Dockless shared bikes are playing an increasingly important role in slow traffic, and gradually become one of the main mode of “last kilometer”. However, shared bikes often form unreasonable aggregation in operation for the spontaneous aggregation effect, which indicates that reassignment should be applied. Therefore, we constructed a shared-bike traffic network, conducted a quantitative evaluation of the network characteristics, and provided reasonable bases for rebalancing of shared bikes to reduce unreasonable aggregation. In this paper, by using a novel cluster method based on spatial relation, a shared-bike traffic network of about 8000 nodes (clustering centers) was constructed and the it’s topological and statistical characteristics were calculated. The result indicated that the growth of the shared-bike traffic network in Shanghai has reached a stable state and the network presents a significant scale-free feature. This study provided a scientific basis for the reassignment of shared bikes by bike suppliers, thus promoting the optimization and development of the shared-bike traffic network.

CICTP