ISO 10576:2022

Statistical methods - Guidelines for the evaluation of conformity with specified requirements

This document sets out guidelines for checking conformity with quantifiable characteristics using the test or measurement result and its associated measurement uncertainty. This document is applicable whenever the uncertainty may be quantified according to the principles laid down in ISO/IEC Guide‑98‑3 (GUM). The term uncertainty is thus a descriptor for all elements of variation in the measurement result, including uncertainty due to sampling. This document does not give rules for how to act when an inconclusive result of a conformity test has been obtained. NOTE There are not limitations on the nature of the entity subject to the requirements nor on the quantifiable characteristic. Examples of entities together with quantifiable characteristics are given in Table A.1.

Méthodes statistiques — Lignes directrices pour l'évaluation de la conformité à des exigences spécifiques

General Information

• Status: Published
• Publication Date: 04-Aug-2022
• ICS: 03.120.30 - Application of statistical methods
• Technical Committee: ISO/TC 69/SC 6 - Measurement methods and results
• Drafting Committee: ISO/TC 69/SC 6/WG 7 - Statistical methods to support measurement uncertainty evaluation
• Current Stage: 6060 - International Standard published
• Start Date: 05-Aug-2022
• Due Date: 26-Jun-2022
• Completion Date: 05-Aug-2022

Relations

• Consolidated By: ISO 21660-2:2025 - Solid recovered fuels - Determination of moisture content using the oven dry method - Part 2: Determination of total moisture by a simplified method
• Effective Date: 04-Nov-2023
• Revises: ISO 10576-1:2003 - Statistical methods - Guidelines for the evaluation of conformity with specified requirements - Part 1: General principles
• Effective Date: 23-Apr-2020

Overview

ISO 10576:2022 - "Statistical methods - Guidelines for the evaluation of conformity with specified requirements" provides guidance for deciding whether a measured, quantifiable characteristic meets specified limits (limiting values) when the measurement result carries uncertainty. The standard applies whenever measurement uncertainty can be quantified following ISO/IEC Guide‑98‑3 (GUM). It focuses on using the measurement result plus its associated uncertainty to assess conformity and documents recommended reporting and assessment practices. The standard adopts a frequentist perspective and replaces the earlier ISO 10576‑1:2003.

Key topics and requirements

• Scope and definitions: Clear definitions for terms such as limiting values, region of permissible/non‑permissible values, uncertainty interval and confidence level (see ISO 3534 references).
• Specification of limiting values: Recommendations for unambiguous description of the entity and the quantifiable characteristic, requirement for validated or standardized test procedures, and guidance on single (upper or lower) and double limits.
• Measurement uncertainty: Requirement to quantify uncertainty according to the GUM principles; uncertainty is treated as a descriptor of all variation sources, including sampling.
• Conformity assessment procedures:
  • One‑stage conformity assessment - assess the measurement result plus uncertainty against limiting values in a single test.
  • Two‑stage conformity assessment - an initial stage followed by a second stage to reduce the risk of erroneous decisions (used to balance consumer/supplier risk).
  • Use of uncertainty intervals (including confidence intervals) to derive decisions.
• Reporting: How to report measurement results, uncertainty, assurance of conformity or non‑conformity, and handling of inconclusive results (note: the standard does not prescribe actions to take when results are inconclusive).
• Examples and illustrative material: Annex A lists example entities and characteristics; Annex B provides worked examples (updated to include lab‑to‑lab variability).

Applications and users

ISO 10576:2022 is practical for organizations and professionals who must declare or verify conformity in the presence of measurement uncertainty:

• Testing and calibration laboratories
• Quality control and production engineers
• Conformity assessment bodies and third‑party auditors
• Regulatory agencies and environmental/health monitoring programs
• Manufacturers (chemicals, food, pharmaceuticals, materials, oil, etc.)

Use cases include comparing test results to specification limits, generating supplier declarations of conformity, and designing assessment procedures that account for measurement and sampling uncertainty.

