Evaluation of measurement uncertainty and its numerical calculation by a Monte Carlo method

TLDR

The Guide to the Expression of Uncertainty in Measurement (GUM) is the de facto standard for evaluating measurement uncertainty, and a probability‑density‑function based Monte Carlo approach has recently been proposed as an alternative. This study outlines the relationship between the GUM standard method and the PDF‑based Monte Carlo approach. The authors describe the Monte Carlo algorithm for numerically evaluating the PDF approach and demonstrate it with two illustrative examples. For both examples, the Monte Carlo results are compared with GUM results, showing the differences between the two methods.

Abstract

The Guide to the Expression of Uncertainty in Measurement (GUM) is the de facto standard for the evaluation of measurement uncertainty in metrology. Recently, evaluation of measurement uncertainty has been proposed on the basis of probability density functions (PDFs) using a Monte Carlo method. The relation between this PDF approach and the standard method described in the GUM is outlined. The Monte Carlo method required for the numerical calculation of the PDF approach is described and illustrated by its application to two examples. The results obtained by the Monte Carlo method for the two examples are compared to the corresponding results when applying the GUM.