Genetic Algorithm for Scheduling Charging Times of Electric Vehicles Subject to Time Dependent Power Availability

The expected widespread adoption of electric vehicles (EVs) in the near future brings new challenges as for example that of scheduling the charging times under limited power and other technological constraints. In this paper, we tackle a scheduling problem derived from a EV charging model and control system recently proposed to organize the charging times of a large fleet of EVs. This problem may be formalized as that of scheduling a set of tasks with given processing times and due dates on a machine whose capacity varies over time. We first introduce a schedule builder and study its main properties. Then, we propose a genetic algorithm that exploits the schedule builder as decoding algorithm. Experimental results show the suitability of our approach.