Uncertainty evaluation from Monte-Carlo simulations by using Crystal-Ball software

摘要

The first Supplemental Guide to the current edition of the Guide to the expression of uncertainty measurement (GUM) deals with the propagation of distributions, and emphasizes the use of Monte-Carlo Simulation (MCS) for estimating the uncertainty of measurands. In order to carry out MCS we need computer programs for generating pseudo-random numbers and for solving numerically the integral equations arising when simulating the values of the specification variables with a given probability density function (PDF). Crystal-Ball commercial software is very suitable for performing MCS and to estimate the expanded uncertainty for the measurand. The use of Crystal-Ball is illustrated with two working examples dealing with specification models of non-linear features and with correlated variables (such as the slope and intercept of calibration straight lines), respectively.