ISO/TR 16476:2016
(Main)Reference materials — Establishing and expressing metrological traceability of quantity values assigned to reference materials
Reference materials — Establishing and expressing metrological traceability of quantity values assigned to reference materials
ISO/TR 16476:2016 investigates, discusses, and specifies further, the general principles of establishing traceability of measurement results laid down in the Joint BIPM, OIML, ILAC and ISO Declaration on Metrological Traceability [1], in particular for values assigned to (certified) reference materials. The document covers the following topics: a) a study into existing principles for, and requirements to, the traceability of the value assigned to the property of a (C)RM, with a specific view to the current definition of metrological traceability given by the 2007 edition of the VIM (published also as JCGM 200:2008[2] and ISO/IEC Guide 99:2007[21]); b) the development of a sensible, widely applicable approach to the understanding of the traceability of a value assigned to (C)RM property; c) recommendations on how traceability should be established, demonstrated, and reported on certificates and other documents accompanying (C)RM. The developed approach is exemplified for measurement procedures not covered earlier by other guidance documents on the topic.
Matériaux de référence — Etablissement et expression de la traçabilité métrologique de valeurs assignées à des matériaux de référence
General Information
Standards Content (Sample)
TECHNICAL ISO/TR
REPORT 16476
First edition
Reference materials — Establishing
and expressing metrological
traceability of quantity values
assigned to reference materials
Matériaux de référence — Etablissement et expression de la
traçabilité métrologique de valeurs assignées à des matériaux de
référence
PROOF/ÉPREUVE
Reference number
©
ISO 2016
© ISO 2016, Published in Switzerland
All rights reserved. Unless otherwise specified, 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
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. +41 22 749 01 11
Fax +41 22 749 09 47
copyright@iso.org
www.iso.org
ii © ISO 2016 – All rights reserved
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 The VIM definition of metrological traceability . 1
3 Challenges arising from the definition of metrological traceability .2
3.1 Conventions . 2
3.2 (C)RM as the carrier of traceable values. 3
3.3 Implicit traceability to the unit of the measurement scale . 4
3.4 Traceability networks . 5
3.5 Properties expressed in units of measurement scales other than the SI . 5
3.6 Properties other than quantitative . 6
3.7 Summary of an ISO/REMCO event on metrological traceability . 6
4 Approaches to metrological traceability of (C)RM . 7
4.1 General . 7
4.2 Approach A . 7
4.3 Approach B . 8
5 Establishing traceability of (C)RM property values (Approach B) . 9
5.1 Principles . 9
5.2 Traceability pathways . 9
5.3 Steps in establishing traceability .10
5.3.1 General.10
5.3.2 Combining results .10
5.4 Summary .12
6 Reporting traceability .12
6.1 Inquiry .12
6.2 Results of the inquiry .12
6.3 Requirements .13
6.4 Formats .14
6.5 Further recommendations .16
Annex A (informative) Worked-out example .17
Annex B (informative) Catalogue of analytes and measurement areas covered by WHO .19
Annex C (informative) Example for method-independent, SI traceable values obtained by
inter-laboratory comparison .21
Bibliography .22
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 on 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 the following URL: www.iso.org/iso/foreword.html.
ISO/TR 16476 was prepared by the ISO Committee on Reference Materials (ISO/REMCO).
iv PROOF/ÉPREUVE © ISO 2016 – All rights reserved
Introduction
Reference materials (RM), in particular when certified (CRM), are a major tool for assuring the quality
and reliability of results obtained in measurement and testing. CRM property values, in particular used
for assessing the trueness of a measurement procedure as implemented in a laboratory, also establish
traceability of the measurement result. Which reference the property values assigned to (C)RM should
be traceable to, and how this traceability should be established, demonstrated, and reported on
certificates is, therefore, a question of primary importance, mainly for RM producers. However, users of
(C)RMs should also know what the endpoint of their traceability chain is, in particular for all purposes
of cross-border acceptance of measurement results.
It was therefore considered necessary to conduct a study into existing principles for, and requirements
to, the traceability of (C)RM, in particular with a specific view to the current definition of metrological
traceability given by the Vocabulary of International Metrology (VIM), edition 3, 2007.
TECHNICAL REPORT ISO/TR 16476:2016(E)
Reference materials — Establishing and expressing
metrological traceability of quantity values assigned to
reference materials
1 Scope
This Technical Report investigates, discusses, and specifies further, the general principles of establishing
traceability of measurement results laid down in the Joint BIPM, OIML, ILAC and ISO Declaration on
[1]
Metrological Traceability , in particular for values assigned to (certified) reference materials. The
document covers the following topics:
a) a study into existing principles for, and requirements to, the traceability of the value assigned to the
property of a (C)RM, with a specific view to the current definition of metrological traceability given
[2] [21]
by the 2007 edition of the VIM (published also as JCGM 200:2008 and ISO/IEC Guide 99:2007 );
b) the development of a sensible, widely applicable approach to the understanding of the traceability
of a value assigned to (C)RM property;
c) recommendations on how traceability should be established, demonstrated, and reported on
certificates and other documents accompanying (C)RM.
The developed approach is exemplified for measurement procedures not covered earlier by other
guidance documents on the topic.
2 The VIM definition of metrological traceability
[2],[21]
The recent edition of the VIM defines metrological traceability (term 2.41) as
property of a measurement result whereby the result can be related to a reference through a documented
unbroken chain of calibrations, each contributing to the measurement uncertainty
NOTE 1 For this definition, a ‘reference’ can be a definition of a measurement unit through its
practical realization, or a measurement procedure including the measurement unit for a
non-ordinal quantity, or a measurement standard.
NOTE 2 Metrological traceability requires an established calibration hierarchy.
NOTE 3 Specification of the reference must include the time at which this reference was used in
establishing the calibration hierarchy, along with any other relevant metrological infor-
mation about the reference, such as when the first calibration in the calibration hierarchy
was performed.
NOTE 4 For measurements with more than one input quantity in the measurement model, each
of the input quantity values should itself be metrologically traceable and the calibration
hierarchy involved may form a branched structure or a network. The effort involved in
establishing metrological traceability for each input quantity value should be commensu-
rate with its relative contribution to the measurement result.
NOTE 5 Metrological traceability of a measurement result does not ensure that the measurement
uncertainty is adequate for a given purpose or that there is an absence of mistakes.
NOTE 6 A comparison between two measurement standards may be viewed as a calibration if the
comparison is used to check and, if necessary, correct the quantity value and measure-
ment uncertainty attributed to one of the measurement standards.
NOTE 7 The ILAC considers the elements for confirming metrological traceability to be an un-
broken metrological traceability chain to an international measurement standard or a
national measurement standard, a documented measurement uncertainty, a documented
measurement procedure, accredited technical competence, metrological traceability to
the SI, and calibration intervals (see ILAC P-10:2002).
NOTE 8 The abbreviated term “traceability” is sometimes used to mean ‘metrological traceabil-
ity’ as well as other concepts, such as ‘sample traceability’ or ‘document traceability’ or
‘instrument traceability’ or ‘material traceability’, where the history (“trace”) of an item
is meant. Therefore, the full term of “metrological traceability” is preferred if there is any
risk of confusion.
