Research on a Preliminary Assessment Model of Seismic Disaster Risks for Bridges

Bridges, as critical nodes in transportation systems, are essential for postearthquake emergency response and recovery. Although existing models (such as the Zhu Meizhen empirical model and dynamic analysis methods) are widely used for rapid seismic damage assessment of regional bridges, they typically assign equal weights or use empirical weighting for seismic impact factors, failing to scientifically quantify their relative importance. This limitation results in low discriminative accuracy of the assessment outcomes. To address this issue, this study proposes a new preliminary regional bridge seismic risk-assessment model. By integrating the fuzzy analytic hierarchy process (subjective weighting) and the entropy weight method (objective weighting), a combined weighting system that combines expert experience with actual seismic damage data is constructed, achieving scientific quantification of the contributions of multiple indicators. Based on readily available bridge data, the model establishes a standardized and updatable risk-assessment process. Verified by typical historical earthquakes such as the Tangshan and Lushan earthquakes, the model achieves AUC values exceeding 0.9 for all four risk categories, demonstrating excellent discriminative accuracy and classification performance. While maintaining high screening efficiency, it significantly improves the accuracy of assessment compared with previous empirical models and can effectively identify the weak links in seismic resistance. This study provides a reliable basis for pre-earthquake disaster prevention planning and postearthquake emergency resource optimization for regional bridges.