Dynamic Vehicle Allocation and Charging Policies for Autonomous Electric Vehicle Sharing Systems

In the future, vehicle sharing systems will be unmanned, autonomous, and electric. In this paper, we design operating policies for sharing platforms with such vehicles. These platforms must decide which vehicle with which battery level should pick up which type of customer and which vehicle should be charged partially or fully. To evaluate vehicle operating policies, we first model the system as a semi-open queueing network (SOQN) with multiple synchronization stations for matching customer battery-demand classes with semi-shared vehicles holding ample remaining battery capacity. If a vehicle’s battery level drops below a threshold, it is routed probabilistically to a nearby charging station for partial or full charging. We solve the analytical model of the SOQN and obtain approximate system performance, which appears to be accurate based on validation by simulation. Then the SOQN is used in a Markov decision process to minimize total cost, and to find good heuristic policies. The experiments show that a state-dependent policy is near optimal in a small-scale network and outperforms benchmark policies in large-scale realistic scenarios. An interesting finding of our research is that reserving idle vehicles to wait for future short-distance customers even when long-distance customers are waiting for vehicles can be beneficial