NOTE 7 makes clear that the measurement method/procedure is a part of the traceability statement,
but insufficient if taken alone. This implies that a measurement result or the assigned value of a (C)RM
can be traceable to a method or a series of methods used, but not to the method alone, although such
statements can still be found on CRM certificates. Without any doubt, the measurement procedure used
will mostly be reflected in the definition of the measurand, but additionally, the value assigned to the
measurand should be made traceable to stated references given the procedure applied, thus, the latter
cannot be the endpoint of the traceability chain for the assigned value. Recent presentations on the
topic (e.g. Reference [3]) support this viewpoint.
As expressed in NOTE 2, the definition as given above is governed by the (assumed) existence of a
straightforward, single-route top-down reference standard hierarchy. Reference [4], as a guidance to
implement the above VIM definition in chemistry, almost always assumes the existence of higher-order
reference materials, conveniently certified at the highest level by using allegedly primary methods.
This Technical Report does not go into further details of these situations since References [4] to [6]
provide sufficient guidance. Considerations and guidance on traceability hierarchies together with
graphs visualizing and illustrating traceability chains, including branched ones, can also be found in
Reference [7].
The described philosophy works fine for all levels which still have a “higher-order” level above, or fields
where primary methods exist and can readily be used for a valid and reliable value assignment to the
measurand. However, at some point, the top with no “higher-order” is reached. It may also be stated
that for a huge amount of certified reference materials at this level, no primary method is available for
assigning values to the measurands. This causes the general problem of traceability for property values
of (C)RMs allocated at prominent places in the hierarchy. More critical points will be discussed under
Clause 3.
NOTE An annotation document is being developed. Its aim is to give further explanations to the VIM
definitions; it will also provide advice regarding the application of these definitions.
3 Challenges arising from the definition of metrological traceability
3.1 Conventions
For the purposes of this Technical Report, the following conventions apply.
— “Traceability of an RM”
is in common and daily use, it is understood throughout as the traceability of the quantity value
assigned to a (certified) reference material.
— “(Analytical) method”
is used in the sense of defining the instrumental implementation of the (most often physical)
principle of obtaining, from an appropriately pre-processed and/or transformed object under
2 PROOF/ÉPREUVE © ISO 2016 – All rights reserved
investigation, a signal (subject to further processing) reflecting the sought-after property. Such
implementations are, for example, ToF-IDMS, GC-FID, LC-MS/MS, HPLC-DAD, FT-IR, etc.
— “Measurement protocol”
is used to refer to measurement procedures prescribed or standardized to an extent that the
value(s) assigned to the material becomes senseless without direct reference to these prescriptions,
i.e. where not only the conditions under which measurements have to be taken but also form,
structure, shape, size and/or composition of the specimen are prescribed.
— “RM document”, sometimes also called “property value sheet” or “product information sheet” (see
ISO Guide 31)
is used as an analogue to, and distinction from, the term “certificate” as defined in ISO Guide 31.
Certificates refer to CRM, while an RM document provides the necessary information on the
properties of a (non-certified) RM.
— “Matrix (C)RM”
an RM made out of natural-born substance(s) or synthetically re-constituted ingredients,
characterised for composition.
— “Property (C)RM”
an RM characterized for a property other than the content of main components and/or impurities
as e.g. tensile strength or Charpy impact for an alloyed steel.
NOTE This Technical Report refers to the requirements of ISO Guide 34, in force at the time of publication of
this Technical Report. For traceability issues, the future ISO 17034 will also follow the principles of the named
ISO Guide 34.
3.2 (C)RM as the carrier of traceable values
In the context of (C)RM production, a basic problem of the definition of metrological traceability (see
Clause 2) is that it refers to traceability of a result of a measurement. (C)RMs are normally considered
as artefacts providing traceability of a measurement result.
NOTE An RM which comes without a (measured) value attributed to its properties, e.g. in cases where
the material has an intended purpose not requiring such attributed value, does not experience the need of
establishing traceability.
The value and uncertainty carried by an (C)RM are, in virtually almost all cases, combinations of
results of various measurements. These results may refer to the different steps of RM production,
namely homogeneity and stability estimation, and to measurements taken in the characterization step
based upon independent implementations of the same measurement procedure, or implementations of
different independent measurement procedures.
Even in cases when all of the single results obtained in an RM certification are traceable, it remains
unclear to which extent and endpoint of the traceability chain a combined result is traceable. This
problem increases considerably if the results to be combined are traceable to different endpoints, or at
least via different pathways all having different lengths and reliabilities.
EXAMPLE Karl Fischer titration and the oven (drying) method have different endpoints of their traceability
chains. It might be sensible to include the method in the definition of the measurand which solves the problem of
the traceability endpoint.
However, for the seemingly clear specification “water in a matrix”, the above mentioned problem arises.
A sensible traceability statement for the value combined from results of both methods might be based
upon the more direct oven method (see also 5.3.2).
3.3 Implicit traceability to the unit of the measurement scale
The measurement procedure is a convention, it most often also includes transformation(s) of the
measurand. This holds for most areas in chemistry, biology, or life sciences. Different conventions (i.e.
different measurement procedures) for the same measurand may lead to different results, i.e. they
turn out to be incompatible. This is reflected in NOTE 5 to the definition saying that traceability is a
necessary condition for comparability of measurement results, but insufficient for their compatibility.
Generally speaking, the measurement procedure has a non-negligible influence on the value assigned.
The principal approach is that traceability can only be established “given the specified measurement
procedure used”. The specification of this procedure in a written standard, an SOP, etc., is a nominal
prescription. Any implementation in a specific place, by a specific operator using specific equipment
will cause inevitable deviations from the prescription, no matter whether these are negligibly small
or introduce a real contribution to the total uncertainty. The deviations should be assessed in specific
investigations virtually considered as calibrations against the nominal prescription.
EXAMPLE ISO 148-3 describes the production of samples for Charpy impact tests. ISO 148-3 provides
nominal values for the size of the sample and the location and shape of the V notch. Despite the fact that the
instrument measuring the size (e.g. a calliper) should duly be calibrated and introduces a measurement
uncertainty, the machining tool also has a certain variability. Both should be considered.
NOTE In chemical analysis, a ruggedness test as part of a thorough method validation assesses most of the
deviations occurring from implementation of a method under real, and varying within specified limits, laboratory
conditions.
All of these calibrations will normally not introduce real corrections to the measured value but
contribute to the overall uncertainty which makes the approach compatible with the VIM definition.
It is commonly accepted that the combination of metrological traceability and proper measurement
[8]
uncertainty is the only way how measurement results can legitimately be compared. Moreover,
measurement uncertainty estimation of the calibration steps is a mandatory prerequisite for the
establishment of traceability. The concept of calibration against a nominal requirement closes the gap in
cases where the routes to measurement scales (SI and others) are considered “indirect”. Demonstration
of compliance with nominal requirements has to be carried out using measuring instruments which
are, for the measurand they tackle, traceable to the corresponding unit of the scale (callipers, balances,
volumetric flasks, etc.). This concept is formalised as approach B under Clause 4.
The significance and influence of indirect pathways to measurement scales is also recognized in the Joint
[1]
Declaration on Metrological Traceability stating that “In general, … references are the International
System of Units (SI), but where such traceability is not yet feasible, measurement results should be traceable
to other internationally agreed references…”
It might seem viable to attribute all the peculiarities of the measurement method or procedure to the
definition of the measurand as proposed in Reference [9]. In general, one should remember that the
VIM defines the measurand as the “quantity intended to be measured”, not as the procedures necessary
for accomplishing the intention. At the same time, the faults that might happen when the quantity is
measured are not considered.
