EARTHQUAKE RESILIENCE ASSESSMENT AND IMPROVEMENT FRAMEWORK OF THE HIGH-SPEED RAILWAY SYSTEM

High-speed railway (HSR) transportation is one of the most critical ways for passenger and freight transportation between cities/communities in many countries. However, earthquakes can cause significant damage to the entire HSR system, since interruption of any part will decrease the performance of the whole HSR system and further hamper the transportation of medical personnel, the wounded, and emergency supplies, causing more economic losses and human casualties. The paper proposes an urban earthquake disaster simulation method based on machine learning and a function evaluation and improvement framework of the HSR system. Where the HSR system infrastructure and urban building portfolio’s earthquake damage, and further, the transportation needs for the wounded, medical personnel, and emergency supplies could be obtained in near real-time by the urban earthquake disaster simulation method. Optimization models are proposed to improve the usage of the HSR system's inherent function each day in the function evaluation and improvement framework, where the train operation and passenger rerouting plan with the objection of least impact on passenger travel and maximize meet the disaster relief transportation demand post-earthquake are obtained. Moreover, to enhance resilience in the recovery process, the optimized repair sequence of damaged infrastructure in the post-earthquake recovery period is generated using a heuristic algorithm. A case study of the proposed model is applied to the great bay area, China, the result reveals that the method can effectively and accurately predict the earthquake damage in the study area, and assess and improve the resilience of the HSR system during recovery.