Optimization Model of Dockless Bike-Sharing Delivery in Subway Station Based on GPS Deviation Correction

The dockless bike-sharing (DBS) travel model is favored by people because of the advantages of low-carbon emission, environmental protection, and appropriate flexibility, so it can effectively solve the "last mile" problem. The purpose of this paper is to explore the delivery optimization of DBS around subway stations. The study first matches the map data with DBS GPS data to reduce the error caused by different data sources. After the subway station points are discretized, the effective parking area is determined according to the parking density of DBS around the station. Then, a full day DBS profit model is established at a time interval of 15 min through the supply-demand relationship model. Finally, taking the GPS data of 35 stations of Shanghai Metro Line 11 and one day of Mobike as an example, the newly established model is used to formulate the delivery scheme of bike-sharing around stations and lines.