Two other points have to be considered.
— Firstly, the approach of Reference [9] will work only with one single method of determination which
then (according to the requirements of ISO Guide 34) should be a primary method, a restriction
limiting the applicability of the approach to special cases.
— Secondly, it will limit the field of application of the material or its commutability and, thus,
considerably reduce its technical and commercial value.
A sensible and balanced distribution of method impact on the measurement result between the definition
of the measurand and the traceability chain(s) to units of scale is therefore crucial (see also 5.4).
4 PROOF/ÉPREUVE © ISO 2016 – All rights reserved
3.4 Traceability networks
NOTE 4 in the definition of metrological traceability (Clause 2) suggests that for measurements
with more than one input quantity in the measurement model (a situation which is daily practice in
chemical analysis and virtually all fields of testing), each of the input quantity values should itself be
metrologically traceable and the calibration hierarchy involved may form a branched structure or a
network. The effort involved in establishing metrological traceability for each input quantity value
should be commensurate with its relative contribution to the measurement result.
Figure 1 — “Horizontal” traceability network relating the measured quantities in a
measurement procedure model to a set of SI units
It is assumed that this explicitly allows a “horizontal” networked strategy for establishing the
traceability of a particular implementation of a measurement procedure as visualized in Figure 1.
The concept formalized under approach B of Clause 4 is a consequence of NOTE 4 to the definition of
metrological traceability and implements this “horizontal” strategy. It is the only feasible approach
when no higher-order reference is available in the vertical direction.
3.5 Properties expressed in units of measurement scales other than the SI
Most of the scales other than the SI have nevertheless similarities with the latter, namely that they are
realised by a single or a set of materialisations/artefacts with assigned values expressed predominantly
in real numbers (e.g. pH scale). Furthermore, fractions or multiples of the basic unit exist. Some scales
use ordinal numbers which express a cardinal “smaller-larger” relationship between the realizations of
the points on the scale (e.g. Mohs hardness, see Reference [10]) rather than an explicit proportionality
or a counting result. Here, specific problems with the resolution of the scale may arise. However, for
establishment of traceability to these scales the same rules and recommendations may be applied as
given in this Technical Report for traceability to the endpoint SI.
A prominent, widely used non-SI measurement scale is the series of natural numbers. It is the basic
measurement scale in all areas of measurement where counting is involved, e.g. of specified objects
(pollen in a certain amount of air, E.coli bacteria in a specified volume of a food product, etc.). The
peculiarity of this scale is that no materialization of the unit exists to which traceability might be
established by direct comparison (calibration).
On the other hand, the unit “unity” can hardly be misinterpreted, and an interpretation of the unit in
one implementation will exactly match the interpretation in any other. The problems are rather object-
specific (misidentification of objects, double-counting, object overlap, etc.), and the resolution of the
scale should always be taken into account. Furthermore, depending on the kind of the objects and the
background of the measurement, statistics different from those commonly accepted for continuous-
scale measurements may apply. Thus, the procedure and its specifications gain even more importance
for measurement results expressed in the unit of this scale, and the impacts of deviations from the
specified procedure should thoroughly be evaluated. However, also with these peculiarities, the
principles of Clauses 4 and 5 also apply here.
3.6 Properties other than quantitative
It is assumed that this is covered by NOTE 1 in the definition of metrological traceability (Clause 2)
which uses the term “non-ordinal quantity” meaning that these measurement results are traceable to
the measurement procedure alone. However, more guidance is needed for the distinction of ordinal
and non-ordinal quantities, in particular with a specific view to the fact that modern measurement
procedures virtually always involve measurements of fully quantifiable (and, thus, not non-ordinal)
quantities and some nominal-valued decisions on the set of quantified measurement results.
Examples are identity of a substance which materializes through measurement, and the sequence of
objects which involves identity and a series of ordinal numbers.
This Technical Report solely deals with quantitative results.
3.7 Summary of an ISO/REMCO event on metrological traceability
In June 2012, ISO/REMCO held a brainstorming session on recent views and approaches to (C)RM
th
traceability in Vienna/Austria, in connection with its 35 General Assembly. The session could be
joined by interested parties world-wide via on-line (video and telephone) connections.
Issues covered during the session included the role of (C)RM in establishing traceability and traceability
statements in certificates. Three major perspectives have been considered in presentations, namely
those of users of (C)RM, the accreditation bodies, and the reference material producers (RMP). The
presentations and a summary are available on the ISO/REMCO webpage.
The user’s perspective referred to the field of geoanalysis (in particular XRF analysis of minerals) and
emphasized a) the use of (C)RM specifically tailored for use in the area and b) the impact of calibration
pathways (standard solutions versus matrix (C)RM), in particular unresolved inconsistent results
when using different calibrants.
Accreditation bodies (AB) identify increased and evolving expectations to RMP (accredited under
ISO Guide 34), statements of traceability (required under ISO Guide 34), additional information on the
certification procedure (certification report), and the intended use of the (C)RM. Users (accredited
under ISO/IEC 17025) are required to describe the role of (C)RM in establishing the traceability of
their results. (C)RM are critical consumables requiring a specific traceability evaluation if not sourced
from accredited RMP or a material included in KCDB Appendix C or the JCTLM RM database. It was also
stated that the AB implementation policy needs to be consistent with respect to traceability.
An RMP scrutinized the role of (C)RM in the delivery of traceability, providing examples of how (C)
RM can be used to validate results including measurement uncertainty assignments, demonstrate the
equivalence of measurements, establish comparability (in the VIM sense) to a measurement scale, and
evaluate and correct for bias.
Fully consistent with this Technical Report, participants of the event concluded that traceability
— cannot be established to an institution,
— establishes comparability, not necessarily trueness of results, and
— is defined by, and potentially limited to, the certification method (for method-defined measurands).
6 PROOF/ÉPREUVE © ISO 2016 – All rights reserved
Full information from the RMP on the certification procedure and the intended use is critical and
should be required by the users (e.g. in the form of a certification report). The challenge of propagating
a quantity through a traceability chain when the measurand changes is generally underestimated (see
3.3), implying a need to check with care for inconsistencies between “claimed” and “actual” quantity
measured.
The following conclusions have been drawn.
— Current practice of reporting traceability on CRM certificates is very often not consistent or
sufficient.
— A need for evaluation (by RMP and accreditors) of traceability statements to “higher-order”
references (including but not limited to SI) exists.
— Traceability statements should avoid generic claims, and a concise summary of the technical
basis/certification procedure used to obtain the property values should be available (at least on
request of users or accreditors).
— Intended-use statements are a critical component for appropriate use by end users.
— Further work is needed on minimum requirements for the content of certificates and supplementary
information. For possible approaches, see Clause 6.
4 Approaches to metrological traceability of (C)RM
4.1 General
Given the definition and the considerations made in Clauses 3 and 4, two approaches to (C)RM
traceability seem feasible in principle. Note that their citation (as A and B) should in no way be confused
with preferences given, or hierarchies attributed to, the approaches.
