Limitations and Future Direction for Structural Health Monitoring of Railroad Bridges Subjected to Over-Height Vehicle Impacts

Abstract

Low clearance through-plate girder railroad bridges in the United States are highly susceptible to impacts from over-height vehicles, which can lead to structural damage and disruptions in railroad bridge service. Current post-impact inspection protocols require bridge closures, causing unnecessary delays when minor impacts occur. This study presents a low-cost structural health monitoring-oriented framework for automated impact detection, severity classification, and condition assessment using wireless smart sensors. The proposed structural health monitoring framework leverages machine learning models to detect and characterize impact events and estimate resulting residual displacements using acceleration data alone. Although this framework demonstrates promising performance in both numerical and field settings, some limitations remain. The current approach has been validated only for single-span through-plate girder bridges, relies solely on acceleration data, and does not incorporate vehicle speed or visual confirmation. Future directions include integrating additional sensing modalities such as cameras for event verification and visual assessment, expanding applicability to multi-span or alternative bridge types, and incorporating traffic history and near- miss data to support predictive impact risk modeling. These extensions will further enhance the utility, scalability, and resilience of automated post-impact response strategies for railroad infrastructure.