Reposition optimization in free-floating bike-sharing system: A case study in Shenzhen City

In recent years, the free-floating bike-sharing system has attracted a large number of travelers for its convenience and non-pollution in the urban public transit. Although many researchers have been carried out, there still exist challenges in repositioning optimization to improve the operating efficiency in the bike-sharing system. This study introduces an improved repositioning model for the free-floating bike-sharing system and designs Non-dominated Sorting Genetic Algorithm II (NSGA-II) to conduct optimization. Firstly, the Gaussian mixture model (GMM) is adopted for clustering to obtain the stations in the free-floating bike-sharing system. Then, a multi-objective optimization model is established considering minimizing the total cost of the repositioning operation and maximizing the satisfaction degree of demands of the delivery stations. To solve the model, an improved NSGA-II is proposed, in which different ways of crossover and mutation operators are applied to maintain the diversity of the population. The Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) is used to determine the best compromise solution. Finally, using trajectory data from the freefloating bike-sharing system in Shenzhen City, we further compare the results with traditional Genetic Algorithm (GA). The comparison demonstrates that the proposed method is reasonable and effective in the reposition optimization for free-floating bike-sharing system. (c) 2022 Elsevier B.V. All rights reserved.

Automated summary

This study introduces an improved repositioning model for the free-floating bike-sharing system and designs Non-dominated Sorting Genetic Algorithm II (NSGA-II) to conduct optimization. The proposed method, which adopts Gaussian mixture model (GMM) for clustering and establishes a multi-objective optimization model, is reasonable and effective in reposition optimization for free-floating bike-sharing system.