4.2 Approach A
According to NOTE 2 in the definition of metrological traceability (Clause 2), metrological traceability
requires an established calibration hierarchy. One might be tempted to define (C)RM as being endpoints
of the traceability chain which do not need further traceability “upwards”. In particular, one might look
at (C)RM as being artefacts which establish their own measurement scales.
NOTE A small number of recent CCQM Key Comparisons might be interpreted into this direction although
basically covering other aims. CRM from different NMI representing different amounts of substance of the same
analyte in the same matrix have been compared for equivalence, thus, establishing a “scale” of this specified
analyte in the specified matrix, say NO in N , or amounts of ethanol in water.
The approach has a number of undoubted advantages and is used for the reference standards produced
and issued by the World Health Organization (WHO). Both the measurement uncertainty calculation
and the traceability chain start from the internationally accepted reference standard. The validity of
[11]
the standard is assured by a series of technical and assessment procedures, namely:
— The need is recognized by scientific and medical community worldwide and a case formally made
by the WHO Secretariat to the Expert Committee on Biological Standardization (ECBS) on the basis
of public health impact.
— Working groups of experts are involved in setting the priorities and characteristics for selection of
the candidate reference preparations.
— An international collaborative study has to be carried out before any candidate reference preparation
can be considered for establishment by the WHO ECBS.
— The goal of such a study is to determine which candidate material is suitable to serve as a WHO
reference material for the standardization of a biological product or of an in vitro diagnostic tests.
— An internationally agreed unit is attributed to the first WHO Biological Reference Material for
biological activity characterization. The continuity of such a unit is ensured by replacement with a
new batch of reference material which is calibrated against the first or previous reference material.
— A requirement to be met by any batch of a WHO Biological Reference Material is that the content
in every ampoule in the batch should be identical in terms of composition, quantity, potency and
stability.
— The Biological Standardization document which reports the international multi-method collaborative
study is peer-reviewed before being submitted to the WHO ECBS. It has to be approved by the ECBS
for final release of the material.
[12]
An overview of the fields of analytes and measurement areas covered by WHO standards is given
in Annex C, and an example of accompanying documentation for a primary standard shown. The WHO
creates as many scales as is needed in reality.
Propagating and multiplying this fit for the specific purposes approach to the full spectrum of
measurement and testing activities would create a very large (at least infinite) number of scales, one
for each feasible analyte-matrix (in chemistry) or property-of-substance combination (in testing).
Although this treatment of the requirement for calibration hierarchies does also not fully coincide with
the network idea of NOTE 4 in the definition of metrological traceability (Clause 2), where calibrations
may be at the same level but go into different “horizontal” directions, there are situations (in particular
for qualitative-property RM) which will make approach A mandatory. This might be covered by format
D in 6.2. However, this is not further elaborated in this Technical Report (see also 3.6).
4.3 Approach B
[20]
This approach is fully in line with the principles of the ISO Guide 35 and described by the options
given under ISO Guide 35:2006, 9.2. Traceability is stated “given the specified measurement procedure
used”, i.e. the pathway and the endpoint of the traceability chain have to be specified. The specification
of the procedure is nominal. Any implementation in a specific place will cause deviations from the
nominal prescription. These need to be assessed in specific investigations virtually considered as
calibrations against the nominal (written) standard. For more detailed considerations, see the example
in Annex A.
The complete model formula describing the measurement procedure, at least for the majority of
measurement procedures commonly used in chemical analysis, but also in testing, takes the form (see,
for example, Reference [13]):
k
p
∏
i
n
i=1
x =⋅ Fkwith 1 ≤≤ mn≤ (1)
∏
meas i
m
im=+1
p
∏
i
ik=+1
where the p represent the directly measured/determined (explicit) parameters, and the F the influences
i i
from grouped/combined sources, both with and without corrective influence (i.e. values differing from
unity or not), and all with uncertainty contributions. Special cases, where model formulae include
additive terms or are non-linear, have to be considered separately. However, the division into directly
observed and indirect influential parameters equally applies.
The F influences may, for example, represent the (used for correction or not) bias term from a calibration,
i
or the degree of compliance with a nominal prescription. As compliance is normally assumed, the value
assigned to the corresponding F term would be unity, but the uncertainty connected with this term
would be assessed in a virtual “calibration” procedure by deliberately deviating from the prescription
and evaluating the influence on the measured value. In method validation, this is normally called
robustness or ruggedness test. Note that the model equation as above can be treated as a consequence
of NOTE 4 in the definition of metrological traceability (Clause 2).
8 PROOF/ÉPREUVE © ISO 2016 – All rights reserved
With the model formula above, the relative uncertainty attributed to the measurement result would be
m n
22 2
ux()=+up() uF() (2)
∑∑
rrmeas i ri
i==1 im+1
except possible correlation terms which have to be accounted for if the influential parameters are not,
or not fully, independent of each other.
NOTE For RM coming without a (measured) value, see 3.2.
5 Establishing traceability of (C)RM property values (Approach B)
5.1 Principles
Under approach B, the traceability of a (C)RM can be established within a framework of reasonable
interpretations of the basic VIM definition, namely:
— Traceability of the assigned value of a (C)RM is a property of this value whereby the single or a
consolidated set of measurement results obtained for the (C)RM can be related to a single or a
set of references through a documented unbroken chain of calibrations, each contributing to the
measurement uncertainty.
— A reference can be an established and well-understood method/procedure, a written description
and specification of an operational procedure, or an artefact/artefacts realizing a point or a range
on the measurement scale of the quantity under consideration. Explicit reference should be made
to the measurement procedures applied by citing the standard, protocol (e.g. AOAC Peer Reviewed
Method), publication or text book.
— Members of a set of references may refer to the same quantity (when defining a measurement
scale) or of different but essential in the measurement process quantities (realizing a traceability
network). This is particularly important for more-dimensional measurands (e.g. spectra).
— A consolidated set of measurement results is a measurement result combined from several
measurement results using appropriate procedures which assure full compatibility between the
results combined and the result consolidated from the former, all within an appropriately assessed
uncertainty. The establishment of the total measurement uncertainty budget of the consolidated
value follows the commonly accepted rules and includes allowances for the “procedure impact”
according to Clause 4.
Compatibility between the results combined may arise per se or may have to be established by
introducing uncertainty components which account for data discrepancies. Decision on the level of
admissible discrepancy is case-sensitive and subject to expert judgment, meaning that starting from
a certain level of discrepancy, the measurement results may seem non-commensurable (and thus no
longer traceable to the same or the same set of references).
Establishment and statement of traceability of a CRM is mandatory; it is not for an RM fulfilling the
basic requirements only (i.e. coming without an assigned value). However, if any values are assigned to
an RM, their traceability should be assured as well (see ISO Guide 33:2015, 6.4.2).
5.2 Traceability pathways
ISO Guide 34 and ISO Guide 35 in their current editions accept four general approaches for the
characterization of reference materials, namely
a) measurement by a single (primary) method in a single laboratory;
b) measurement by two or more independent reference methods in one laboratory;
c) measurement by a network of laboratories using one or more methods of demonstrable accuracy;
d) a method-specific approach giving only method-specific assessed property values, using a network
of laboratories.
Traceability pathways may have different targets, namely
1) (direct) traceability to the unit of the measurement scale,
2) traceability to the unit of the measurement scale via, and given by, a measurement method or
procedure, and
3) traceability to a protocol.
