In this paper we propose an improved mathematical model for locating EV charging stations. We consider the successive activities of the travelers, i.e., a person with two main stops during the day should charge the vehicle in just one of the parking spots, hence avoiding double counting the demand. The model is tested for the city of Coimbra (Portugal), where there is a network of nine stations. We conclude that our solution is better than the one that was implemented in reality, moreover we are able to conclude that demand transference has a rather significant impact on the solutions.

In this chapter we propose a method to plan the location of charging stations for electric vehicles (EV) in a city in which the objective is to maximize the number of satisfied vehicles under a fixed budget for building the stations. We take into consideration the maximum capacity of each possible site for installing a station, in terms of the number of plugs that each one can have, and the distance from that location and each demand point, which is measured in walking time. To be able to apply these models, we develop a charging demand model for based on parking data, considering that the higher the parking time, the greater the probability of charging. We also take in consideration the relation between the demand at different points, e.g., if a vehicle can charge at home, the probability of needing to charge at work will be significantly reduced. We test our mathematical models for the case of the city of Coimbra, where there is already a network of charging stations. We first use an existing mobility survey to extract parking data and establish a demand grid, and then we apply the models that gives us the optimal location for charging stations for the entire city allowing us to compare both.