Probabilistic service life prediction updating with inspection information for RC structures subjected to coupled corrosion and fatigue

•Probabilistic service life prediction based on coupled corrosion-fatigue deterioration.•Updating probabilistic parameters in the coupled corrosion-fatigue deterioration model.•Determination of the most appropriate updating parameters of the deterioration model.•Updating process considering all possible detection results. The performance and remaining service life of deteriorating reinforced concrete (RC) structures decrease with time due to increased critical damage propagation based on coupled corrosion and fatigue. Since damage propagation prediction is highly uncertain, an updating process after each inspection and monitoring is required to improve the accuracy and applicability of the structural performance and service life prediction. This paper proposes a probabilistic approach for updating the service life prediction of deteriorating structures under coupled corrosion-fatigue deterioration processes. The probabilistic parameters involved in the coupled corrosion-fatigue deterioration model are updated using Bayesian inference (BI). In the updating process, the Markov Chain Monte Carlo (MCMC) sampling method is applied to integrate all the possible inspection results associated with no damage detection, corrosion damage, and/or fatigue crack damage. The most appropriate updating parameters of the deterioration model are determined by adopting a comparison-based analysis. Finally, the remaining service life of the deteriorating structures based on the updated coupled damage prediction model is re-estimated. The application of the proposed probabilistic approach is illustrated on an RC bridge girder. This paper shows that (a) the coupled corrosion-fatigue damage can reduce the structural service life compared to corrosion or fatigue damage separately, and (b) the errors and uncertainty in corrosion-fatigue damage prediction can be reduced through the parameter updating integrating multiple inspection outcomes.