An advanced reliability procedure for lifetime assessment of structures: application to reinforced concrete beams

Abstract

Lifetime assessment of civil engineer structures is a framework which helps operators in performing its maintenance by evaluating their behaviour during the whole life. In order to do so, a numerical model, which should be a good representation of reality, is usually considered. This paper presents a methodology, based in numerical identification and Bayesian inference techniques, that provides a likely probabilistic model. This is validated with a set of reinforced concrete beams which were loaded up to failure in laboratory. Obtained results point out the advantages of integrating it within such a framework, in detriment of deterministic models.