Multi-objective optimal scheduling model for shared bikes based on spatiotemporal big data

Station-free bike sharing is one of the most important short-distance means of transportation. However, with the surge in the number of shared bikes, an imbalance of supply and demand in time and space caused by the disorderly parking of shared bikes has also emerged. This research combined massive spatiotemporal trajectory data of shared bikes and user travel demands to propose a feasible multi-objective optimal scheduling method. Specifically, this research presented a model that utilizes extensive order data to analyze user travel patterns, divides shared bike operation areas into internally connected communities by Geohash coding, and analyzes the shared bike dispatch hotspots in each community segment. Then, a novel model for multi-objective optimal scheduling of shared bikes was proposed based on NSGA-II. The model takes the number of transport vehicles participating in shared bike dispatching and the actual number of dispatch points as the decision variables, and its optimization goal is to reduce the cost of dispatching shared bikes and improve the utilization rate of shared bikes. The optimization effect of the model before and after the improvement of the genetic algorithm was analyzed, and the proposed optimized Pareto front solution set and optimal scheduling routes for shared bikes were given. Some of the results are noteworthy. First, the factors that affect the scheduling cost and utilization rate of shared bikes vary, so setting the variables reasonably is helpful for improving the optimization ability of the scheduling optimization model. Second, exploring the connections between shared bikes can effectively improve the scheduling efficiency of shared bikes. The research results are of great significance for optimizing dispatching routes and formulating low-carbon shared bike dispatching strategies.