Multi‐level fatigue reliability assessment of reinforced concrete railway bridges

Abstract This paper presents a multi‐level reliability framework for assessing the fatigue life of reinforced concrete (RC) railway trough bridges subjected to cyclic loading. The framework incorporates increasing levels of analytical complexity and real‐world data in four steps. First, an analytical model applies S–N curves and the Palmgren–Miner rule with constant stress assumptions. Second, monitored strain data refine stress estimates. Third, a calibrated finite element (FE) model is used to simulate degradation and structural response. Fourth, survival information conditions the reliability on observed performance. The framework is applied to a RC trough bridge tested under representative railway loading using traffic data from Sweden's Iron Ore Line. Results demonstrate the value of combining monitoring, FE modeling, and probabilistic methods for evaluating remaining service life (RSL). From step 1 to step 3, the methodology extended the RSL estimates by 39 years, allowing an increase in mean axle load by approximately 20%.