01627827

Component

Service disruptions are among the greatest enemies of attractiveness and trust in public transport. In many cities in the world, plenty of effort has been put for policies, technologies, information services and travel options, in order to raise attractiveness of rail, metro and bus services for travelers preferring private vehicles, but it often takes only a few really bad experiences to fall back to less sustainable options. While preventing all disruptions may be out of reach due to their nature (e.g. power failures, inclement weather, medical emergencies, intrusions), one can focus on their consequences. Yet, such assessment needs to cover the entire transportation system as it may significantly affect the performance of other modes as well, especially in urban areas. With a comprehensive and well- calibrated simulation model, one could assess impacts of potential disruption and test mitigation strategies. This is, however, a type of scenario where careful representation of just-in-time agent- interaction dynamics and limited access to information needs to be taken into consideration. In other words, simulation of travel disruptions requires a representation of within-day behavior with incomplete information and en-route routing, which are known to be a challenge in travel models. This paper demonstrates the use of a multi-modal simulation tool, the SimMobility Mid-term model (SimMobilityMT), in the study of a disruption scenario occurred previously in Singapore’s mass-rapid transit (MRT) system. In order to simulate within-day behavior and agent interactions, SimMobilityMT uses a publish/subscribe mechanism that is able to asynchronously propagate information across agents and trigger en-route rerouting decision making processes, as agents become aware of the disruption (e.g. when they arrive to the MRT station). For empirical analysis purposes, the authors test a bus-bridging disruption management strategy, using shuttles between pairs of MRT stations. The authors compared the results with a baseline, where no disruption is occurring, according to several metrics, namely travel times, dwelling time, waiting times, denied boarding and mode share. Unsurprisingly, the bus bridging strategy substantially outperforms a scenario where there is no pro-active support from the operators.

This paper was sponsored by TRB committee AP065 Standing Committee on Rail Transit Systems.

01618707

17-06587

English

17p

2017

Transportation Research Board 96th Annual Meeting

Digital/other

Figures; Maps; References; Tables

Intelligent agents; Rapid transit; Service disruption; Shuttle buses; Simulation; Urban areas

Singapore Mass Rapid Transit (MRT)

Singapore

Operations and Traffic Management; Planning and Forecasting; Public Transportation

Transportation Research Board Annual Meeting 2017 Paper #17-06587

TRIS, TRB, ATRI

Dec 8 2016 12:41PM