On the Use of Numeric Integration for Uncertainty Evaluation in Indirect Measurements

Abstract

According to the last revision of international recommendations, numerical methods are nowadays claimed/or the estimation of measurement uncertainty in indirect measurements. The paper, in particular, proposes the use of numeric integration of measurement models as a fast and reliable method for uncertainty estimation. Starting from the knowledge of the probability density functions of the input quantities, the method applies traditional techniques of numeric integration to the measurement model in order to achieve an estimate the output quantity variance, and, consequently, of the output standard uncertainty. A number of tests have been carried out to assess the performance of the proposed method. In particular, the concurrence between the estimates of output quantity expectation and standard uncertainty provided by the method and those granted by Monte Carlo simulations or method based on unscented transform has been verified along with a comparison of computational times. The obtained results highlight the efficacy of the method and suggest it as an attractive alternative to other approaches currently adopted.