In recent years, Bike sharing systems have been installed in many cities around the world and are increasing in popularity. However, the parking of bike sharing everywhere also brings many problems to the city. In order to cope with the challenges brought by shared bicycle parking to urban management, we use complex network analysis methods, build a shared bicycle OD travel network based on two weeks of shared bicycle data in Beijing,2017, and analyze the relevant indicators of the network. Further,we select different functional zones to measure the number and proportion of traid in the network within this local area.The results show that the nodes with large shared bicycle network are concentrated in the vicinity of the subway bus station, which indicates that the circulation of shared bicycles in the vicinity of public transportation stations is large and the riding is frequent. A small-scale network is selected to calculate the average path length between 2-3, and the transmission performance and efficiency of the network are high. The transitive clustering coefficient and the cyclic clustering coefficient respectively can obtain the transitive clustering coefficient larger than the cyclic clustering coefficient, which further proves that the mutual communication between the nodes in the bike sharing OD network is insufficient. The circulation is not strong, and the flow of the vehicle is mainly based on the transmission mode, with obvious directionality. By establishing a local network for three different functional areas of residential area, commercial area and school district, the triad isomorphism is measured, and the bike sharing flow has obvious directionality. More nodes are agglomeration points and dispersion points, and the transfer model is The proportion of vehicles is large, and the flow of vehicles between nodes is not balanced. Only a small number of nodes have two-way flow between vehicles, so scheduling is required.

CICTP