Genetic Algorithm-Based Relocation Scheme in Electric Vehicle Sharing Systems

Abstract

This paper designs a genetic algorithm-based relocation scheme for electric vehicle sharing systems, which suffer from the stock imbalance problem due to different rent-out and return patterns in different stations. To improve the service ratio, the relocation scheme explicitly moves vehicles from overflow stations to underflow stations. Each relocation plan is encoded to an integer-valued vector, based on two indexes, one for the overflow list, and the other for the underflow list. In each list, stations are bound to specific locations according to the number of surplus or needed vehicles. For a vector element, its location is the overflow station index, while the value is the underflow index. Iterative genetic operations improve the population quality, computed by the relocation distance, generation by generation. The simulation result shows that the proposed relocation scheme finds an efficient relocation plan in the early stage of iterations for the given parameter set.

This research was financially supported by the Ministry of Knowledge Economy (MKE), Korea Institute for Advancement of Technology (KIAT) through the Inter-ER Cooperation Projects.