Optimal Vehicle Scheduling for Bike Re-Balancing

Public Bicycle Sharing System (PBSS) has been introduced into many cities and regions around the world. In this paper, in order to overcome the uneven demand problem in PBSS, a novel multi-objective optimization model aiming at maximizing satisfaction and minimizing cost is established. Moreover, a prediction-based ant colony optimization approach is proposed to solve the imbalance problem. Experimental results show that the modified algorithm can not only guarantee user satisfaction, but also effectively reduce the driving distance of dispatched vehicles.