Trajectory-interception based method for electric vehicle taxi charging station problem with real taxi data

Abstract

Taxi fleets are suitable for the first stage of deployment of electric vehicles (EV) because EV taxis can provide chances for more people to experience EV. However, the lack of charging facilities and limited operation range are barriers to the adoption of EV taxis. As EV taxis can charge when there is no passenger in the vehicle, the conventional origin-destination (OD)-based flow-interception method for the charging station problem is not appropriate. In this study, we present a new EV taxi charging facility planning model that uses the trajectory-interception method with real taxi trajectory data and electric vehicle operation data. Global positioning system (GPS) data from 1,000 vehicles (taxis) and taxi battery performance data from three EVs are used. We developed an optimization algorithm to find optimal charger distribution considering charger installation cost and waiting delay simultaneously based on the evolution algorithm. The results show that the maintenance cost of the chargers and the charging stations is more sensitive than the opportunity cost of taxis, including additional travel distance and time to minimize the total cost. The methodology proposed in this study provides a robust result for the charging station location problem and can be widely applied to other cities with similar situations.