As a rule, pathways and certification schemes combine as shown in Table 1.
Table 1 — Certification/characterization schemes and pathways of traceability
Scheme Pathway
1) 2) 3)
a) X — X
b) X X -
c) — X -
d) — X X
5.3 Steps in establishing traceability
5.3.1 General
[20]
General steps to be taken, and provisions to be made are described in ISO Guide 35 and should be
followed strictly. Major points are as follows:
— Transformation (of the measurand): Although the determination of the property value itself can
be made traceable to appropriate units through, for example, calibration of the measurement
equipment used, steps like the transformation of the sample from one physical (chemical) state to
another cannot. Such transformations may only
...
TECHNICAL ISO/TR
REPORT 16476
First edition
2016-06-01
Reference materials — Establishing
and expressing metrological
traceability of quantity values
assigned to reference materials
Matériaux de référence — Etablissement et expression de la
traçabilité métrologique de valeurs assignées à des matériaux de
référence
Reference number
©
ISO 2016
© ISO 2016, Published in Switzerland
All rights reserved. Unless otherwise specified, 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
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. +41 22 749 01 11
Fax +41 22 749 09 47
copyright@iso.org
www.iso.org
ii © ISO 2016 – All rights reserved
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 The VIM definition of metrological traceability . 1
3 Challenges arising from the definition of metrological traceability .2
3.1 Conventions . 2
3.2 (C)RM as the carrier of traceable values. 3
3.3 Implicit traceability to the unit of the measurement scale . 4
3.4 Traceability networks . 5
3.5 Properties expressed in units of measurement scales other than the SI . 5
3.6 Properties other than quantitative . 6
3.7 Summary of an ISO/REMCO event on metrological traceability . 6
4 Approaches to metrological traceability of (C)RM . 7
4.1 General . 7
4.2 Approach A . 7
4.3 Approach B . 8
5 Establishing traceability of (C)RM property values (Approach B) . 9
5.1 Principles . 9
5.2 Traceability pathways .10
5.3 Steps in establishing traceability .10
5.3.1 General.10
5.3.2 Combining results .11
5.4 Summary .12
6 Reporting traceability .12
6.1 Inquiry .12
6.2 Results of the inquiry .12
6.3 Requirements .13
6.4 Formats .14
6.5 Further recommendations .16
Annex A (informative) Worked-out example .17
Annex B (informative) Catalogue of analytes and measurement areas covered by WHO .19
Annex C (informative) Example for method-independent, SI traceable values obtained by
inter-laboratory comparison .21
Bibliography .22
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 on 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 the following URL: www.iso.org/iso/foreword.html.
ISO/TR 16476 was prepared by the ISO Committee on Reference Materials (ISO/REMCO).
iv © ISO 2016 – All rights reserved
Introduction
Reference materials (RM), in particular when certified (CRM), are a major tool for assuring the quality
and reliability of results obtained in measurement and testing. CRM property values, in particular used
for assessing the trueness of a measurement procedure as implemented in a laboratory, also establish
traceability of the measurement result. Which reference the property values assigned to (C)RM should
be traceable to, and how this traceability should be established, demonstrated, and reported on
certificates is, therefore, a question of primary importance, mainly for RM producers. However, users of
(C)RMs should also know what the endpoint of their traceability chain is, in particular for all purposes
of cross-border acceptance of measurement results.
It was therefore considered necessary to conduct a study into existing principles for, and requirements
to, the traceability of (C)RM, in particular with a specific view to the current definition of metrological
traceability given by the Vocabulary of International Metrology (VIM), edition 3, 2007.
TECHNICAL REPORT ISO/TR 16476:2016(E)
Reference materials — Establishing and expressing
metrological traceability of quantity values assigned to
reference materials
1 Scope
This Technical Report investigates, discusses, and specifies further, the general principles of establishing
traceability of measurement results laid down in the Joint BIPM, OIML, ILAC and ISO Declaration on
[1]
Metrological Traceability , in particular for values assigned to (certified) reference materials. The
document covers the following topics:
a) a study into existing principles for, and requirements to, the traceability of the value assigned to the
property of a (C)RM, with a specific view to the current definition of metrological traceability given
[2] [21]
by the 2007 edition of the VIM (published also as JCGM 200:2008 and ISO/IEC Guide 99:2007 );
b) the development of a sensible, widely applicable approach to the understanding of the traceability
of a value assigned to (C)RM property;
c) recommendations on how traceability should be established, demonstrated, and reported on
certificates and other documents accompanying (C)RM.
The developed approach is exemplified for measurement procedures not covered earlier by other
guidance documents on the topic.
2 The VIM definition of metrological traceability
[2],[21]
The recent edition of the VIM defines metrological traceability (term 2.41) as
property of a measurement result whereby the result can be related to a reference through a documented
unbroken chain of calibrations, each contributing to the measurement uncertainty
NOTE 1 For this definition, a ‘reference’ can be a definition of a measurement unit through its
practical realization, or a measurement procedure including the measurement unit for a
non-ordinal quantity, or a measurement standard.
NOTE 2 Metrological traceability requires an established calibration hierarchy.
NOTE 3 Specification of the reference must include the time at which this reference was used in
establishing the calibration hierarchy, along with any other relevant metrological infor-
mation about the reference, such as when the first calibration in the calibration hierarchy
was performed.
NOTE 4 For measurements with more than one input quantity in the measurement model, each
of the input quantity values should itself be metrologically traceable and the calibration
hierarchy involved may form a branched structure or a network. The effort involved in
establishing metrological traceability for each input quantity value should be commensu-
rate with its relative contribution to the measurement result.
NOTE 5 Metrological traceability of a measurement result does not ensure that the measurement
uncertainty is adequate for a given purpose or that there is an absence of mistakes.
NOTE 6 A comparison between two measurement standards may be viewed as a calibration if the
comparison is used to check and, if necessary, correct the quantity value and measure-
ment uncertainty attributed to one of the measurement standards.
NOTE 7 The ILAC considers the elements for confirming metrological traceability to be an un-
broken metrological traceability chain to an international measurement standard or a
national measurement standard, a documented measurement uncertainty, a documented
measurement procedure, accredited technical competence, metrological traceability to
the SI, and calibration intervals (see ILAC P-10:2002).
NOTE 8 The abbreviated term “traceability” is sometimes used to mean ‘metrological traceabil-
ity’ as well as other concepts, such as ‘sample traceability’ or ‘document traceability’ or
‘instrument traceability’ or ‘material traceability’, where the history (“trace”) of an item
is meant. Therefore, the full term of “metrological traceability” is preferred if there is any
risk of confusion.
NOTE 7 makes clear that the measurement method/procedure is a part of the traceability statement,
but insufficient if taken alone. This implies that a measurement result or the assigned value of a (C)RM
can be traceable to a method or a series of methods used, but not to the method alone, although such
statements can still be found on CRM certificates. Without any doubt, the measurement procedure used
will mostly be reflected in the definition of the measurand, but additionally, the value assigned to the
measurand has to be made traceable to stated references given the procedure applied, thus, the latter
cannot be the endpoint of the traceability chain for the assigned value. Recent presentations on the
topic (e.g. Reference [3]) support this viewpoint.
