Optimal Railway Disruption Bridging Using Heterogeneous Bus Fleets

Bojic, Iva; Luo, Chunling; Li, Xinrong; Zehe, Daniel; Eckhoff, David; Ratti, Carlo · 2021 · DOAJ

DOI: 10.1109/ACCESS.2021.3091576

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of reactive transportation planning during unplanned Mass Rapid Transit (MRT) disruptions. As urban ridership increases, the impact of service interruptions grows, necessitating rapid and comprehensive bridging strategies to minimize passenger delay and maximize the number of served individuals. The authors develop an optimization model that determines the optimal deployment of heterogeneous bus fleets—accounting for different vehicle types and capacities—to bridge disrupted MRT lines. The study is motivated by the need for real-time, computationally efficient solutions that can handle the complexity of mixed vehicle fleets and prioritize passenger groups, aligning with Mobility as a Service (MaaS) concepts. The methodology employs a mathematical optimization model that minimizes total passenger travel delay and the number of unserved passengers. The model utilizes historical smart card data to reconstruct origin-destination demand and existing bus capacities. It discretizes time into slots and considers both existing bus routes and candidate bridging routes parallel to the disrupted MRT line. Key decision variables include route selection, vehicle headways, and the assignment of specific vehicle types to routes. The model incorporates a penalty parameter for unserved passengers, allowing for prioritization based on willingness to wait or pay. To validate the analytical results, the authors extended the microscopic mobility simulator CityMoS, which captures dynamic traffic effects like bus bunching and dwell times that the simplified optimization model assumes away. The study was evaluated using a hypothetical case study of a disruption on Singapore’s MRT purple line during morning peak hours, affecting approximately 6,000 passengers. The optimization model successfully generated bridging plans within minutes, demonstrating its suitability for real-time application. The results showed that the analytical model effectively reduced travel delays and increased served passengers compared to baseline strategies. However, validation via microscopic simulation revealed deviations from the analytical predictions, primarily due to the model’s simplifications regarding static travel times and unlimited bus stop capacities. The simulation indicated that while the optimization model provides a strong initial plan, a combined approach integrating simulation feedback yields more effective bridging strategies. The significance of this work lies in its contribution to reactive disruption management by introducing a model that supports heterogeneous vehicle fleets and passenger prioritization. This capability allows for personalized bridging services, where premium passengers can receive prioritized service. The paper highlights the trade-off between computational efficiency and modeling realism, suggesting that while simplified analytical models are viable for rapid planning, they should be complemented by simulation to account for dynamic traffic interactions. The findings provide a framework for transportation operators to design resilient, cost-effective bridging plans that mitigate the negative impacts of inevitable public transit disruptions.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• passenger motion sickness comfort