建成环境对共享单车轨道接驳的非线性作用研究

Research on the Nonlinear Effect of the Urban Built Environment on Shared Bicycles Connecting to Rail Transit

摘要: 为探究建成环境对共享单车与轨道交通接驳的复杂影响，本研究以上海市为案例，提出结合聚类分析与分位数回归的分析框架。研究首先利用多源数据和K-means算法，将轨道站点依建成环境特征划分为“城市中心型”与“城市边缘型”。通过构建分位数回归模型，深入剖析了建成环境要素对接驳量与骑行距离的非线性影响及阈值效应。研究核心发现：建成环境的影响具有显著的非线性特征，且因站点类型而异。对城市边缘型站点，人口、就业及主干路网密度是关键的积极因素；而城市中心型站点则更依赖于土地利用混合度的提升。此外，模型揭示了部分变量在不同分位点上影响效应会发生变化，体现了传统线性模型无法捕捉的阈值关系。本研究成果可为优化共享单车设施布局及制定精细化城市规划策略提供科学依据。

Abstract: To investigate the complex effects of the built environment on the connection between shared bicycles and rail transit, this study uses Shanghai as a case study to propose an analytical framework that combines cluster analysis and quantile regression. The research first utilizes multi-source data and the K-means algorithm to classify rail transit stations into “urban center” and “urban fringe” types based on their built environment characteristics. By constructing a quantile regression model, the study provides an in-depth analysis of the non-linear impacts and threshold effects of built environment factors on the volume and riding distance of these connection trips. The core findings reveal that the impact of the built environment is significantly non-linear and varies by station type. For urban fringe stations, population density, employment density, and main road network density are key positive factors. In contrast, the effect of land use mix is more prominent for urban center stations. Furthermore, the model reveals that the influence of certain variables changes across different quantiles, demonstrating threshold effects that traditional linear models cannot capture. The results of this study can provide a scientific basis for optimizing the layout of shared bicycle facilities and for formulating refined urban planning strategies.