Train rescheduling for minimizing passenger travel time under disruption for metro lines

Abstract

In case of a metro disruption, the collaborative optimization for train rescheduling combined with passenger flow demand is quite essential since passenger flow is not adequately considered. To handle this gap, this study first establishes two separated optimization models, i.e. train rescheduling model and passenger flow model. On train side, short-turnings, fully cancellation or partial cancellation are taken into account to construct the disrupted timetable with the aims of minimizing train delays and timetable deviations from the original timetable. On passenger side, time-dependent OD passenger demand is incorporated into passenger’s arrival, passenger’s loading and alighting process. Then, an iterative framework is proposed to interact these two models. In each iteration loop, random short-turnings and cancellations are applied to calculate new departure and arrival for the services. Next, the corresponding departure time and arrival time is transferred to passenger flow model to evaluate the impacts on passenger’s travel time and the numbers of stranded passengers. The iterative results illustrates that less short-turning can effectively mitigate the numbers of stranded passengers at platforms but it comes at the cost of increasing train delays and passenger travel time. Cancelling trains can help to reduce passenger travel time as well but it should avoid causing heavier train delays to other non-cancelled services. Therefore, reasonable numbers of short-turnings and cancellations during disruption is trade-off between train delays and passenger delays during disruptions.