基于交通生成的共享单车投放量估计方法研究OA

To address the issue of uneven distribution of shared bicycles in operations,the article proposes a method for estimating the deployment volume of shared bicycles.It screens and classifies shared bicycle data,summarizes the generation and attraction volumes at different parking stations,and evaluates the supply-demand balance of shared bicycles.Using gray relational analysis and correlation coefficients,it identifies the indicators affecting travel rates,selecting the number of residential units,land area,and building area as input variables to construct a BP neural network travel volume prediction model.By combining residential building travel rates and the travel share rate of shared bicycles,the deployment volume of bicycles is estimated.This method was applied in an empirical study in Hefei,and the results show that the estimation error of deployment volume is 6.48%,with prediction performance better than models such as CNN and PSO-SVM.This method can effectively generate shared bicycle deployment volumes,providing a reference for shared bicycle operation management.