Improved reliability approximate method combining Kriging and importance sampling

Approximation methods are widely used to alleviate the computational burden of structural reliability analyses. Engineering problems involve more and more complex computer codes and the evaluation of the probability of failure may require very time-consuming computations. To assess reliability, the most popular approach remains the numerous variants of response surfaces. Widespread in optimization, Kriging has just started to appear in uncertainty propagation and reliability studies. This paper investigates an Optimized Kriging method by using the artificial bee colony algorithm combining importance sampling for structural reliability problems. An example is performed to illustrate the methodology to prove its high accuracy and efficiency, particularly for problems of high non-linearity, high dimensionality and implicit performance functions.