Improved active learning probabilistic approach for the computation of failure probability

This paper presents a cost-effective probabilistic approach to be used in engineering applications. The proposed consists of an improved Kriging-based method aiming at reducing minimum the number evaluations true performance function when computing failure probability. It is kind variant classical Active learning combining Kriging and Monte Carlo Simulation (AK-MCS) developed by Echard et al. (2011) [1], where some improvements are introduced enhance process. Some illustrative practical examples presented discussed. has shown great efficiency as compared AK-MCS approach.