Raw data matrix.

Carsharing is regarded as a new mode of transportation that can meet the diversity of travel demands. This study develops a bi-level programming model to design operating strategies aiming to maximize economic and social benefits by determining the site location, number, size and carsharing pricing. In the upper model, a new mixed integer linear programming model is pro-posed to plan the site location, number, size, and pricing of carsharing jointly. Based on the stochastic user equilibrium principle, the lower model is a variational inequality model of travel mode and route choice, to understand the evolution of the transportation system structure after the introduction of shared cars. The results of the case study show the designed scheme can increase the economic benefits of carsharing enterprises and the social benefits of the government, and the site location, number, size, and pricing scheme impact benefits significantly. Furthermore, the setting of shared car sites at public transfer points can promote the interaction between shared cars and other modes of transportation. This study is helpful for carsharing enterprises to increase their economic benefits, enhance their capacity for sustainable growth, and improve the accessibility and flexibility of urban public transportation through sound operational strategies.