| 基于多源数据的夜间出行需求空间效应及其异质性分析 |
| |
| 引用本文： | 程小云,张学宇,施澄,杨琦,刘钰倩,姚振兴,丁玲.基于多源数据的夜间出行需求空间效应及其异质性分析[J].中国公路学报,2021,34(12):288-301. |
| |
| 作者姓名： | 程小云 张学宇 施澄 杨琦 刘钰倩 姚振兴 丁玲 |
| |
| 作者单位： | 1. 长安大学 运输工程学院, 陕西 西安 710064;2. 长安大学 生态安全屏障区交通网设施管控及循环修复技术交通运输行业重点实验室, 陕西 西安 710064;3. 同济大学 建筑与城市规划学院, 上海 200092;4. 长安大学 经济与管理学院, 陕西 西安 710064 |
| |
| 基金项目： | 国家自然科学基金青年科学基金项目（52002030，51808392） |
| |
| 摘 要： | 为了揭示多维因素对出租车夜间出行需求的空间影响机制与局部关联特征，基于多源数据，利用空间计量模型和多尺度地理加权回归（MGWR）模型分别从全局和局部视角研究多维因素对夜间出行需求影响的空间效应及其异质性。首先，利用出租车GPS数据在分析夜间出行时空分布特征的基础上，明确夜间场景下的研究区域及时段划分；其次，从建成环境、人口分布、路网结构3个维度提出夜间出行需求影响因素指标体系，并基于POI数据、手机信令数据和路网数据进行指标量化。在交通小区尺度下分时段构建空间杜宾模型，从全局视角探讨不同时段下主要影响因素对夜间出行需求的空间效应；通过构造距离负指数衰减形式的空间权重矩阵，量化分析关键影响因素的空间溢出效应随地理距离变动的过程。最后，借助MGWR模型从局部视角分析主要影响因素对夜间出行需求影响的空间异质特征。研究结果表明：各解释变量对夜间出行需求的直接效应大多在0.05水平下显著，而溢出效应差异明显；美食与停车设施密度影响最为显著，两者的溢出效应在不同时段分别表现为负向和正向影响，即美食设施呈现出虹吸现象，停车设施具有诱发作用；停车和美食设施对夜间出行需求的空间溢出效应随距离变化均具有非线性衰减特征，分别呈现先增强后减弱的倒“U”形过程和负向先减弱后增强的过程，并在地理距离分别为1 500 m和1 000 m时，空间溢出效应达到最强；MGWR模型拟合效果优于经典GWR模型，局部回归系数的空间分布模式表明关键影响因素对夜间出行需求空间作用的异质特征显著。研究结论从夜间设施影响范围及强度的空间差异等方面为夜间交通需求的科学预测提供理论依据，进而为夜间商业及交通配套设施的合理配置提供方法支撑。
|
| 关 键 词： | 交通工程 空间溢出效应 空间计量模型 夜间出行 建成环境 |
| 收稿时间： | 2021-04-30 |
Analysis of Spatial Effect and Its Heterogeneity on Night-time Travel Based on Multi-source Data |
| |
| CHENG Xiao-yun,ZHANG Xue-yu,SHI Cheng,YANG Qi,LIU Yu-qian,YAO Zhen-xing,DING Ling.Analysis of Spatial Effect and Its Heterogeneity on Night-time Travel Based on Multi-source Data[J].China Journal of Highway and Transport,2021,34(12):288-301. |
| |
| Authors: | CHENG Xiao-yun ZHANG Xue-yu SHI Cheng YANG Qi LIU Yu-qian YAO Zhen-xing DING Ling |
| |
| Affiliation: | 1. School of Transportation Engineering, Chang'an University, Xi'an 710064, Shaanxi, China;2. Key Laboratory of Transport Industry of Management, Control and Cycle Repair Technology for Traffic Network Facilities in Ecological Security Barrier Area, Chang'an University, Xi'an 710064, Shaanxi, China;3. College of Architecture and Urban Planning, Tongji University, Shanghai 200092, China;4. School of Economics and Management, Chang'an University, Xi'an 710064, Shaanxi, China |
| |
| Abstract: | To reveal the influence mechanism and local relationship of multidimensional factors on taxi night-time travel demand, spatial econometric models and a multi-scale geographically weighted regression (MGWR) model were utilized to study the spatial effect and its heterogeneity of multidimensional factors on night-time travel demand from the global and local perspectives based on multi-source data. First, the spatiotemporal distribution characteristics of night-time travel were analyzed and then the study area and time period were defined based on the taxi GPS data. Second, a comprehensive indicator system of factors influencing the night-time travel demand was proposed from three dimensions-built environment, population distribution, and road network structure. These indicators were measured using POI data, cellular mobile phone data, and road network data. This paper constructed the spatial Durbin model for different time periods at the traffic analysis zone (TAZ) scale and discussed the spatial effect of influencing factors on night-time travel demand from a global perspective. The spatial spillover effect of main influencing factors varying with geographical distance was quantified by constructing a spatial weight matrix based on the distance negative exponential decay function. Finally, by using the MGWR model, the spatial heterogeneity of the main influencing factors on night-time travel demand was analyzed from a local perspective. The results show that most of the explanatory variables have significant direct effects on night-time travel demand at 0.05 significant level, but the spillover effects vary greatly. The density of catering service and parking facilities have the most significant impact. Their spillover effects are negative and positive respectively in different time periods, which means catering service facilities show the siphon phenomenon, and parking facilities have an inducing effect. The spatial spillover effects of parking and catering service facilities present nonlinear characteristics with the increasing geographic distance. The former shows an inverted U-shaped pattern of strengthening first and then weakening with the distance increasing and reaches the maximum at 1 500 m. While, the latter appears a negative pattern of weakening first and then strengthening and reaches the maximum at 1 000 m. The fitting effect of the MGWR model is better than that of the GWR model. The spatial distribution pattern of the local regression coefficient shows that the spatial heterogeneity of the impact of key influencing factors on night-time travel demand is significant. The conclusions can provide a theoretical basis for the scientific prediction of night-time traffic demand from the aspects of the spatial difference of attraction range and intensity of night-time facilities, and then offers methodological support for the reasonable allocation of night-time commercial and transportation facilities. |
| |
| Keywords: | traffic engineering spatial spillover effect spatial econometric model night-time travel built environment |
|
| 点击此处可从《中国公路学报》浏览原始摘要信息 |
|
点击此处可从《中国公路学报》下载全文 |
|
