An analytical probability distribution model for extreme bridge traffic load effects based on WIM data and extreme value theory

Abstract Accurate prediction of extreme bridge traffic load effects is essential for ensuring structural safety under increasing heavy-vehicle traffic. This study develops an analytical probability distribution model that directly links traffic condition parameters, bridge configuration parameters, and vehicle weight statistics to the parameters of the Type III Extreme Value Distribution (EVD). Using WIM data and MCS, short-term measurements were extended to long-term traffic, and bridge load effects were obtained through influence-line analysis. By correlating the EVD parameters with site-specific characteristics, semi-empirical analytical equations were derived that generalize Bailey’s model from European bridges to bridges in China. Results show that upper bound parameters remain consistent with theoretical limits, while the shape and position parameters require regional calibration due to heavier truck distributions. Validation using measured and simulated data confirms the model’s accuracy and applicability. The proposed analytical model provides a novel, efficient framework for region-specific prediction of extreme traffic load effects and supports reliability-based bridge design and assessment.