Research On Shared Bicycle Vehicle Scheduling Based On Travel Data

With the development of the sharing economy,Bike-sharing has become a vital tool for the "last mile" of the destiny in a city because of its flexibility,low carbon and environmental friendly,and good accessibility.However,Bike-sharing also has the problem of "no car to borrow" and "no room to place".Under the condition of controlling the total amount of Bike-sharing,how to properly deploy and improve the urban bike-sharing system has become an urgent problem for the current bike-sharing model.Based on the theoretical research of public bicycles,this paper studies the demand forecasting and optimization of scheduling path of the bike-sharing.Firstly,the paper defines the bike-sharing and clarifies the position of the bike-sharing.On this basis,through the travel data of Beijing Mobike bicycle in April 2018,this paper analyzes the borrow and return features of the bike-sharing.The placement points are divided into several categories to explore their change rules of demand.According to the Mobike bicycle travel data characteristics,BP neural network is selected to predict the travel volume of different placement points.According to the changing rules of demand of different placement points,the current situation and existing problems of vehicles,scheduling in bike-sharing systems are analyzed.Then the vehicle scheduling optimization method is discussed,and the solution idea of bike-sharing scheduling is determined.According to the predicted scheduling demand,an optimal scheduling model based on multi-vehicle time window requirements with minimum scheduling vehicle and minimum scheduling cost is constructed.According to the characteristics of the model,the tabu search algorithm is designed to solve the model,taking the Mobike bicycle system in Haidian District of Beijing as an example for analysis.The algorithms and models proposed in this paper help to improve the scheduling efficiency of bike-sharing.