Estimating a probability of failure with the convex order in computer experiments

TL;DR

This paper introduces a convex order inequality between two Bayesian estimators for failure probability in expensive black-box models, enabling comparison of their efficiency and guiding experiment design.

Contribution

It presents a novel convex order inequality between failure probability estimators and develops a sequential design method based on this principle for computer experiments.

Findings

Convex order inequality compares estimator efficiency.

Proposed sequential design improves experiment efficiency.

Application to real case demonstrates practical benefits.

Abstract

This paper deals with the estimation of a failure probability of an industrial product. To be more specific, it is defined as the probability that the output of a physical model, with random input variables, exceeds a threshold. The model corresponds with an expensive to evaluate black-box function, so that classical Monte Carlo simulation methods cannot be applied. Bayesian principles of the Kriging method are then used to design an estimator of the failure probability. From a numerical point of view, the practical use of this estimator is restricted. An alternative estimator is proposed, which is equivalent in term of bias. The main result of this paper concerns the existence of a convex order inequality between these two estimators. This inequality allows to compare their efficiency and to quantify the uncertainty on the results that these estimators provide. A sequential procedure for the construction of a design of computer experiments, based on the principle of the Stepwise Uncertainty Reduction strategies, also results of the convex order inequality. The interest of this approach is highlighted through the study of a real case from the company STMicroelectronics.