Enhancing the evacuation efficiency through the two-step optimization of train timetable and response vehicles during metro disruptions

Train timetable rescheduling and bus bridging service are two common measures in metro disruption management. Existing literature on train timetable rescheduling typically does not address the issue of transporting stranded passengers, and the use of a fixed operational mode in bus bridging services can lead to a loss in transport capacity. In this paper, we present a two-step optimization model that involves train timetable rescheduling and response vehicle scheduling. In the first step, the train timetable rescheduling (TTR) model aims to minimize the interests of both metro operators and passengers. Stranded passengers are distributed based on the platform holding capacity at each station, and three rescheduling strategies (cancellation, short-turning, and adjustment of the train departure time) are also adopted. In the second step, the response vehicle scheduling (RVS) model aims to provide flexible auxiliary transportation service for stranded passengers, with the goal of transporting them to their destinations as quickly as possible. Both models are coordinated to ensure the effective transport of passengers during the disruption period. Furthermore, we develop a decomposition-based heuristic framework that performs well for solving our models. Finally, we conduct several numerical experiments to test the validity of the models and algorithms. The results show that: (1) the combined rescheduling strategies can decrease the passenger waiting time and alleviate the platform pressure at each station; (2) compared with the standard bus bridging service, the response vehicle performs better in reducing both passenger total travel time and vehicle empty driving.