A Novel Genetic Algorithm-based Routing Approach for Electric Vehicles

Recent studies show that transportation accounts for 20 % of the total CO2 emissions of the world, placing it as the main contributor to climate change; therefore, decarbonization of road transport is necessary. Promoting the use of less polluting transport modes such as electric vehicles (EVs) is an important step toward achieving this objective. As it is expected that the use of EVs will rise significantly, this paper aims to help EVs’ drivers to have a better and less stressful driving experience through an innovative routing approach. It consists in leveraging genetic algorithms (GAs) to route an EV effectively based on multiple different factors including route length, its duration and the experienced wait-times. A novel ‘branching’ methodology is developed which takes a random point of an existing route, and attempts to find a unique sub-route to the destination from this point, creating a new additional route for the population to balance vast exploration and exploitation through allowing effective crossover. The preliminary simulation results obtained, using the traffic simulator SUMO, highlight that our proposed approach outperforms A* under congested traffic scenarios.