An Original Fuzzy Method for the Comparison of Measurement Results Represented as Random-Fuzzy Variables

Generally, the final result of a measurement process is used to make decisions according to the result of a comparison between the measurement result and a threshold, which could be a single numerical value or another measurement result. Of course, the latter case is the more general one since the single-value threshold is, from a metrological point of view, a measurement result with zero associated uncertainty or, more specifically, a measurement result whose associated uncertainty is much lower, or negligible, with respect to the uncertainty associated to the other measurement result. Moreover, in recent years, because of the great diffusion of instruments based on digital signal processing, decisions can be required not only at the end but also within the measurement process itself so that the comparison of measurement results becomes part of the measurement process itself. The comparison of measurement results or, in other words, decision making, is a very important task. This paper proposes a method based on the use of suitable fuzzy operators for comparing measurement results, which are expressed in terms of random-fuzzy variables, taking into account measurement uncertainty.