Evaluating methods of calculating measurement uncertainty

Abstract

This communication demonstrates the need for independent validation when an uncertainty calculation procedure is applied to a particular type of measurement problem. A simple measurement scenario is used to highlight differences in the performance of two general methods of uncertainty calculation, one from the Guide to the Expression of Uncertainty in Measurement (GUM) and one from Supplement 1 to the "Guide to the Expression of Uncertainty in Measurement"-Propagation of Distributions using a Monte Carlo method. The performance of these methods is investigated in terms of the long-run success-rate when applied to many simulated measurements in the scenario. An individual application of the method is deemed successful if an uncertainty interval containing the measurand is obtained. The alternative approach to validation taken in the Supplement, that an uncertainty interval calculated by a Monte Carlo method can be used to validate the GUM method, is not consistent with the results of this study.

Can coverage factor 2 be interpreted as an equivalent to 95% coverage level in uncertainty estimation? Two case studies

“…There have been reports of using MCM for uncertainty calculation [4,5,[14][15][16][17][18] many of them emphasizing its advantages over the GUM uncertainty framework. There has, however also been a report on an opposite finding [19]. Examples on the application of the Bayesian approach about different measurement fields have also been published [20][21][22].…”

Invited Article: Concepts and tools for the evaluation of measurement uncertainty

“…142,143 In some cases this is a consequence of a choice of coverage interval that is not fit for the intended purpose, as Possolo, Toman, and Estler 144 explain. However, since other issues or limitations may arise, 3 next we describe an illustrative, cautionary tale that should be kept in mind when choosing a particular expression of uncertainty to summarize the scatter of the sample drawn from the probability distribution of the output quantity.…”

Type A evaluation of measurement uncertainty when the sample size is not predetermined

“…This presents the objective-Bayesian statistician with a number of related problems when considering Type A evaluation of uncertainty, as we now demonstrate. These problems can be added to a growing list of difficulties associated with the use of Bayesian ideas in the evaluation of measurement uncertainty [12]- [17].…”