摘要
为探究建成环境与网约车出行需求之间的互动关系,基于上海市网约车订单数据,围绕人口密度、道路密度、 POI密度、公交临近度等24种多源地理大数据构建城市建成环境指标。分别针对工作日与休息日的早高峰、午高峰和晚高峰3个时段建立考虑不同影响因素适应性带宽的多尺度地理加权回归(MGWR)模型,研究了建成环境对网约车出行需求的时空异质性影响。结果表明:与普通最小二乘法(OLS)和地理加权回归(GWR)模型相比,MGWR模型在同一时段内对网约车需求的解释力最高可达77.1%,比OLS模型提高0.426,比GWR模型提高0.082,展现出更优的拟合效果;地铁站点、次干道密度对网约车出行需求呈现出较强的空间异质性;工作日、休息日早高峰,城市郊区商务住宅与地铁站点对网约车的正向影响更大;各个高峰时段,市中心次干道密度对网约车需求的正向影响普遍大于其他区域,国际机场附近建成环境因素对网约车需求的正向影响程度都较高,虹桥机场略高于浦东机场。
To investigate the interaction between the built environment and travel demand of online car-hailing,this study utilized Shanghai's online car-hailing order data and constructed the urban built environment indicators with 24 parameters,including population,road,and POI densities,and distance to transit.A multi-scale geographically weighted regression(MGWR) model was developed for the three periods of peak hours of morning,noon and afternoon on weekdays and weekends,considering adaptive bandwidths of influencing factors,and to explore the spatiotemporal heterogeneity of the impact of built environment on the travel demand of online car-hailing.Results indicated that the interpreting ability of MGWR model for online car-hailing demand upped to 77.1%,which was 0.426 more than the OLS model and 0.082 more than the GWR model,showing a better fitting effect.Subway stations and sub-branch roads density showed higher spatial heterogeneity on the demand of online car-hailing.During morning rush hours of weekdays and weekends,suburban business residential areas and subway stations had a stronger positive effect on online car-hailing demand.During every period of rush hours,the positive impact on online car-hailing demand by sub-branch roads density of city center was generally greater than other areas,and the factors of built environment near the international airports also had a greater positive influence on online car-hailing demand,and the impact of Hongqiao Airport was slightly higher than that of Pudong Airport.
作者
李淑庆
雷宇寰
赖辉涛
李胤浩
LI Shuqing;LEI Yuhuan;LAI Huitao;LI Yinhao(School of Traffic and Transportation,Chongqing Jiaotong University,Chongqing 400074,China)
出处
《西南大学学报(自然科学版)》
北大核心
2025年第6期188-200,共13页
Journal of Southwest University(Natural Science Edition)
基金
国家自然科学基金项目(52078070)。
关键词
网约车出行需求
建成环境
空间异质性
多尺度地理加权回归
travel demand of online car-hailing
built environment
spatial heterogeneity
multi-scale geographically weighted regression
