Version HTML de base
“Uncertainty-noise” Le Mans
5
Acoustique
&
Techniques n° 40
any important decisions are based on measurement
or test results : the results can be used for example to
assess a noise level, to check material against specification
or statutory limits…Whenever decisions are based on
measurement, it is important to have some indication on
the quality of the results. The uncertainty of measurement
is the simplest way to express the reliability of the result.
Since the standard ISO / IEC 17025 “General requirements
for the competence of testing and calibration laboratories”
has been published in 2000, the concept of uncertainty of
measurements or test results seems to be a key issue for
laboratories and their clients. Up to now, only calibration
laboratories were deeply involved in uncertainty evaluation.
Theconcept of uncertainty is certainly difficult tounderstand
and to admit by people not familiar with technical activities.
It is now recognized in the testing community that it is as
important to communicate the uncertainty related to a
specific measurement as it is to report the measurement
result itself. It is also clear that a measurement or a test
result without an assessment of its reliability is completely
useless. Without knowing uncertainty it is impossible to
assess the comparison between different measurements
with the same parameter or to compare a result with a
specification limit. Measurement and test results represent
the basic information for the conformity statement of many
products or activities.
The evaluation of uncertainty is often regarded as a
difficult and arduous task. The Guide to the Expression of
Uncertainty in Measurement (GUM) provides international
agreed concepts, recommendations and a procedure
for the evaluation of uncertainty. consistency with the
GUM requires the basic concept to be accepted and the
recommendation to be followed.
The conference will present the three levels of reading of
the GUM: Concepts, recommendations and a procedure.
The different methods and tools for the evaluation of
uncertainty components will also be introduced: modelling
approach, single laboratory validation approach, inter-
laboratory validation approach and proficiency testing
approach. These presentation will be based on the new
International or European document EA 4/16 EA Guidelines
on the expression of uncertainty in quantitative testing
and ISO / TS 21748 Guidance for the use of repeatability,
reproducibility and trueness estimates in measurement
uncertainty estimation.
The three levels of reading of the GUM
The concept of uncertainty and principles are defined
in the GUM (Guide for the Expression of Uncertainty in
Measurement). GUM [1] is based on sound theory and
provides a consistent and transferable evaluation of
measurement uncertainty.
The approaches for measurement
uncertainties evaluation
Marc Priel
Laboratoire national de métrologie et d’essais (LNE),
1 rue Gaston Boissier
75724 Paris CEDEX 15
FRANCE
Different methods and tools are now available for measurement uncertainty
evaluation. These new methods comply with the concepts and recommendations
of Guide to the expression of uncertainty in measurement (GUM). During the
presentation, the author will introduce several alternatives for laboratories and
notably those based on inter-laboratory comparisons : modelling approach, single
laboratory validation approach, inter-laboratory validation approach and proficiency
testing approach. The new approaches for uncertainty evaluation will certainly
streamline the measuring results or test reports, their comparability and their
traceability to SI units.
Plusieurs outils et méthodes sont actuellement disponibles pour évaluer
l’incertitude de mesure. Ces nouvelles méthodes obéissent essentiellement aux
concepts et aux recommandations du GUM (Guide to the expression of Uncertainty
in Measurement).
Dans cet article, l’auteur présente plusieurs solutions d’évaluation à l’attention des
laboratoires et notamment celles concernant les comparaisons entre laboratoires.
Il s’agit des approches par modélisation, par validation dun seul laboratoire,
par validation inter-laboratoires et enfin de la démarche des tests d’aptitude.
Ces nouvelles démarches d’évaluation de l’incertitude permettront sûrement
de rationaliser les résultats de mesure ou les rapports d’essais, ainsi que leur
comparabilité et leur traçabilité aux unités SI.
M