An Agent-based Simulation for Shared Automated Electric Vehicles with Vehicle Relocation

Abstract

This paper presents a new vehicle relocation strategy for shared automated electric vehicles (SAEVs). An agent-based modeling methodology is employed. We consider two types of agents: SAEV agents and customer agents. SAEVto-customer assignment rules and SAEV relocation rules are both defined. The proposed approach allows for charging of SAEVs while en route to rebalancing stations. Treating charging stations as waypoints constitutes a significant departure from present techniques. Simulation results confirm the improvements in performance of an SAEV system with en route relocation when compared to no relocation, in terms of reductions in average waiting times and lost customers.