ABSTRACT
In this contribution, an existing post-tensioned concrete bridge is considered in a probabilistic framework. The most severely corroded bridge girder is analysed using an extended Direct Stiffness Method (DSM) tool, circumventing the use of advanced finite element models and hence facilitating the implementation of the developed framework in daily engineering practice. Therefore, the traditional DSM approach is extended to accommodate for (corroded) post-tensioned elements. A probabilistic framework is constructed which is able to predict the reliability index of the element, as well as the end-of-service life of the girder. Currently, the considered bridge is part of a large-scale monitoring campaign, continuously measuring the strains along the bridge. Hence, the approach is coupled with Bayesian updating, which can reduce the uncertainties on the input and output of the DSM tool. As a result, more accurate predictions can be made of the structural performance of the girder, contributing to the decision-making process regarding the considered bridge.
