Updating Performance and Reliability of Concrete Structures Using Discrete Empirical Bayes Methods

Abstract

When performing a probabilistic assessment of the reliability of deteriorating structures, we often need to integrate the results of different inspections in time, within the models used to analyze the progress of deterioration. A new framework is described in this paper. It rests on a special case of the empirical Bayes method where the non-observable parameter is a discrete random variable with a relatively small number of outcomes. Various likelihood functions are derived. They are based on mixtures of deterioration scenarios. It is shown how the method can be used to calibrate the response of a stochastic deterioration model and to merge with a time-dependent reliability analysis. Examples relating to the long-term chloride corrosion in a reinforced concrete slab are presented in the paper.