Comprehensive digital twin for infrastructure: A novel ontology and graph-based modelling paradigm

The inherent uncertainty and complexity of infrastructure systems present significant challenges for precise management. By providing real-time status tracking and decision support capabilities, Digital Twin (DT) has shown significant promise in enhancing decision-making and improving efficiency and quality of management. To create a knowledge-rich and scalable DT capable of tracing the distinctive features and performance of infrastructure, this article develops a comprehensive IDT (Infrastructure DT) modelling paradigm. The ontologybased paradigm comprises five elements: scenario, virtual model, physical entity, relation, and component, supporting interoperability and multidisciplinary collaboration. By employing a graph-based modelling approach, the paradigm integrates various types of discrete data to link infrastructure entities into a cohesive system, thereby effectively capturing their dynamic characteristics. A case study comprising supported foundation excavation, subway tunnels, and pipe networks demonstrates the IDT's capability to facilitate coordination across multidisciplinary, multi-scale, and multi-stage contexts, thereby establishing a transparent, effective, and secure pattern of infrastructure management.