Abstract

This paper designs a team-based relocation scheme for electric vehicle sharing systems in which stock imbalance can lead to serious service quality degradation. For an efficient operation plan for a relocation team consisting of multiple service staffs, stations are geographically grouped first and the planner merges as many equivalent relocation pairs as possible, considering the number of simultaneously movable vehicles. To obtain a reasonable quality plan within a limited time bound using genetic algorithms even for a large-scale sharing system, each relocation plan is encoded to an integer-valued vector. A vector index points to a overflow station while the vector element points to an underflow station, according to predefined association maps built by the number of surplus and lacking vehicles. The experiment result obtained by a prototype implementation shows that each addition of a service staff cut down the relocation distance by 38.1 %, 17.2 %, and 12.7 %, respectively.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call