Digital Twin-Based Climate Resilience Assessment for Transport Infrastructures: An Integrated Civil Engineering Framework

Transport infrastructures are increasingly exposed to climate-related hazards, including extreme rainfall, flooding and heat stress, with potential consequences for physical assets and service continuity. Existing Digital Twin applications mainly focus on asset representation, condition monitoring and predictive maintenance, while climate exposure, infrastructure vulnerability and operational consequences are rarely integrated within a single assessment environment. This study proposes a Digital Twin-enabled framework for assessing the climate resilience of railway infrastructure and demonstrates its application to four nodes along a selected railway corridor.The framework integrates three analytical dimensions: climate exposure, infrastructure and operational vulnerability, and adaptive capacity. The assessment considers indicators related to rainfall intensity, flood susceptibility and temperature stress; infrastructure characteristics, asset condition, network redundancy and accessibility; and operational variables, including service frequency, potential delay propagation and recovery capacity. Infrastructure and network characteristics are derived from infrastructure databases and open spatial data, including OpenStreetMap, while environmental and climate information is obtained from Copernicus datasets. Operational characteristics are reconstructed using railway timetable data. The indicators are normalised and combined through the Analytic Hierarchy Process (AHP) to produce a composite resilience index and a comparative classification of the four railway nodes. Current conditions are compared with the SSP2-4.5 climate scenario for 2030. Sensitivity analysis of indicator weights is used to examine the robustness of the node classification and address methodological uncertainty.The application demonstrates how heterogeneous spatial, climate, infrastructure and timetable data can be integrated to identify differences in resilience across the four railway nodes. The framework produces node-level exposure, vulnerability and adaptive-capacity scores, together with an overall resilience classification. It also enables the identification of the factors driving the relative criticality of each node and supports comparison between current conditions and the 2030 SSP2-4.5 scenario. The study contributes a replicable method that extends conventional Digital Twin applications by connecting asset-level climate stressors with their potential operational consequences at railway-node level. The proposed environment supports the identification of critical assets and the prioritisation of maintenance and adaptation measures under future climate conditions.