Resilience-based restoration sequence optimization of disrupted transportation networks: A novel matheuristic approach

Transportation networks are crucial components of modern infrastructure but are highly vulnerable to disruptions caused by frequent, unpredictable disasters, such as earthquakes and rainstorms, which severely compromise connectivity and mobility. Developing resilient restoration plans is thus essential for minimizing disruption impacts and expediting recovery. However, existing approaches primarily depend on experience-driven or importance-based methods, which struggle to identify critical disrupted links and fail to provide optimal sequences. To tackle these challenges, this study proposes a general sequencing framework featuring multi-stage restoration modes and formulates an optimization problem as a mixed-integer nonlinear programming model. To improve computational tractability, a bipartition-based simplification strategy is introduced. Additionally, a novel matheuristic approach combining heuristic flexibility with mathematical programming precision is developed, enabling effective decision-making across diverse scenarios. The framework is validated through the Tongzhou transportation network, demonstrating its robustness and efficiency under varying disruption scenarios, offering valuable insights into resilience-based restoration.