As expressed in NOTE 2, the definition as given above is governed by the (assumed) existence of a
straightforward, single-route top-down reference standard hierarchy. Reference [4], as a guidance to
implement the above VIM definition in chemistry, almost always assumes the existence of higher-order
reference materials, conveniently certified at the highest level by using allegedly primary methods.
This Technical Report does not go into further details of these situations since References [4] to [6]
provide sufficient guidance. Considerations and guidance on traceability hierarchies together with
graphs visualizing and illustrating traceability chains, including branched ones, can also be found in
Reference [7].
The described philosophy works fine for all levels which still have a “higher-order” level above, or fields
where primary methods exist and can readily be used for a valid and reliable value assignment to the
measurand. However, at some point, the top with no “higher-order” is reached. It may also be stated
that for a huge amount of certified reference materials at this level, no primary method is available for
assigning values to the measurands. This causes the general problem of traceability for property values
of (C)RMs allocated at prominent places in the hierarchy. More critical points will be discussed under
Clause 3.
NOTE An annotation document is being developed. Its aim is to give further explanations to the VIM
definitions; it will also provide advice regarding the application of these definitions.
3 Challenges arising from the definition of metrological traceability
3.1 Conventions
For the purposes of this Technical Report, the following conventions apply.
— “Traceability of an RM”
is in common and daily use, it is understood throughout as the traceability of the quantity value
assigned to a (certified) reference material.
— “(Analytical) method”
is used in the sense of defining the instrumental implementation of the (most often physical)
principle of obtaining, from an appropriately pre-processed and/or transformed object under
2 © ISO 2016 – All rights reserved
investigation, a signal (subject to further processing) reflecting the sought-after property. Such
implementations are, for example, ToF-IDMS, GC-FID, LC-MS/MS, HPLC-DAD, FT-IR, etc.
— “Measurement protocol”
is used to refer to measurement procedures prescribed or standardized to an extent that the
value(s) assigned to the material becomes senseless without direct reference to these prescriptions,
i.e. where not only the conditions under which measurements have to be taken but also form,
structure, shape, size and/or composition of the specimen are prescribed.
— “RM document”, sometimes also called “property value sheet” or “product information sheet” (see
ISO Guide 31)
is used as an analogue to, and distinction from, the term “certificate” as defined in ISO Guide 31.
Certificates refer to CRM, while an RM document provides the necessary information on the
properties of a (non-certified) RM.
— “Matrix (C)RM”
an RM made out of natural-born substance(s) or synthetically re-constituted ingredients,
characterised for composition.
— “Property (C)RM”
an RM characterized for a property other than the content of main components and/or impurities
as e.g. tensile strength or Charpy impact for an alloyed steel.
NOTE This Technical Report refers to the requirements of ISO Guide 34, in force at the time of publication of
this Technical Report. For traceability issues, the future ISO 17034 will also follow the principles of the named
ISO Guide 34.
3.2 (C)RM as the carrier of traceable values
In the context of (C)RM production, a basic problem of the definition of metrological traceability (see
Clause 2) is that it refers to traceability of a result of a measurement. (C)RMs are normally considered
as artefacts providing traceability of a measurement result.
NOTE An RM which comes without a (measured) value attributed to its properties, e.g. in cases where
the material has an intended purpose not requiring such attributed value, does not experience the need of
establishing traceability.
The value and uncertainty carried by an (C)RM are, in virtually almost all cases, combinations of
results of various measurements. These results may refer to the different steps of RM production,
namely homogeneity and stability estimation, and to measurements taken in the characterization step
based upon independent implementations of the same measurement procedure, or implementations of
different independent measurement procedures.
Even in cases when all of the single results obtained in an RM certification are traceable, it remains
unclear to which extent and endpoint of the traceability chain a combined result is traceable. This
problem increases considerably if the results to be combined are traceable to different endpoints, or at
least via different pathways all having different lengths and reliabilities.
EXAMPLE Karl Fischer titration and the oven (drying) method have different endpoints of their traceability
chains. It might be sensible to include the method in the definition of the measurand which solves the problem of
the traceability endpoint.
However, for the seemingly clear specification “water in a matrix”, the above mentioned problem arises.
A sensible traceability statement for the value combined from results of both methods might be based
upon the more direct oven method (see also 5.3.2).
3.3 Implicit traceability to the unit of the measurement scale
The measurement procedure is a convention, it most often also includes transformation(s) of the
measurand. This holds for most areas in chemistry, biology, or life sciences. Different conventions (i.e.
different measurement procedures) for the same measurand may lead to different results, i.e. they
turn out to be incompatible. This is reflected in NOTE 5 to the definition saying that traceability is a
necessary condition for comparability of measurement results, but insufficient for their compatibility.
Generally speaking, the measurement procedure has a non-negligible influence on the value assigned.
The principal approach is that traceability can only be established “given the specified measurement
procedure used”. The specification of this procedure in a written standard, an SOP, etc., is a nominal
prescription. Any implementation in a specific place, by a specific operator using specific equipment
will cause inevitable deviations from the prescription, no matter whether these are negligibly small
or introduce a real contribution to the total uncertainty. The deviations should be assessed in specific
investigations virtually considered as calibrations against the nominal prescription.
EXAMPLE ISO 148-3 describes the production of samples for Charpy impact tests. ISO 148-3 provides
nominal values for the size of the sample and the location and shape of the V notch. Despite the fact that the
instrument measuring the size (e.g. a calliper) should duly be calibrated and introduces a measurement
uncertainty, the machining tool also has a certain variability. Both should be considered.
NOTE In chemical analysis, a ruggedness test as part of a thorough method validation assesses most of the
deviations occurring from implementation of a method under real, and varying within specified limits, laboratory
conditions.
All of these calibrations will normally not introduce real corrections to the measured value but
contribute to the overall uncertainty which makes the approach compatible with the VIM definition.
It is commonly accepted that the combination of metrological traceability and proper measurement
[8]
uncertainty is the only way how measurement results can legitimately be compared. Moreover,
measurement uncertainty estimation of the calibration steps is a mandatory prerequisite for the
establishment of traceability. The concept of calibration against a nominal requirement closes the gap in
cases where the routes to measurement scales (SI and others) are considered “indirect”. Demonstration
of compliance with nominal requirements has to be carried out using measuring instruments which
are, for the measurand they tackle, traceable to the corresponding unit of the scale (callipers, balances,
volumetric flasks, etc.). This concept is formalised as approach B under Clause 4.
The significance and influence of indirect pathways to measurement scales is also recognized in the Joint
[1]
Declaration on Metrological Traceability stating that “In general, … references are the International
System of Units (SI), but where such traceability is not yet feasible, measurement results should be traceable
to other internationally agreed references…”
It might seem viable to attribute all the peculiarities of the measurement method or procedure to the
definition of the measurand as proposed in Reference [9]. In general, one should remember that the
VIM defines the measurand as the “quantity intended to be measured”, not as the procedures necessary
for accomplishing the intention. At the same time, the faults that might happen when the quantity is
measured are not considered.
Two other points have to be considered.
— Firstly, the approach of Reference [9] will work only with one single method of determination which
then (according to the requirements of ISO Guide 34) should be a primary method, a restriction
limiting the applicability of the approach to special cases.
— Secondly, it will limit the field of application of the material or its commutability and, thus,
considerably reduce its technical and commercial value.
A sensible and balanced distribution of method impact on the measurement result between the definition
of the measurand and the traceability chain(s) to units of scale is therefore crucial (see also 5.4).
