Dynamic Planning of Bicycle Stations in Dockless Public Bicycle-sharing System Using Gated Graph Neural Network

Benefiting from convenient cycling and flexible parking locations, the Dockless Public Bicycle-sharing (DL-PBS) network becomes increasingly popular in many countries. However, redundant low-utility stations waste public urban space maintenance costs of DL-PBS vendors. In this article, we propose a Bicycle Station Dynamic Planning (BSDP) system to dynamically provide optimal bicycle station layout for network. The BSDP contains four modules: drop-off location clustering, bicycle-station graph modeling, prediction, recommendation. clustering module, candidate are clustered each spatio-temporal subset large-scale trajectory records. modeling weighted digraph model is built based on results inferior with low revenue utility filtered. Then, models across time periods combined create sequence model. prediction GGNN used train data predict next period. recommendation predicted fine-tuned according government management plan, which ensures that recommended conducive city management, vendor revenue, user convenience. Experiments actual networks verify effectiveness, accuracy, feasibility proposed system.