Int. J. Metrol. Qual. Eng.
Volume 3, Number 2, 2012
|Page(s)||71 - 77|
|Published online||14 November 2012|
Approaches to evaluating measurement uncertainty
National Physical Laboratory,
Accepted: 27 June 2012
The Guide to the expression of measurement uncertainty, (GUM, JCGM 100) and its Supplement 1: propagation of distributions by a Monte Carlo method, (GUMS1, JCGM 101) are two of the most widely used documents concerning measurement uncertainty evaluation in metrology. Both documents describe three phases (a) the construction of a measurement model, (b) the assignment of probability distributions to quantities, and (c) a computational phase that specifies the distribution for the quantity of interest, the measurand. The two approaches described in these two documents agree in the first two phases but employ different computational approaches, with the GUM using linearisations to simplify the calculations. Recent years have seen an increasing interest in using Bayesian approaches to evaluating measurement uncertainty. The Bayesian approach in general differs in the assignment of the probability distributions and its computational phase usually requires Markov chain Monte Carlo (MCMC) approaches. In this paper, we summarise the three approaches to evaluating measurement uncertainty and show how we can regard the GUM and GUMS1 as providing approximate solutions to the Bayesian approach. These approximations can be used to design effective MCMC algorithms.
Key words: Measurement uncertainty / Monte Carlo / Markov chain Monte Carlo
© EDP Sciences 2012
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.