基于MGWR的共享单车空间异质性分析  

Spatial Heterogeneity Analysis of Bikesharing Based on MGWR

作　　者：朱岑远 郑乐 张毅萌 ZHU Cenyuan;ZHENG Yue;ZHANG Yimeng(School of Modern Posts,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]南京邮电大学现代邮政学院,江苏南京210003

出　　处：《物流科技》2024年第24期72-77,共6页Logistics Sci Tech

基　　金：国家自然科学基金(52102381);中国博士后科学基金(2023M731778);江苏省高校哲学社会科学一般项目(TJZ221042)。

摘　　要：基于南京市多源数据对共享单车出行需求密度进行了分析。选取社会经济人口变量、土地利用变量和公共交通变量三类变量为解释变量,构建了多尺度地理加权回归模型,揭示了共享单车需求密度的影响因素及其空间异质性关系。研究表明:第一,多尺度地理加权回归模型的拟合结果显著优于最小二乘回归模型与地理加权回归模型;第二,与主城区不同,郊区居住人口密度越大,共享单车订单需求密度越小;第三,市中心的共享单车投放密度已经趋于饱和;第四,地铁站点对共享单车的需求起促进作用;第五,私人自行车拥有率大部分情况下对共享单车的需求起促进作用,而私家车拥有率对共享单车的需求起抑制作用。Based on the multi-source data of Nanjing,this paper analyzed the travel demand density of shared bicycles.Three kinds of variables including socio-economic population variables,land use variables and public transport variables were selected as explanatory variables,and a multi-scale geographical weighted regression model was constructed to reveal the influencing factors and spatial heterogeneity relationship of the demand density of bike-sharing.The results show that:(i)the fitting results of the multi-scale geographically weighted regression model are significantly better than those of the least squares regression model and geographically weighted regression model;(ii)different from the main urban area,the greater the residential population density in the suburbs,the smaller the demand density of bike-sharing orders;(iii)the density of bike-sharing in the city center has become saturated;(iv)metro stations promote the demand for shared bicycles;(v)in most cases,private bicycle ownership promotes the demand for shared bicycles,while private car ownership inhibits the demand for shared bicycles.

关 键 词：城市交通 公共交通 共享单车 需求密度 空间异质性 多尺度地理加权回归模型 

分 类 号：F57[经济管理—产业经济] U121[交通运输工程]