A tour recommendation service for electric vehicles based on a hybrid orienteering model

Abstract

This paper designs a tour recommendation scheme for electric vehicles, aiming at reducing time waste induced from long charging time and finally accelerating their penetration into our daily lives. Not just deciding the visiting and charging schedule for the user-selected tourist attractions, our scheme recommends more places having chargers as well as providing tour activities to save the waiting time. Genetic operations are tailored to create a tour plan consisting of essential selected and optional recommended spots by means of combining legacy traveling salesman problem and orienteering problem solvers. Its encoding scheme represents a visiting order, which may have variable number of tour spots, by a fixed-length integer-valued vector, while the fitness function estimates time waste considering the distance between tour places and stay time. The performance measurement result obtained from a prototype implementation discovers that our recommendation service can reduce the time waste by up to 67 % for given parameter setting.