| 摘要: |
| 轨道交通站点客流作为轨道交通网络
化运营的基础，其时空需求和特征分布直接关
系到城市综合交通规划、城市土地使用、空间
结构和设施布局。既有研究对建成环境与轨
道交通站点客流量之间关系的分析相对成熟，
但对不同时段站点客流量影响因素的研究相
对不足。基于多源地理空间数据构建“5D+N”
维度的建成环境指标体系，采用OLS、GWR、
SGWR模型解析建成环境对不同时段站点客流
量的影响效果。针对天津的案例研究表明：不同
时段站点客流量影响因素类型及其作用方向存
在差异，站点开通时长、出入口数量、行政办公
设施POI密度、公交站点密度等因素对多个时段
客流量都有正向作用，而POI混合度与距公交站
点平均距离则有负向作用；建成环境对站点客
流量的影响呈现全局效应与局部效应的差异，
其中密度维度的建成环境因素多为局部影响变
量；局部影响变量对站点客流量的作用方向及强度表现出空间异质性特征。不同时段客流量影响因素类型与作用方向的差异性以及影响效应的
空间异质性特征为客流效能的提升与TOD高质量开发提供了分时分区的差异化策略指引, 也为站
点周边建成环境的更新优化提供了引导。 |
| 关键词: 城市轨道交通 客流量 建成环境 半参数地理加权回归 空间异质性 |
| DOI：10.13791/j.cnki.hsfwest.20240612 |
| 分类号: |
| 基金项目:国家自然科学基金面上项目（52278070） |
|
| Analysis of the influence mechanism of the built environment on the passenger flow ofurban rail transit stations during different time periods |
|
PANG Lei,REN Lijian,JIANG Yuxiao,SUN Zhong,YUN Yingxia
|
| Abstract: |
| The passenger flow of urban rail transit stations reflects the agglomeration effect of
passenger flow. As the foundation for operating an urban rail transit network, the spatial and
temporal demand and characteristic distribution of station passenger flow are directly linked to
comprehensive transportation planning, urban land use, spatial structure, and facility layout. In
this context, exploring the impact mechanism of the built environment on the passenger flow of
urban rail transit stations and accurately identifying key built environment factors that affect station
passenger flow play a crucial guiding role in revitalizing the urban built environment organically and
enhancing the efficiency of station passenger flows in urban rail transit systems. This has become one
of the focal points in studying urban transportation planning and transportation geography. While
existing studies on the relationship between the built environment and passenger flow of urban
rail transit stations are relatively mature, there is a relative lack of research on factors influencing
station passenger flow during different time periods. To address the lack of existing research, we
developed a comprehensive “5D+N” index system for the built environment based on multi-source
geospatial data (including land use data, mobile signaling data, POI data, building footprint data,
road network data, bus station and line network data, urban rail transit station and line network data,
etc.). The study then selected average daily passenger flow and peak inbound/outbound passenger
flow during morning and evening periods as dependent variables for urban rail transit stations.
Using multicollinearity testing and OLS stepwise regression, it identified significant independent
variables that influence station passenger flow. Finally, by comparing the fitting effects of OLS,
GWR, and SGWR models, it determined the superior model to analyze how the built environment
impacts passenger flow during different time periods. The case study of Tianjin yields the following
conclusions. 1) The average daily passenger flow of urban rail transit stations exhibits a distribution
pattern characterized by high passenger flow in central areas and low ridership in peripheral areas.
Additionally, the distribution of station passenger flow during peak hours also displays spatial
heterogeneity. Stations with high passenger flow during morning and evening peaks are primarily
transportation transfer hubs and the first and last stations located in the western periphery of the
city. Moreover, stations with significant passenger flow during these peak periods are concentrated
in the core area of the city. The heterogeneity in station passenger flow is mainly attributed to
variations in urban land development intensity, diversity in station spatial function and business mix,
as well as differences in regional rail network density. 2) There exist both similarities and notable
differences regarding factors influencing passenger flow at different time periods. Among them,
factors such as opening time of stations, number of entrances/exits, density of administrative office
facilities’ points-of-interest (POI), density of bus stops have a positive impact on passenger flow
across multiple periods. Conversely, POI mixing degree and average distance from bus stops exert anegative effect on passenger flow. Notably, significant factors affecting morning peak inbound/outbound passenger flow exhibit relative similarity while those
impacting evening peak outbound/inbound passenger flow also demonstrate some level of similarity. 3) The impact of the built environment on passenger
flow at the station demonstrates a distinction between global and local effects, with the density dimension of the built environment primarily influencing
locally. The direction and intensity of local variables on passenger flow exhibit spatial heterogeneity. Variations in factors influencing passenger flow during
different periods and their spatial heterogeneity offer guidance for enhancing passenger flow efficiency and promoting high-quality TOD development, as
well as informing updates and optimizations to the built environment surrounding the station. Taking Tianjin City as a case study, this paper conducted an
empirical investigation to explore the methodology and approach for collaborative optimization and enhancement of the built environment and urban rail
transit passenger flow efficiency. The research findings not only broaden the research perspective on the relationship between the built environment and
urban rail transit passenger flow but also advance the frontier exploration direction of multidisciplinary integration encompassing transportation, planning,
and geography. Simultaneously, it enriches the methodological framework of rational spatial planning research under a “flow” logic, enhances the scientific
nature |
| Key words: urban rail transit passenger flow built environment semi-parametric geographically weighted regression spatial heterogeneity |
