Dockless shared bicycles accumulation point analysis based on DBSCAN

Abstract. As bike sharing becomes more and more popular, how to optimize the use experiences is now turning into a problem. In many cities, how to allocate the dockless shared bicycles delivery is important for providing convenience for citizens. Based on sufficient data which is originated from a shared bicycle company called Mobike in Shanghai, this paper used Density-Based Spatial Clustering of Applications with Noise (DBSCAN) method to analysis the major accumulation points to help rebalance the allocation. In order to intuitively feel the concentration level, this study divided the points into 3 levels: the most accumulated, the more and the relatively least, which was generally corresponded with the previous study. The most accumulated points were concentrated on Yangpu District, followed by some points in Jing'an and Minhang District. The outer ring of city was relatively less accumulated, but also more accumulated than other areas. The 3 levels significantly differentiated the data, making convenience for staffs to optimize the allocation.