Digital Twins for Bridge Health Monitoring

Abstract

Structural integrity and safety are paramount for bridges, vital components of our transportation infrastructure. Traditional structural health monitoring methods often struggle to manage and interpret large, heterogeneous datasets from sensors, inspections, and environmental sources. Digital twins (DTs) offer a dynamic solution, acting as virtual replicas of physical bridges that continuously update with real-time data and enable predictive insights. This research synthesizes contributions up to 2020, covering approaches that integrate Building Information Modeling (BIM), finite element (FE) models, SensorML frameworks, and statistical and physics-based modeling. A notable study by Ye et al. (2019) lays a comprehensive foundation, exploring real-time data management via BIM, physics-driven FE modeling, data-driven analytics, and hybrid frameworks synthesizing these paradigms for actual railway bridges dpi-proceedings.comIOPscience. Similarly, Dang et al. (2018) present a digital twin maintenance strategy combining CAD data, inspection records, and photographic mapping to monitor deterioration IOPscience. Methodologies like BrIM, BIM–SensorML integration, federated models, and Gaussian Process-based temperature behavior prediction underscore evolving complexity in DT frameworks IOPscience. These digital twins support visualization, condition assessment, simulation, and risk-based ―what-if‖ analyses, transforming passive monitoring into proactive management IOPscience. This abstract sets the stage by highlighting how DT architectures established before 2020 laid the groundwork for smart, responsive, and predictive bridge health monitoring systems.