An intelligent method for dynamic distribution of electric taxi batteries between charging and swapping stations

Abstract Taxis are often motivated to drive in crowded areas to earn more money by serving customers. Clean taxicab fleets in cities will have a significant impact on reducing air pollution and cutting emissions. Use of electric taxis is a highly efficient solution to address the issue of greenhouse effects, because electric cars are cleaner and cheaper than gasoline-powered cars. Battery swapping is an efficient and fast recharging method enabling taxi drivers to go to a battery swapping station (BSS) and replace their empty batteries with full ones. We study a new dynamic optimization model that helps the planning process in joint consideration of inventory, routing, and dynamic pricing to distribute electric taxi batteries between a BSS and a battery charging station (BCS). A novel dynamic programming (DP) model is proposed to incorporate a Markov decision process (MDP) with an actual demand function, operator cost, customer delay, and a dynamic pricing strategy using a social optimization function through a look-ahead policy. The results indicate that the average response time to transfer batteries from BCS nodes to BSS nodes under the look-ahead policy is reduced by up to 9% compared to the myopic case. We also observe that the average social welfare under the look-ahead policy increases by 22 % compared to a policy without look-ahead. The numerical analyses highlighted the benefits of supplying batteries at a socially efficient price instead of the standard monopoly price.