Reliability Analysis based on Bootstrap Information Criteria

Abstract

Reliability is the probability of satisfying the target performance. Reliability analysis is applying the reliability estimation methods to the system to achieve better overall performance. Many methods for evaluation of statistical models like information criteria can be found in literature. Basic Bootstrap Information Criteria (EIC) skips the use of analytic expressions for bias correction and instead uses computer generated calculations to estimate bootstrap samples for the bias correction. Advanced bootstrapping uses variance reduction method for improving the bias correction term. Combination of Bootstrap Information criteria methods with reliability analysis is done and numerical examples are shown considering different approaches.