Related standards

• ISO/IEC Guide‑98‑3 (GUM) - Guide to the expression of uncertainty in measurement (required)
• ISO 3534‑1 / 3534‑2 - Statistics vocabulary and symbols
• ISO 5725‑1 to ‑6 - Accuracy (bias and precision) of measurement methods and results
• ISO/IEC Guide 98‑4 - Bayesian perspective on uncertainty (complementary)
• ISO/TR 13587 - Comparison of frequentist, Bayesian and fiducial approaches

Keywords: ISO 10576:2022, conformity assessment, measurement uncertainty, GUM, confidence interval, limiting values, statistical methods, one‑stage assessment, two‑stage assessment, specification limits, quality control.

Standards Content (Sample)

INTERNATIONAL ISO
STANDARD 10576
First edition
2022-08
Statistical methods — Guidelines for
the evaluation of conformity with
specified requirements
Méthodes statistiques — Lignes directrices pour l'évaluation de la
conformité à des exigences spécifiques
Reference number
© 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: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Specification of recommendations .2
4.1 Recommendations for definition of limiting values . 2
4.2 Reporting of limiting values . 3
5 Uncertainty of results . .4
5.1 General . 4
5.2 Reporting the measurement results and the measurement uncertainty . 4
6 Assessing conformity to requirements . 4
6.1 General . 4
6.2 The two-stage conformity assessment . 5
6.2.1 Stage 1 . 5
6.2.2 Stage 2 . 5
6.3 The one-stage conformity assessment . 5
6.4 The uncertainty interval given in the form of a confidence interval . 6
6.5 Inconclusive result of the conformity assessment . 6
7 Reporting the result of the conformity assessment . 8
7.1 General . 8
7.2 Assurance of conformity . 8
7.3 Assurance of non-conformity . 8
7.4 Inconclusive result . 8
Annex A (informative) Examples of entities and quantifiable characteristics .9
Annex B (informative) Examples .10
Bibliography .14
iii
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 6, Measurement methods and results.
This first edition of ISO 10576 cancels and replaces ISO 10576-1:2003, which has been technically
revised.
The main changes are as follows:
— examples were updated to incorporate lab-to-lab variability in the uncertainty calculations, see
Annex B.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.
iv
Introduction
Conformity assessment is a systematic examination of the extent to which an entity conforms to a
specified criterion. The objective is to provide assurance of conformity, either in the form of a supplier’s
declaration, or of a third-party certification (see ISO/IEC Guide 2, 2004). A specification is usually
formulated as a single limiting value, LV, or as a set of (upper and lower) limiting values for a measurable
characteristic. When the specification refers, e.g. to health-related characteristics, the limiting values
are sometimes termed threshold limit value TLV, or permissible exposure limits, PEL.
Whenever conformity assessment involves measurement or sampling uncertainty, it is common practice
to invoke elements from the theory of statistical hypothesis testing to provide a formal procedure. With
the knowledge of the measurement procedure and of its behaviour with regard to the uncertainty of its
outcomes, it is possible to estimate and control the risk of making erroneous declarations of conformity
or non-conformity to the specifications. An operational way of formulating requirements of assurance
is to require that the risk of (erroneously) declaring a non-conforming entity to be conforming should
be small. Consequently, it is necessary to tolerate a (large) risk that an entity, which only marginally
conforms, will fail to be declared as conforming. Applying a two-stage procedure instead of a one-stage
procedure can decrease this risk.
When a test for non-conformity is performed, similar considerations apply.
In this document, this issue is addressed in respect of the testing of output from production or service
processes for conformity and non-conformity with specifications.
Because of the apparent similarity to acceptance sampling procedures, it is sometimes seen that
acceptance sampling plans are used in conformity assessment activities. Acceptance sampling and
[2]
conformity assessment activities both utilize elements of hypothesis testing (see e.g. ISO 2854 ). It
is, however, important to realise that the objectives of the two activities are fundamentally different
[2]
and in particular the two activities imply different approaches to the risk involved (see ISO 2854 and
