Research papers - Modelling the dynamic one-way car sharing system

The one-way car sharing scheme allows users to return a rented car to any designated spot, which may be different from the origin station. The existing studies mainly focus on the vehicle relocation problem in order to deal with the travel demand fluctuation over time and demand imbalance in space. However, the strategic plan for the station deployment and its capacity design for the one-way car sharing have not been well studied, especially when taking the vehicle relocation into consideration. To solve the car-sharing station location and capacity problem, this paper developed a Mixed-integer Non-linear Programming (MINLP) model. Firstly, the relocation operation and relocation cost are taken into consideration to address the imbalance between demand and availability of vehicles in different time steps, which is a key part in car sharing dispatch. Secondly, the flexible travel demand based on the different time steps are taken as the inputs to the dynamic optimization model. Thirdly, the elastic service rate is constructed using the ratio of car sharing utility and private car utility. An illustrative example is constructed to demonstrate the effectiveness of the model and to draw insights on policy implications to promote car sharing.