An improved GRASP for the bike-sharing rebalancing problem

A bike-sharing system is a service in which bicycles are made available for shared use to individuals on a very short term basis. Due to varying user demands in bike-sharing systems, employees need to actively shift bicycles between stations by a fleet of vehicles. In this paper, an improved greedy randomized adaptive search procedure(GRASP) algorithm is utilized to find efficient vehicle tours. Firstly, a greedy construction heuristic is used to construct an initial solution, and then a local search algorithm is used to improve the solution. In addition, some improvements are made in some phases of the algorithm according to the feature of rebalancing problem. Practice examples and comparison with the typical algorithm in the fields are made. The results show that the proposed algorithm is efficient and it can produce good results. The research result has been implemented in Hangzhou.