Joint Dynamic Bus Dispatching and Bus Bridging Timetabling for Mass Rapid Transit Disruption Management

Abstract

The mass rapid transit (MRT) systems play a pivotal role for urban mobility services but are frequently susceptible to various disruptions. The bus bridging service is a widely-applied substitute transit service in response to the malfunction of MRT systems caused by MRT disruptions. As MRT systems serve the majority demands in urban, a significant number of buses are needed when MRT systems break down. However, in the literature, very little attention has been paid to the source of buses used in bus bridging services, and fleets of buses were impractically assumed to be available as soon as MRT disruptions occur. To bridge this gap, this study explores a feasible source of buses, i.e., nearby bus lines. Specifically, when MRT disruptions occur, the buses operating along nearby bus lines are dispatched to perform bus bridging services. The bus dispatching decisions affect the service quality of bus system and timetables of bus bridging system, which brings a joint optimization problem integrating bus dispatching and bus bridging timetabling. In response to the real-time requirements, the investigated problem is modeled with a rolling horizon approach. For each horizon, the problem is formulated as a mixed integer linear programming model to minimize total penalties caused by affected onboard passengers and cumulative waiting passengers in both bus and bus bridging systems. An effective decomposition method is tailored to find high-quality solutions efficiently. Through decomposition method, the model can be split into three sub-problems, and three efficient methods are proposed to solve these sub-problems based on their unique features. The efficiency of proposed model and method is demonstrated using a case study in Singapore. The computational results show our model and method can meet real-time requirements and guide buses in an efficient way. Finally, the impacts of duration, bus frequency, and passenger demand are analyzed.

Keywords: MRT disruption management, Bus bridging service, Bus dispatching, Bus bridging timetabling, Rolling horizon, Decomposition method