An Adaptive Tabu Search Optimisation Algorithm for Solving E-Scooters Battery Swapping Problem

Abstract

E-scooters have become a popular mode of transportation for first and last-mile excursions in recent years. Their usage as a short-distance public transit system is commonly regarded as an efficient solution to minimise carbon emissions while also being handy for individuals on the go. The need for charging, however, has become problematic as a result of the popularity of e-scooters because it effectively renders the scooter that is presently being charged inoperable. A relatively new method being used involves swapping batteries in e-scooters rather than transferring full scooters to be recharged, reducing out-of-service time to a few minutes rather than hours. To reduce trip distances and maximise fuel economy for battery swapping operators, a system for determining the most effective path to switch the e-scooters’ batteries will be required. This paper aims to do this through the use of Tabu Search (TS) algorithm to determine the optimal number of battery swapping operators for an area and then to ascertain the most efficient routes for each operator. This method will then be compared to Simulated Annealing in order to determine which method is the most optimal for this scenario. The data used to evaluate this method was obtained from the 2019 Chicago pilot program . The results showed an adapted tabu search in the total distance travelled, leading to shorter charging trips comparing to simulated annealing.