4 © ISO 2016 – All rights reserved
3.4 Traceability networks
NOTE 4 in the definition of metrological traceability (Clause 2) suggests that for measurements
with more than one input quantity in the measurement model (a situation which is daily practice in
chemical analysis and virtually all fields of testing), each of the input quantity values should itself be
metrologically traceable and the calibration hierarchy involved may form a branched structure or a
network. The effort involved in establishing metrological traceability for each input quantity value
should be commensurate with its relative contribution to the measurement result.
Figure 1 — “Horizontal” traceability network relating the measured quantities in a
measurement procedure model to a set of SI units
It is assumed that this explicitly allows a “horizontal” networked strategy for establishing the
traceability of a particular implementation of a measurement procedure as visualized in Figure 1.
The concept formalized under approach B of Clause 4 is a consequence of NOTE 4 to the definition of
metrological traceability and implements this “horizontal” strategy. It is the only feasible approach
when no higher-order reference is available in the vertical direction.
3.5 Properties expressed in units of measurement scales other than the SI
Most of the scales other than the SI have nevertheless similarities with the latter, namely that they are
realised by a single or a set of materialisations/artefacts with assigned values expressed predominantly
in real numbers (e.g. pH scale). Furthermore, fractions or multiples of the basic unit exist. Some scales
use ordinal numbers which express a cardinal “smaller-larger” relationship between the realizations of
the points on the scale (e.g. Mohs hardness, see Reference [10]) rather than an explicit proportionality
or a counting result. Here, specific problems with the resolution of the scale may arise. However, for
establishment of traceability to these scales the same rules and recommendations may be applied as
given in this Technical Report for traceability to the endpoint SI.
A prominent, widely used non-SI measurement scale is the series of natural numbers. It is the basic
measurement scale in all areas of measurement where counting is involved, e.g. of specified objects
(pollen in a certain amount of air, E.coli bacteria in a specified volume of a food product, etc.). The
peculiarity of this scale is that no materialization of the unit exists to which traceability might be
established by direct comparison (calibration).
On the other hand, the unit “unity” can hardly be misinterpreted, and an interpretation of the unit in
one implementation will exactly match the interpretation in any other. The problems are rather object-
specific (misidentification of objects, double-counting, object overlap, etc.), and the resolution of the
scale always has to be taken into account. Furthermore, depending on the kind of the objects and the
background of the measurement, statistics different from those commonly accepted for continuous-
scale measurements may apply. Thus, the procedure and its specifications gain even more importance
for measurement results expressed in the unit of this scale, and the impacts of deviations from the
specified procedure has to be thoroughly evaluated. However, also with these peculiarities, the
principles of Clauses 4 and 5 also apply here.
3.6 Properties other than quantitative
It is assumed that this is covered by NOTE 1 in the definition of metrological traceability (Clause 2)
which uses the term “non-ordinal quantity” meaning that these measurement results are traceable to
the measurement procedure alone. However, more guidance is needed for the distinction of ordinal
and non-ordinal quantities, in particular with a specific view to the fact that modern measurement
procedures virtually always involve measurements of fully quantifiable (and, thus, not non-ordinal)
quantities and some nominal-valued decisions on the set of quantified measurement results.
Examples are identity of a substance which materializes through measurement, and the sequence of
objects which involves identity and a series of ordinal numbers.
This Technical Report solely deals with quantitative results.
3.7 Summary of an ISO/REMCO event on metrological traceability
In June 2012, ISO/REMCO held a brainstorming session on recent views and approaches to (C)RM
th
traceability in Vienna/Austria, in connection with its 35 General Assembly. The session could be
joined by interested parties world-wide via on-line (video and telephone) connections.
Issues covered during the session included the role of (C)RM in establishing traceability and traceability
statements in certificates. Three major perspectives have been considered in presentations, namely
those of users of (C)RM, the accreditation bodies, and the reference material producers (RMP). The
presentations and a summary are available on the ISO/REMCO webpage.
The user’s perspective referred to the field of geoanalysis (in particular XRF analysis of minerals) and
emphasized a) the use of (C)RM specifically tailored for use in the area and b) the impact of calibration
pathways (standard solutions versus matrix (C)RM), in particular unresolved inconsistent results
when using different calibrants.
Accreditation bodies (AB) identify increased and evolving expectations to RMP (accredited under
ISO Guide 34), statements of traceability (required under ISO Guide 34), additional information on the
certification procedure (certification report), and the intended use of the (C)RM. Users (accredited
under ISO/IEC 17025) are required to describe the role of (C)RM in establishing the traceability of
their results. (C)RM are critical consumables requiring a specific traceability evaluation if not sourced
from accredited RMP or a material included in KCDB Appendix C or the JCTLM RM database. It was also
stated that the AB implementation policy needs to be consistent with respect to traceability.
An RMP scrutinized the role of (C)RM in the delivery of traceability, providing examples of how (C)
RM can be used to validate results including measurement uncertainty assignments, demonstrate the
equivalence of measurements, establish comparability (in the VIM sense) to a measurement scale, and
evaluate and correct for bias.
Fully consistent with this Technical Report, participants of the event concluded that traceability
— cannot be established to an institution,
6 © ISO 2016 – All rights reserved
— establishes comparability, not necessarily trueness of results, and
— is defined by, and potentially limited to, the certification method (for method-defined measurands).
Full information from the RMP on the certification procedure and the intended use is critical and
should be required by the users (e.g. in the form of a certification report). The challenge of propagating
a quantity through a traceability chain when the measurand changes is generally underestimated (see
3.3), implying a need to check with care for inconsistencies between “claimed” and “actual” quantity
measured.
The following conclusions have been drawn.
— Current practice of reporting traceability on CRM certificates is very often not consistent or
sufficient.
— A need for evaluation (by RMP and accreditors) of traceability statements to “higher-order”
references (including but not limited to SI) exists.
— Traceability statements should avoid generic claims, and a concise summary of the technical
basis/certification procedure used to obtain the property values should be available (at least on
request of users or accreditors).
— Intended-use statements are a critical component for appropriate use by end users.
— Further work is needed on minimum requirements for the content of certificates and supplementary
information. For possible approaches, see Clause 6.
4 Approaches to metrological traceability of (C)RM
4.1 General
Given the definition and the considerations made in Clauses 3 and 4, two approaches to (C)RM
traceability seem feasible in principle. Note that their citation (as A and B) should in no way be confused
with preferences given, or hierarchies attributed to, the approaches.
4.2 Approach A
According to NOTE 2 in the definition of metrological traceability (Clause 2), metrological traceability
requires an established calibration hierarchy. One might be tempted to define (C)RM as being endpoints
of the traceability chain which do not need further traceability “upwards”. In particular, one might look
at (C)RM as being artefacts which establish their own measurement scales.
NOTE A small number of recent CCQM Key Comparisons might be interpreted into this direction although
basically covering other aims. CRM from different NMI representing different amounts of substance of the same
analyte in the same matrix have been compared for equivalence, thus, establishing a “scale” of this specified
analyte in the specified matrix, say NO in N , or amounts of ethanol in water.
