Physics-based predictive assessment and domination of uncertainties: The RelySoft method and software tool

Abstract

Often there is in-depth criticism of random sampling models, such as the Poisson series, for computing the prediction of the behavior of physical systems in operation. In this work we introduce a new method called RelySoft alternative to Monte Carlo and the First Order Reliability Method (FORM) to examine the influence of uncertainties inherent to a physical process in order to calculate the success probability of the same physical process and eventually its evolution over time. The method is based on the introduction of a set of functions and operator over probability distributions; the method has no limits to the number of parameters and allows for the uncertainties of the exponents appearing in equations. Two examples are reported concerning aero-spatial field: a time-independent and a time-dependent case studies.