Research On Operation Law And Advance Scheduling Method Of Public Bicycle System

As urban environmental pollution and traffic congestion become increasingly prominent,guided by the green and sharing concepts,and supported by the rapid development of Internet and GPS positioning technologies,guided by the concept of green and sharing,supported by the rapid development of technologies such as the Internet and GPS positioning,public bicycles have emerged and developed rapidly as representatives of low-carbon,environmentally friendly,and healthy travel modes.At present,there is a problem of vehicle deployment lag in the operation process of the public bicycle system,which seriously affects the system operation efficiency and user experience.How to predict the demand of outlets for vehicles,accurately assess the importance of each outlet in the system,give priority to the supervision and dispatch of bicycles in outlets with higher importance is an effective measure to solve the problem of vehicle deployment lag.This article takes the Shenzhen public bicycle system as an example.Firstly,based on the user travel data,analyze the space-time rules of the system operation,discuss the impact of rainfall,temperature,air quality and other factors on the system's vehicle volume through historical weather data.Secondly,based on BP neural network model,a method for predicting the demand for outlets is combined with factors such as the time of use,weather,temperature,etc.Finally,a method of evaluating the importance of directed weighted network nodes using a tree-like data structure is proposed to provide theoretical support for solving the problem of scheduling lag by improving the scheduling efficiency by balancing important nodes in the system first.The main work and conclusions of this study are as follows:1.Based on the user's card swipe data of the public bicycle system,descriptive statistical analysis is performed on the system's vehicle usage rules at different times of the day,the user's vehicle length rule,the bicycle usage rate of different types of outlets,and the flow of vehicles between outlets.Combine historical weather data to analyze the impact of natural factors such as rainfall,temperature,and pollution on the number of users borrowing cars.2.Combined with the factors that influence the user's car usage,the method for predicting the demand of public bicycle outlets based on the BP neural network model is proposed,which takes into account the historical characteristics,time period characteristics,date characteristics,weather characteristics,and temperature characteristics of user vehicle usage.The validity of the method is proved by comparing the difference between the predicted amount and the actual value of each part of the outlets obtained by using this method in different days and weather characteristics.3.This paper proposes a construction tree-based node importance evaluation method for directed weighted networks.This method separately evaluates the breadth-first traversal tree and depth-first traversal tree of the evaluated nodes to investigate the local influence and global influence of the evaluated nodes.In the evaluation process,consider the influence of the in-degree and out-degree of the nodes to be evaluated on the importance of the outlets.Therefore,the evaluation method can provide more comprehensive information on the importance of different nodes in the network.The robustness experiment based on ARPA network proves the effectiveness of the method.Applying the above method to the importance evaluation of the outlets of the public bicycle system can achieve significant results.