ABSTRACT
This paper aims to develop a system reliability analysis method to accurately determine the time-dependent changes in the probability of failure of a girder system. The system reliability analysis incorporates the spatial variability of corrosion rate, the effect of corrosion on mechanical properties and multiple failure modes. Notably, the spatial variation of corrosion rate has been determined through random field theory with conditional sampling. The degradation effect of corrosion on mechanical properties has been determined through in-situ destructive tests and lab-based mechanical tests. The girder system is subdivided into different girder segments for system reliability analysis and time-dependent changes in the probability of failure of each girder segment under multiple failure modes are determined through Monte Carlo simulation. A working example is then provided to demonstrate the application of the proposed methodology. The paper concludes that the developed method can assist bridge engineers and asset managers in accurately predicting the failure of a bridge network and minimizing the costs of bridge rehabilitation and maintenance.
