Dispatching strategy of shared bicycles for morning peak tidal travel based on spatial distribution

Bicycle sharing has rapidly gained popularity as a low-carbon mode of transportation, with the advantages of low cost and sustainability. However, its tidal phenomenon during the morning peak period leads to a mismatch between demand and regional distribution, which brings challenges to urban traffic management. In this paper, we address this problem and propose a bike-sharing scheduling strategy for morning peak tidal travel. The study identifies the spatial distribution characteristics of the tidal phenomenon by analyzing the riding order and electronic fence data and combining them with the KD-Tree algorithm; subsequently, a demand prediction model based on the XGBoost algorithm and a hierarchical scheduling model based on the greedy algorithm are constructed to dynamically optimize the spatial distribution of the bicycle resources with the demand fluctuation as the guide. Empirical studies based on Xiamen Island shared bicycle data show that the scheduling model can effectively alleviate the phenomena of excessive resource concentration and idleness, and significantly improve the resource utilization rate and operational efficiency of the bicycle system.