Modeling routing behavior for vacant taxicabs in Urban traffic networks

Taxicabs account for a significant portion of traffic in many Asian cities, and route choice by taxi drivers is an active research area. The routing objectives of a taxi driver vary, depending on taxi occupancy. If a taxi is occupied by customers, then a least-cost path is usually sought. Several paradigms in the literature are related to such a routing objective. However, the taxi driver's route choice behavior when a taxi is vacant is not well understood. A routing model for vacant taxis is proposed in which taxi drivers are assumed to minimize the expected search time for customers when making routing decisions at intersections. A probabilistic dynamic programming formulation of the problem and the solution algorithm are presented. A numerical analysis was conducted on a hypothetical network resembling the traffic network structure in the city of Taipei, Taiwan. The proposed model exhibited realistic and reasonable properties. This research sheds light on the routing decisions of taxi drivers and therefore directly supports areawide traffic management.