The approach has a number of undoubted advantages and is used for the reference standards produced
and issued by the World Health Organization (WHO). Both the measurement uncertainty calculation
and the traceability chain start from the internationally accepted reference standard. The validity of
[11]
the standard is assured by a series of technical and assessment procedures, namely:
— The need is recognized by scientific and medical community worldwide and a case formally made
by the WHO Secretariat to the Expert Committee on Biological Standardization (ECBS) on the basis
of public health impact.
— Working groups of experts are involved in setting the priorities and characteristics for selection of
the candidate reference preparations.
— An international collaborative study has to be carried out before any candidate reference preparation
can be considered for establishment by the WHO ECBS.
— The goal of such a study is to determine which candidate material is suitable to serve as a WHO
reference material for the standardization of a biological product or of an in vitro diagnostic tests.
— An internationally agreed unit is attributed to the first WHO Biological Reference Material for
biological activity characterization. The continuity of such a unit is ensured by replacement with a
new batch of reference material which is calibrated against the first or previous reference material.
— A requirement to be met by any batch of a WHO Biological Reference Material is that the content
in every ampoule in the batch should be identical in terms of composition, quantity, potency and
stability.
— The Biological Standardization document which reports the international multi-method collaborative
study is peer-reviewed before being submitted to the WHO ECBS. It has to be approved by the ECBS
for final release of the material.
[12]
An overview of the fields of analytes and measurement areas covered by WHO standards is given
in Annex C, and an example of accompanying documentation for a primary standard shown. The WHO
creates as many scales as is needed in reality.
Propagating and multiplying this fit for the specific purposes approach to the full spectrum of
measurement and testing activities would create a very large (at least infinite) number of scales, one
for each feasible analyte-matrix (in chemistry) or property-of-substance combination (in testing).
Although this treatment of the requirement for calibration hierarchies does also not fully coincide with
the network idea of NOTE 4 in the definition of metrological traceability (Clause 2), where calibrations
may be at the same level but go into different “horizontal” directions, there are situations (in particular
for qualitative-property RM) which will make approach A mandatory. This might be covered by format
D in 6.2. However, this is not further elaborated in this Technical Report (see also 3.6).
4.3 Approach B
[20]
This approach is fully in line with the principles of the ISO Guide 35 and described by the options
given under ISO Guide 35:2006, 9.2. Traceability is stated “given the specified measurement procedure
used”, i.e. the pathway and the endpoint of the traceability chain have to be specified. The specification
of the procedure is nominal. Any implementation in a specific place will cause deviations from the
nominal prescription. These need to be assessed in specific investigations virtually considered as
calibrations against the nominal (written) standard. For more detailed considerations, see the example
in Annex A.
The complete model formula describing the measurement procedure, at least for the majority of
measurement procedures commonly used in chemical analysis, but also in testing, takes the form (see,
for example, Reference [13]):
k
p
∏
i
n
i=1
x =⋅ Fkwith 1 ≤≤ mn≤ (1)
∏
meas i
m
im=+1
p
∏
i
ik=+1
where the p represent the directly measured/determined (explicit) parameters, and the F the influences
i i
from grouped/combined sources, both with and without corrective influence (i.e. values differing from
unity or not), and all with uncertainty contributions. Special cases, where model formulae include
additive terms or are non-linear, have to be considered separately. However, the division into directly
observed and indirect influential parameters equally applies.
The F influences may, for example, represent the (used for correction or not) bias term from a calibration,
i
or the degree of compliance with a nominal prescription. As compliance is normally assumed, the value
assigned to the corresponding F term would be unity, but the uncertainty connected with this term
8 © ISO 2016 – All rights reserved
would be assessed in a virtual “calibration” procedure by deliberately deviating from the prescription
and evaluating the influence on the measured value. In method validation, this is normally called
robustness or ruggedness test. Note that the model equation as above can be treated as a consequence
of NOTE 4 in the definition of metrological traceability (Clause 2).
With the model formula above, the relative uncertainty attributed to the measurement result would be
m n
22 2
ux()=+up() uF() (2)
∑∑
rrmeas i ri
i==1 im+1
except possible correlation terms which have to be accounted for if the influential parameters are not,
or not fully, independent of each other.
NOTE For RM coming without a (measured) value, see 3.2.
5 Establishing traceability of (C)RM property values (Approach B)
5.1 Principles
Under approach B, the traceability of a (C)RM can be established within a framework of reasonable
interpretations of the basic VIM definition, namely:
— Traceability of the assigned value of a (C)RM is a property of this value whereby the single or a
consolidated set of measurement results obtained for the (C)RM can be related to a single or a
set of references through a documented unbroken chain of calibrations, each contributing to the
measurement uncertainty.
— A reference can be an established and well-understood method/procedure, a written description
and specification of an operational procedure, or an artefact/artefacts realizing a point or a range
on the measurement scale of the quantity under consideration. Explicit reference has to be made
to the measurement procedures applied by citing the standard, protocol (e.g. AOAC Peer Reviewed
Method), publication or text book.
— Members of a set of references may refer to the same quantity (when defining a measurement
scale) or of different but essential in the measurement process quantities (realizing a traceability
network). This is particularly important for more-dimensional measurands (e.g. spectra).
— A consolidated set of measurement results is a measurement result combined from several
measurement results using appropriate procedures which assure full compatibility between the
results combined and the result consolidated from the former, all within an appropriately assessed
uncertainty. The establishment of the total measurement uncertainty budget of the consolidated
value follows the commonly accepted rules and includes allowances for the “procedure impact”
according to Clause 4.
Compatibility between the results combined may arise per se or may have to be established by
introducing uncertainty components which account for data discrepancies. Decision on the level of
admissible discrepancy is case-sensitive and subject to expert judgment, meaning that starting from
a certain level of discrepancy, the measurement results may seem non-commensurable (and thus no
longer traceable to the same or the same set of references).
Establishment and statement of traceability of a CRM is mandatory; it is not for an RM fulfilling the
basic requirements only (i.e. coming without an assigned value). However, if any values are assigned to
an RM, their traceability should be assured as well (see ISO Guide 33:2015, 6.4.2).
5.2 Traceability pathways
ISO Guide 34 and ISO Guide 35 in their current editions accept four general approaches for the
characterization of reference materials, namely
a) measurement by a single (primary) method in a single laboratory;
b) measurement by two or more independent reference methods in one laboratory;
c) measurement by a network of laboratories using one or more methods of demonstrable accuracy;
d) a method-specific approach giving only method-specific assessed property values, using a network
of laboratories.
Traceability pathways may have different targets, namely
1) (direct) traceability to the unit of the measurement scale,
2) traceability to the unit of the measurement scale via, and given by, a measurement method or
procedure, and
3) traceability to a protocol.
As a rule, pathways and certification schemes combine as shown in Table 1.
Table 1 — Certification/characterization schemes and pathways of traceability
Scheme Pathway
1) 2) 3)
a) X — X
b) X X -
c) — X -
d) — X X
5.3 Steps in establishing traceability
5.3.1 General
[20]
General steps to be taken, and provisions to be made are described in ISO Guide 35 and should be
followed strictly. Major points are as follows:
— Transformation (of the measurand): Although the determination of the property value itself can
be made traceable to appropriate units through, for example, calibration of the measurement
equipment used, steps like the transformation of the sample from one physical (chemical) state to
another cannot. Such transformations may only be compared with a reference (when available), or
among themselves. For some transformations, reference methods have been defined and may be
used in certifica
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.
Loading comments...