土地利用视角下城市轨道交通线路分类及其客运量影响因素研究

Classification of urban rail transit lines and determinants of the ridership from the perspective of land use

理解城市轨道交通客流与土地利用的依赖关系有助于合理配置资源以推动TOD(Transit-oriented development)模式的实施。既有研究多以车站客流及其建成环境为研究对象揭示客流的关键影响因素,从全局视角下分析线路客流对其沿线土地利用依赖关系的研究不够充分。论文基于2019年中国36座主要城市的多源数据构建包含线路功能和建成环境两方面的土地利用以及交通服务等三类指标,并以线路功能和建成环境两类指标为聚类特征划分线路类型,采用XGBoost(eXtreme Gradient Boosting)模型分析不同类型线路客运量的影响因素。案例研究表明:根据线路服务范围及城市发展水平的差异,城市轨道交通线路可分为市区线I型、市区线II型和郊区线。3类线路重要度排序前5的解释变量对客运量的贡献至少达到60%,其中就业岗位密度和中心城区线路覆盖率是3类线路共同的重要土地利用变量。重要度排序高的解释变量对客运量的影响具有阈值效应,且相同类型线路重要变量的阈值趋同,而不同类型线路之间的阈值存在显著差异。因此,TOD客流廊道的构建需因地制宜地统筹职住分布及其配套资源。研究结果有助于理解土地利用影响城市轨道交通线路客流的机理,可为促进TOD客流廊道的发展提供参考。

The comprehension of the interdependence between urban rail transit ridership and land use is beneficial for rational resource allocation to facilitate the implementation of transit-oriented development(TOD).Existing studies have primarily focused on ridership and the built environment of stations to identify key influencing factors,while there remains a lack of research from a global perspective regarding the dependence of ridership on land use along urban rail transit lines.To this end,this study used multi-source urban geographic data from 36 major cities in China's mainland in 2019 to establish line function and built environment factors that depict land use,and classify line types using the two as clustering features,while also establishing transportation service factors.XGBoost(eXtreme Gradient Boosting)was employed to analyze the influencing factors of ridership across different types of lines.Taking seven supercities such as Beijing,14 megacities such as Wuhan,and 15 big cities such as Nanning and their urban rail transit systems as research objects,the results show that urban rail transit lines can be categorized into downtown line typeⅠ,downtown line type II,and suburb line,based on the different service scopes of the lines and variations in urban development levels.The cumulative contribution of the explanatory variables in the top five,which are of utmost importance to ridership,accounts for at least 60%across all three types of lines.Among these,both employment density and the coverage level of line on the downtown area emerged as significant land use explanatory variables shared by all three types of lines.Moreover,it was observed that important explanatory variables exhibited a threshold effect on ridership,with converging thresholds among important variables within the same type of line but notable difference between different types.Therefore,the construction of TOD corridor should align the distribution of employment and residential areas,as well as supporting resources,in accordance with local conditions.The research findings can help understanding the mechanism of land use affecting urban rail transit ridership and provide a reference for promoting the development of corridor TOD.