Fleet Sizing for Electric Car Sharing Systems in Discrete Event System Frameworks

This paper proposes a two-level strategy to determine the optimal fleet size of electric car (EC) sharing systems (ECSSs) in a networks of a set of stations. At the first level, the system is modeled as a discrete event system in a closed queueing network framework that allows describing the asymptotic system behavior and determining the optimal fleet size that maximizes the network profit. At the second level the ECSS dynamics is modeled by timed Petri net, in order to take into account some particular aspects, such as the user flows in different time periods of the day or the exit of the customers from the stations when they do not find available ECs. A simulation campaign analysis enlightens the effectiveness of the presented complementary design strategy.