Data-driven dynamic accessibility to public service in multi-modal transportation network with uncertainty

Abstract

Conventional accessibility of public service is concerned about the coverage of alternative public service resources, which is a static indicator without consideration of individual demands and travel time uncertainties. A data-driven dynamic accessibility method is proposed in this paper to investigate the effects of travel time uncertainty on accessibility to public service with consideration of multi-modal transportation network topology and on-time arrival probability. Travel behaviors of travelers taking taxi and metro for access to the public hospital in Guangzhou, China are investigated for evaluation of the place-based accessibility. With the use of Floating Car Data of taxi and Smart Card Record from Metro in Guangzhou, the variation of travel times by taxi and metro for access to the Guangzhou general hospital are analyzed. Empirical results reveal the dynamic impacts of travel time uncertainty on the place-based accessibility coverage area by mode, on-time arrival probability and time of day.