摘要
【背景】随着我国城市轨道交通线网的快速扩张,轨道交通与常规公交在共同服务区内的客流竞争日益加剧,传统研究往往忽视环境因素的空间异质性影响,难以解释二者客流竞争时空分布格局的复杂作用机理。【目标】将表征社会经济指标、地铁公交线网属性及城市建成环境的外生因素纳入指标体系,通过引入多尺度地理加权回归(MGWR)模型,克服传统回归方法参数全局恒定和单一尺度的局限。【方法】以成都市中心城区“5+1”区域为例,分别统计工作日与非工作日期间地铁公交客流分担比,以揭示二者间客流竞争空间格局。通过整合多源时空数据,经变量多重共线性、空间自相关性及显著性检验,筛选12个关键影响因素构建普通最小二乘回归(OLS)、传统地理加权回归(GWR)与MGWR模型,定量分析了轨道交通与常规公交客流竞争格局的时空分异机理。【结果】轨道交通在整体客流分担中占主导,常规公交则在部分线网节点处分担地铁站影响范围内近35%的客流。MGWR模型的拟合优度接近0.8,综合性能最优,证实了二者客流竞争空间格局的形成机制高度依赖建成环境等外生因素的多尺度作用,其中轨道交通在其换乘站点、高房价区域、就业岗位集聚区、公共文体场馆与城市复合功能区等区域竞争优势显著,常规公交则在人口和教育资源密集、绿地景点周边更具相对竞争力。【应用】为基于地理空间差异,因地制宜构建轨道交通与常规公交优势互补、协同服务的城市公共交通系统提供了实证参考。
[Background]The rapid expansion of metro networks in China has intensified the competition between metro and bus in overlapping service areas.Previous studies typically overlook the spatial heterogeneity of environmental factors and thus cannot adequately explain the complex spatiotemporal mechanisms underlying this competition.[Objective]This study integrates spatial influences from socioeconomic features,network topology,and the built environment into an analytical framework using a multiscale geographically weighted regression(MGWR)model to overcome the limitations of globally fixed parameters and single-scale analysis in conventional models.[Methods]Focusing on the“5+1”central areas of Chengdu,this study calculated the ridership ratio between metro and bus on both weekdays and non-weekdays to reveal their spatial competition patterns.By integrating multi-source spatiotemporal data and performing tests for multicollinearity,spatial autocorrelation,and significance,12 key factors were selected to construct OLS,GWR,and MGWR models,thus quantitatively uncovering spatiotemporal heterogeneity in the ridership competition.[Results]The findings indicate that metro dominates the overall ridership share,whereas bus constitutes approximately 35%of the total ridership within the buffer zones of metro stops at certain network nodes.The goodness-of-fit of the MGWR model is close to 0.8 and its overall performance is superior,thus confirming that the formation mechanism of the ridership competition pattern between metro and bus depends significantly on the multiscale effects of exogenous factors such as the built environment.Metro shows distinct competitive advantages in transfer stops,high-priced residential areas,employment clusters,public cultural and sports facilities,and mixed-use zones,whereas bus dominates densely populated neighborhoods,schools,and tourism areas.[Application]The empirical evidence derived from this study contributes to the formulation of a geographically contextualized public transit system and the strategic enhancement of the intermodal complementarity between metro and bus networks.
作者
姜小美
徐占东
陈正贤
刘晓波
JIANG Xiaomei;XU Zhandong;CHEN Zhengxian;LIU Xiaobo(School of Transportation and Logistics,Southwest Jiaotong University,Chengdu 611756,China;National Engineering Laboratory of Integrated Transportation Big Data Application Technology,Chengdu 611756,China;National United Engineering Laboratory of Integrated and Intelligent Transportation,Chengdu 611756,China)
出处
《交通运输工程与信息学报》
2025年第3期213-230,共18页
Journal of Transportation Engineering and Information
基金
国家自然科学基金项目(72201220)
四川省自然科学基金项目(2024NSFSC1056)。
关键词
城市公共交通
客流竞争
多尺度地理加权回归
轨道交通
常规公交
空间异质性
urban public transit
ridership competition
multiscale geographically weighted regression
metro
bus
spatial heterogeneity
