Citation

Nourinejad M, Roorda MJ. Transportation (Amst) 2015; 42(3): 497-518.

Copyright

(Copyright © 2015, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s11116-015-9604-3

PMID

unavailable

Abstract

Carsharing organizations are shared-mobility services which allow users to share vehicles of a fleet. These services are divided into one-way and two-way systems. Two-way systems require users to return the vehicles to their original pick up stations whereas one-ways systems do not. We present a hybrid system that has features of both. Two integer programming optimization models are formulated. The first model, called the fleet size problem (FSP), is for tactical planning and calculates the required fleet size in order to answer all the demand in various market segments where every market segment is composed of some direct-trip and some round-trip users. This model also accounts for vehicle relocation between stations. In reality, however, fleet size is fixed for day-to-day operations. With a specified fleet size, not all users are necessarily served unless the fleet size is quite large compared to the demand. Therefore, a second operational model, called profit maximization problem (PMP), selects from a list of user requests those who profit the service-provider the most. Data from Autoshare, a carsharing company in Toronto, is used to assess the models.

RESULTS of the FSP model indicate that one-way systems require the lowest fleet size but highest vehicle relocation hours. Two-way systems are relatively insensitive to the market segment and the hybrid system's performance is dependent on the market segment scenario. The PMP model shows that the one-way (two-way) system is most beneficial to direct-trip (round-trip) users. The hybrid system is the only system without user surplus costs.

Language: en