[9]
Holst ).
This document examines conformity assessment from a frequentist perspective. ISO/IEC Guide 98-4
examines conformity assessment from a Bayesian perspective. A comparison of these two approaches
plus the fiducial approach is given in ISO/TR 13587.
v
INTERNATIONAL STANDARD ISO 10576:2022(E)
Statistical methods — Guidelines for the evaluation of
conformity with specified requirements
1 Scope
This document sets out guidelines for checking conformity with quantifiable characteristics using the
test or measurement result and its associated measurement uncertainty.
This document is applicable whenever the uncertainty may be quantified according to the principles
laid down in ISO/IEC Guide-98-3 (GUM). The term uncertainty is thus a descriptor for all elements of
variation in the measurement result, including uncertainty due to sampling.
This document does not give rules for how to act when an inconclusive result of a conformity test has
been obtained.
NOTE There are not limitations on the nature of the entity subject to the requirements nor on the quantifiable
characteristic. Examples of entities together with quantifiable characteristics are given in Table A.1.
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 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
3 Terms and definitions
For the purposes of this document, the terms and definitions given in ISO 3534-1, ISO 3534-2 and the
following apply.
ISO and IEC maintain terminology 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
limiting values
specification limits
L
specified values of the characteristic giving upper and/or lower bounds of the permissible values
[SOURCE: ISO 3534-2:2006, 3.1.5]
3.2
lower specification limit
L
SL
lower bound of the permissible values of the characteristic
3.3
upper specification limit
U
SL
upper bound of the permissible values of the characteristic
3.4
conformity assessment
systematic evaluation by means of testing of the extent to which a product, process or service does or
does not fulfil specified requirements
3.5
region of permissible values
interval or intervals of all permissible values of the characteristic
Note 1 to entry: Unless otherwise stated in the specification, the limiting values belong to the region of
permissible values.
3.6
region of non-permissible values
interval or intervals of all values of the characteristic that are not permissible
Note 1 to entry: Figure 1 displays various possibilities for the partitioning of the region of possible values of the
characteristic in regions of permissible and non-permissible values.
3.7
uncertainty interval
interval derived from the actual measurement of the characteristic and its uncertainty, covering the
values that could reasonably be attributed to this characteristic at a given probability
Note 1 to entry: An uncertainty interval may be the symmetric interval around the measurement result as
defined in ISO/IEC Guide 98-3:2008, 6.2.1.
Note 2 to entry: When the uncertainty has been obtained only by Type A evaluations of uncertainty components,
the uncertainty interval may be in the form of a confidence interval for the value of the characteristic (see e.g.
ISO 3534-1:2006, 2.57 and ISO/IEC Guide 98-3:2008, G.3).
3.8
two-sided confidence interval
when T and T are two functions of the observed values such that, θ being a population parameter to
1 2
be estimated, the probability P (T ≤ θ ≤ T ) is at least equal to (1 − α) [where (1 − α) is a fixed number,
r 1 2
positive and less than 1], the interval between T and T is a two-sided (1 − α) confidence interval for θ
1 2
[SOURCE: ISO/IEC Guide 98-3:2008, C.2.27]
3.9
confidence level
the value (1 − α) of the probability associated with a confidence interval or a statistical coverage interval
[SOURCE: ISO/IEC Guide 98-3:2008, C.2.29]
4 Specification of recommendations
4.1 Recommendations for definition of limiting values
4.1.1 The entity should be clearly and unambiguously specified.
4.1.2 The quantifiable characteristic of the entity should be clearly and unambiguously specified.
The value of the characteristic should be determined by means of a measurement or test procedure
that enables an assessment of the measurement uncertainty to be made. Examples of entities and their
quantifiable characteristics are given in Annex A.
4.1.3 The measurement or test procedure should be a standardized or validated procedure.
4.1.4 The measurement uncertainty should neither explicitly nor implicitly be referred to in the
designation of the limiting values.

