Shared travel demand forecasting and multi-phase vehicle relocation optimization for electric carsharing systems

Electric shared mobility is flourishing in urban transportation. However, the problem of uneven vehicle distribution and untimely vehicle charging hampers user trip experience and system operation efficiency. To overcome these challenges, this study proposed a multi-phase vehicle relocation optimization approach for one-way station-based carsharing systems. In phase one, a micro-level shared travel demand forecasting model was developed to capture the number of orders in the short-term future. In phase two, stations were divided into different categories based on the results of user travel demand forecast. In phase three, the minimization of driving mileage and carbon emissions was taken as the optimization objective, and a solution method combining Gurobi solver and charging priority ranking was designed. Finally, the effectiveness and advantages of the proposed model and algorithm were comprehensively validated through a case study using real passenger orders and geographic data from the city of Shanghai, China.