A comparative analysis of the statistical and random-fuzzy approaches in the expression of uncertainty in measurement

The present practice for uncertainty expression and estimation in measurement, endorsed in the IEC-ISO Guide to the Expression of Uncertainty in Measurement, is based on a statistical approach, which is also the basis for the Monte Carlo method generally employed to overcome the problems met in the strict application of the guide. More recently, methods based on the fuzzy theory have been proposed too, with encouraging results. This paper compares the results obtained, in the expression of uncertainty, by the use of the Monte Carlo method and the random-fuzzy variable method. Both methods are applied to a real, digital signal processing-based instrument for electric power quality measurement, and the obtained results are compared and discussed.