Often serving as the backbone for public transport in metropolitan areas, any major disruption in the metro system will have a severe impact, affecting tens of thousands of passengers. Even for the most reliable metro systems, such as the Mass Transit Railway (MTR) in Hong Kong,on average there are hundreds of service disruptions every year, with some lasting for a few hours. It is, therefore, imperative to develop contingency plans for disruption management. This is particularly important for rail systems running on single tracks, without parallel lines, wherein any rail blockage would require substitute bus (SB) for bridging the disrupted railway sections. Despite the importance of bus bridging for rail disruption management, in reviewing the literature, it is surprising that this pro...[ Read more ]

Often serving as the backbone for public transport in metropolitan areas, any major disruption in the metro system will have a severe impact, affecting tens of thousands of passengers. Even for the most reliable metro systems, such as the Mass Transit Railway (MTR) in Hong Kong,on average there are hundreds of service disruptions every year, with some lasting for a few hours. It is, therefore, imperative to develop contingency plans for disruption management. This is particularly important for rail systems running on single tracks, without parallel lines, wherein any rail blockage would require substitute bus (SB) for bridging the disrupted railway sections. Despite the importance of bus bridging for rail disruption management, in reviewing the literature, it is surprising that this problem has not attracted the level of efforts it deserves.

This thesis will address four important issues in the development of SB contingency plans. The first issue regards the initiation time for SB service. As the duration of a metro disruption varies and cannot be predicted perfectly ahead of time, the response timing is critical. Initiating SB service too early would confuse passengers and have cost implications; initiating too late, on the other hand, would exacerbate the problem as the unsatisfied demand accumulates. The second one pertains to the routings of SB service and the fleet sizes upon their initiation. The analysis should consider demand redistribution due to the disruption, which introduces uncertainty to the problem. Two types of SB services to cater for the affected demand are proposed: regular SB service and flexible SB service. The third concern regards the negotiation between the metro company and the bus company to put together of a fleet to serve as SB. The analysis will focus on developing cost-effective portfolios to supply the needed SB fleet. The last issue concerns about the operation of the SB, as bus bunching is one of the most serious problems in deteriorating bus operations, a two-way-looking self-equalizing method is introduced to control the SB, therefore better regulating the bus headways.

Two advancements will be introduced in this thesis for the development of SB contingency plans: the first is the proper treatment of uncertainty, including the uncertainty in disruption duration and that of the demand pattern in the aftermath of a disruption, as well as the bus running time during operations; the other is capturing the time-dimension explicitly in the analysis as the disruption unveils itself over time. Both advancements are critical for developing robust contingency plans that adapt to the situation as the disruption continues. We will use Hong Kong and Shanghai case studies to demonstrate the modeling framework. This research will open up theoretically interesting and practically important topics to enhance the SB service for metro system disruption anagement.