Station planning by simulating user behavior for electric car-sharing systems

Abstract

The planing of a full battery electric car sharing system involves several strategic decisions. These decisions include the placement of recharging stations, the number of recharging slots per station, and the total number of cars. The evaluation of such decisions clearly depends on the demand that is to be expected within the operational area as well as the user behavior. In this work we model this as combinatorial optimization problem and solve it heuristically using a variable neighborhood search approach. For the solution evaluation we use a probability model for the user behavior and approximate the expected profit with a Monte-Carlo method. The proposed algorithm is evaluated on a set of benchmark instances based on real world data of Vienna, Austria. Computational results show that by simulating user behavior the expected profit can increase significantly and that other methods assuming the best case for user behavior are likely to overestimate the profit.