Research Output

Schedule-based hyperpath approaches to transit assignment: the impact of imperfect information.

  In a schedule-based capacitated transit network we analyze optimal passenger strategies, where a strategy is a ranked set of services at each interchange in the transit network. The passenger is assumed to take the first ranked service unless this is full in which case he chooses the second ranked service, etc. In this equilibrium model passengers choose strategies to minimize their expected cost of travel. This cost depends not only on waiting times and travel times but also on failure-to-board probabilities. These probabilities depend on the loading situation of the services. The more passengers try to use an overcrowded service, the less likely it is that they are able to board the vehicle. The model is asymmetric in the sense that passengers that are already on board have priority. This model, originally proposed by Hamdouch et al. (Hamdouch and Lawphongpanich 2008), assumed that passengers are aware of the complete network, the complete schedule and the loading situation, i.e. the failure-to board probabilities. We analyze how imperfect information affects optimal passenger strategies and the overall loading situation in several dimensions. We also analyze how passengers who don't take all the strategies into account fare in comparison to those who do. Once a passenger has missed his preferred connection, he might not be aware of or able to determine the connection that is now best for him. Hence we analyze how optimal strategies alter when passengers consider average headways instead, thereby effectively combining schedule- and frequency-based assignment.

  • Type:

    Conference Paper (unpublished)

  • Publication Status:

    Unpublished

  • Library of Congress:

    HE Transportation and Communications

  • Dewey Decimal Classification:

    388 Transportation; ground transportation

Citation

Rochau, N., Bell, M. G. H., Noeckel, K. & Fonzone, A. Schedule-based hyperpath approaches to transit assignment: the impact of imperfect information.. Paper presented at 5th IMA Conference on Mathematics in Transport, London. (Unpublished)

Authors

Keywords

Training assignment; capacity constraints; passenger strategies;

Available Documents