ABSTRACT
The network level visual inspection of bridges conducted by road agencies in Australia and internationally, generally consists only of discrete condition data. The quantitative links between condition state ratings and various deterioration mechanisms are missing. In this paper, an integrated model is developed in order to link a condition based data-driven model using processed level 2 inspection data, with flexural cracking behavior. The probabilistic Markov Chain process is used to model the discrete data, and is applied on reinforced concrete u-slab bridges with crack defects. A Finite Element Model (FEM) based on a case study is developed and validated with results from Structural Health Monitoring (SHM) of the bridge, with a superload passing through. This FEM is then used to model cracking behavior by increasing levels of superload, through push-down static testing. The cracking model and crack patterns are used to then extract values in order to calculate crack widths according to condition states. These are subsequently integrated with the condition based data-driven model, in order to determine cracking of u-slab bridges over time, thus creating a durability model. The degrees of cracking predicted in this case have more relevance to durability and serviceability rather than load rating. The outcomes of this research can be used for optimizing maintenance and repair activities performed by road authorities, particularly crack repair planning and budgeting, given a bridges age and environmental exposure.
