Cyber-physical resilience modelling and assessment of urban roadway system interrupted by rainfall

Emerging technologies such as the Internet of Things (IoT) and connected vehicles enable urban roadway system more tightly integrated as a Cyber-Physical System (CPS). Accordingly, cyber-physical resilience brings the system with new mitigation strategy interpretation, specifically when being disrupted by rainfall and its induced waterlogging risk. However, modelling and assessment of cyber-physical resilience are not easy for the complicated interaction between 4I(municipal Infrastructure, human Individuality, vehicle Instrument and network Information). In this paper, a Causal Bayesian Network (CBN) framework is employed for resilience modelling and assessment, which is described by absorptive, adaptive and restorative capacity. Forward propagation, backward propagation and sensitivity analysis are being conducted. For validation, a case study via real-world Radio Frequency Identification (RFID) data is utilized to analyze the system's self-healing process. In the underlying CBN model, absorptive capacity is 74.56%, which exhibits a good fit with RFID data analysis that performs a 75% performance index compared with the original state in heavy rainfall scenario. This study can be of great potential for offering practical opportunities on risk preparedness and management as well as investment decisions, for obtaining a more resilient and sustainable future city.