability and reproducibility of a standard measurement
method
[13] ISO 7870-1:2019, Control charts — Part 1: General guidelines
[14] ISO 7870-2:2013, Control charts — Part 2: Shewhart control charts
[15] ISO 10576-1:2003, Statistical methods — Guidelines for the evaluation of conformity with specified
requirements — Part 1: General principles
[16] ISO 22514-2:2017, Statistical methods in process management — Capability and performance —
Part 2: Process capability and performance of time-dependent process models
[17] Chan Tony F., Lewis John Gregg, Computing Standard Deviations: Accuracy. Communications
of the ACM, September 1979, Volume 22, Number 9, pp 526-531.
[18] LIEBERMAN G. J. RESNIKOFF G. J., Sampling plans for inspection by variables. In: Journal of the
American Statistical Association 50. 1955, pp. 457-516.
[19] MELGAARD H. THYREGOD P. Acceptance sampling by variables under measurement uncertainty,
2001, pp. 47-57, In: Lenz, H,J, and Wilrich, P,-Th, pp. 47-57. Frontiers in Statistical Quality Control
6, Physica-Verlag, Heidelberg: New York.
[20] MONTGOMERY D.C., Introduction to Statistical Quality Control, 6th Edition. 2009, New York: John
Wiley & Sons.
[21] RESNIKOFF G.J. LIEBERMAN G.J., Tables of the Non-Central t-Distribution. 1966, California:
Stanford University Press.

ISO 3951-1:2022(E)
[22] SCHILLING E.G., NEUBAUER, D.V., Acceptance sampling in quality control, 2nd Edition. 2009, USA:
CRC Press.
[23] STEPHENS K. S., The handbook of applied acceptance sampling. 2001, Milwaukee: ASQ Quality
Press.
[24] WILRICH P.-TH, Single sampling plans for inspection by variables in the presence of measurement
error. All. Stat. Arch. 2000, pp. 239–250
[25] R Foundation for Statistical Computing [online database]. Available from http://www.R-project
.org [viewed 17 September 2020].

ISO 3951-1:2022(E)
ICS 03.120.30
Price based on 107 pages

International Standard: Third Edition 2022-08

Uploaded by

Copyright:

Available Formats

International Standard: Third Edition 2022-08

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

International Standard: Third Edition 2022-08

Uploaded by

Copyright:

Available Formats

INTERNATIONAL ISO

Sampling procedures for inspection by

Partie 1: Spécification pour les plans d’échantillonnage simples

COPYRIGHT PROTECTED DOCUMENT

ii © ISO 2022 – All rights reserved

© ISO 2022 – All rights reserved iii

Annex B (informative) Accommodating measurement variability..................................................................................58

iv © ISO 2022 – All rights reserved

© ISO 2022 – All rights reserved v

vi © ISO 2022 – All rights reserved

© ISO 2022 – All rights reserved vii

Sampling procedures for inspection by variables —

© ISO 2022 – All rights reserved 1

3 Terms and definitions

Note 2 to entry: See 5.4.

2 © ISO 2022 – All rights reserved

© ISO 2022 – All rights reserved 3

Note 2 to entry: See Clause 6 and Clause 7.

4 © ISO 2022 – All rights reserved

Note 2 to entry: See Clause 4, Clause 6, and Clause 7.

Note 2 to entry: See Clause 4, Clause 6, and Clause 7.

Note 2 to entry: See 5.2.

Note 2 to entry: See Clause 9.

© ISO 2022 – All rights reserved 5

L lower specification limit (as a subscript to a variable, it denotes its value at L)

M unknown process mean

N lot size (number of items in a lot)

n sample size (number of items in a sample)

QL lower quality statistic

QU upper quality statistic

6 © ISO 2022 – All rights reserved

σ process standard deviation

U upper specification limit (as a suffix to a variable, it denotes its value at U)

5 Choice of a sampling plan

5.1 Choice between variables and attributes

© ISO 2022 – All rights reserved 7

5.3 Choice between the s-method and σ-method

8 © ISO 2022 – All rights reserved

5.4 Choice of inspection level and AQL

© ISO 2022 – All rights reserved 9

Table 1 — Sample size code letters and inspection levels

6 Standard procedures for the s-method

6.2 Single specification limits

c) Calculate the quality statistic QU = (U − x )/ s or QL = ( x − L )/ s and compare it with the acceptance

10 © ISO 2022 – All rights reserved

6.3 Double specification limits

7 Standard procedures for the σ-method

7.2 Single specification limits

c) Calculate the quality statistic QU = (U − x ) σ or QL = ( x − L ) σ and compare it with the acceptance

© ISO 2022 – All rights reserved 11

7.3 Double specification limits

c) Estimated fraction nonconforming:

12 © ISO 2022 – All rights reserved

9 Switching between inspection severities

© ISO 2022 – All rights reserved 13

Figure 1 — Outline of the switching rules

9.1 Rules for switching between inspection severities

ii) production is in statistical control (see 5.3), and