Hostname: page-component-7b7567c8c-8tzhz Total loading time: 0 Render date: 2026-11-19T16:06:15.329Z Has data issue: false hasContentIssue false

Approaches to evaluating measurement uncertainty

Published online by Cambridge University Press:  14 November 2012

A.B. Forbes*
Affiliation:
National Physical Laboratory, TW11 0LW, Teddington, UK
*
⋆ Correspondence:Alistair.Forbes@npl.co.uk

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the 'Save PDF' action button.

The Guide to the expression of measurement uncertainty, (GUM, JCGM 100)and its Supplement 1: propagation of distributions by a Monte Carlomethod, (GUMS1, JCGM 101) are two of the most widely used documents concerningmeasurement uncertainty evaluation in metrology. Both documents describe three phases (a)the construction of a measurement model, (b) the assignment of probability distributionsto quantities, and (c) a computational phase that specifies the distribution for thequantity of interest, the measurand. The two approaches described in these two documentsagree in the first two phases but employ different computational approaches, with the GUMusing linearisations to simplify the calculations. Recent years have seen an increasinginterest in using Bayesian approaches to evaluating measurement uncertainty. The Bayesianapproach in general differs in the assignment of the probability distributions and itscomputational phase usually requires Markov chain Monte Carlo (MCMC) approaches. In thispaper, we summarise the three approaches to evaluating measurement uncertainty and showhow we can regard the GUM and GUMS1 as providing approximate solutions to the Bayesianapproach. These approximations can be used to design effective MCMC algorithms.

Information

Type
Research Article
Copyright
© EDP Sciences 2012