NOTE R denotes region of permissible values while R denotes region of non-permissible values.
PV NV
The specification limits are denoted L, L , U , L and L .
SL SL 1 2
Figure 1 — Division of the domain for the characteristic
4.2 Reporting of limiting values
The reporting of limiting values should be the result of the description given in 4.1.1 and 4.1.2.
The range of permissible values of a quantifiable characteristic may be limited to only one side or to
both sides. Limits are therefore of two kinds:
— double limits, consisting of an upper and a lower limit;
— single limit, i.e. either an upper limit or a lower limit.
The possible configuration of limits is illustrated in Figure 1.
EXAMPLE 1 Double limits
For a single item in the form of a barrel of motor oil (i.e. the entity) the requirements for the kinematic
viscosity of the oil (i.e. the characteristic) could be
−5 2 −5 2
the kinematic viscosity should be not less than 0,5 × 10 m /s and not greater than 1,0 × 10 m /s.
EXAMPLE 2 Double limits
For one lot of bottles with frying oil (i.e. the entity) the requirements for the average boiling point at the
atmospheric pressure of 101,6 kPa for the oil in the bottles (i.e. the characteristic) could be
the average boiling point should be within the interval 105,0 °C to 115,0 °C.
EXAMPLE 3 Single upper limit
For a shipment of crude oil (i.e. the entity) the requirements for the sulfur mass fraction (i.e. the
characteristic) in the bulk could be
the sulfur mass fraction should be no greater than 2 %.
EXAMPLE 4 Single upper limit
For an individual (i.e. the entity) the requirements for the concentration of lead in blood (i.e. the
characteristic) could be
the concentration of lead should be no greater than 0,97 µmol/l.
EXAMPLE 5 Single lower limit
For a lot of bitumen (i.e. the entity) the requirements for the solubility of the bitumen in kerosene at
20 °C (i.e. the characteristic) could be
the solubility of the bitumen in kerosene at 20 °C should be not less than a mass fraction of 99 %.
EXAMPLE 6 Single upper limit
For a shipment of apples (i.e. the entity) the requirements for mass fraction of the apples infected with
pests (i.e. the characteristic) could be
the mass fraction of apples infected with pests should be less than 0,2 %.
Due to the variation of the mass of the individual apples, the mass fraction of infected apples will usually
be different from the number fraction of infected apples.
NOTE In many cases (e.g. in the environmental field), an additional implied limit such as 0 %, 0,0 kg/l and
100 % can be ignored when considering a single limit because they are theoretical and/or physical limits and
therefore need not be specified.
5 Uncertainty of results
5.1 General
When comparing a measurement result with the limiting values, it is necessary to take into consideration
the measurement uncertainty of the result. The uncertainty should be determined according to the
provisions of ISO/IEC Guide 98-3. ISO 5725-1 to ISO 5725-6, may also be consulted to help identify some
of the components of uncertainty. Examples are shown in Annex B.
NOTE This discussion implies that the contributions to the uncertainty from all stages in the measurement
procedure is taken into consideration and also includes any uncertainty due to sampling.
5.2 Reporting the measurement results and the measurement uncertainty
The measured value of the characteristic of interest and the measurement uncertainty should be
reported; the measurement uncertainty should be reported as an interval. When this interval is
a confidence interval, the confidence level (1 − α) should be reported together with the interval
(see ISO 3534-1). Otherwise the coverage factor of the uncertainty interval should be reported (see
ISO/IEC Guide 98-3:2008, 6.2.1).
6 Assessing conformity to requirements
6.1 General
A conformity test is a systematic examination (by means of measurement) of whether or not the entity
fulfils the specified requirements.
The objective of the conformity assessment is to provide confidence that the entity does or does not
fulfil the specified requirements.
This document recommends that the conformity assessment be performed as a two-stage procedure.
In the cases where a two-stage procedure either cannot be performed or for other reasons should not
be performed, a one-stage procedure is provided.
When a two-stage procedure is performed, there should be appropriate procedures to evaluate the
consistency of the measurement results from the two stages.
NOTE The advantage of the two-stage procedure over the one-stage procedure is the higher probability
of declaring conformity for entities with permissible values of the quantity of interest, which are close to the
limiting value(s). The disadvantage is a slightly higher probability of declaring conformity for entities with non-
permissible value(s) of the quantity of interest, which are close to the limiting values. If this increased probability
in declaring conformity for non-conforming entities cannot be accepted, a one-stage procedure is provided.
6.2 The two-stage conformity assessment
6.2.1 Stage 1
Perform the measurement procedure and estimate the measurement uncertainty of the resu
...