A Sharing Bike Rebalancing Scheduling Method For Mixed Fleets

Since its inception,the Bike Sharing System(BSS)has gradually become one of the most green and healthy transportation modes through the operating mode of time-sharing leasing.It not only solves the "last mile" travel problem for users,but also effectively alleviates the huge carrying pressure of urban public transportation.The operation optimization of shared bicycle systems has also become one of the research hotspots in the transportation field.Its research mainly focuses on how to solve the problem of supply and demand mismatch between the bicycle supply and user demand of shared bicycle systems caused by traffic asymmetry,which is essentially a problem of rebalancing scheduling.In traditional research,this type of problem is defined as the Vehicle Routing Problem(VRP),which is a generalization of the Traveling Salesman Problem(TSP),where a set of scheduling requirements and a team of scheduling vehicles are given,with the goal of meeting all scheduling requirements with the lowest possible scheduling cost.The specific order in which each scheduling vehicle processes each scheduling requirement can be determined.This article considers the actual situation in the operation of shared bicycle systems,and analyzes and studies the supply-demand rebalancing problem,location selection problem,and rebalancing solution problem of shared bicycle systems based on common pileless shared bicycles.The specific research work in this paper is as follows:(1)A rebalancing scheduling method for shared bicycle systems in hybrid fleets.By analyzing and comparing the stations of the shared bicycle system,scheduling different stations with different types of vehicles can help reduce the scheduling cost and pollution emissions of the shared bicycle system.Firstly,by analyzing the requirements,capacity,and other indicators of different stations,all stations in a system are divided into large and small stations.Then,the two stations are separately scheduled by vehicles with different power modes,carrying capacity,and operating costs,and solved in virtual network and actual network examples.The empirical results indicate that this type of scheduling mode is superior to traditional homogeneous vehicle scheduling in terms of cost reduction and pollution emissions.In addition,the scheduling model proposed in this article also provides new references for operators of shared bicycle companies.(2)Site selection considering geographical characteristics.Using Analytic Hierarchy Process(AHP)and combining relevant geographical features,analyze the importance of different geographical factors to users,assign weights to each region type,and add feature vectors of geographical weights to each sample data based on the region type.Then,the improved hierarchical clustering algorithm is used to generate the initial cluster center.Finally,the improved KMeans clustering algorithm is used to find the optimal number of clusters,and 11 single vehicle station networks with geographical characteristics are generated.In previous studies,people often focused on cost and user needs when it comes to site selection.This article adds the influence of geographical features on site selection,providing a solution for practical operations of enterprises.(3)Improved heuristic algorithm design.Elaborate on the relevant principles,basic models,algorithm steps,and related parameters of the basic ant colony algorithm.Based on relevant research,use the site data generated by the site selection method in Chapter 4 to customize an improved ant colony system suitable for the research topic of this article to solve the model in Chapter 3.Compare and analyze the solution performance of the improved algorithm selected in this article with the basic ant colony algorithm,The results indicate that the improved ant colony system proposed in this paper has better performance in finding optimal solutions compared to the basic ant colony algorithm.This article conducts in-depth analysis on the shared bicycle system from the perspectives of rebalancing scheduling,location selection,and algorithm solving,providing relevant references for algorithm solving solutions for enterprise business activities and optimization problems.