建成环境对轨道站点早高峰步行接驳比例的影响研究
Study on the Influence of Built Environment on the Pedestrian Access Ratio of Rail Transit Stations during Morning Peak
费跃 1温旭丽 1刘云 1毛盈盈1
作者信息
摘要
为了分析城市建成环境对轨道交通站点早高峰步行接驳比例的影响,文章提出利用梯度提升决策树(GBDT)模型研究建成环境与步行接驳比例间的非线性关系.以南京地铁为例,获取开放平台POI数据,结合规划路网数据、接驳调查数据等多源数据,选取并计算站点周边半径 800 m覆盖范围内的建成环境指标,利用GBDT算法进行回归分析,得到各指标对步行接驳比例的相对重要度和部分依赖图.研究结果表明,就业岗位密度、路网密度、用地混合熵指数、出入口方向覆盖率和公交站点密度对早高峰进出站步行接驳比例有明显的正向影响,步行可达性存在双向的影响,而住户密度和出入口数量的影响是负向的.
Abstract
In order to analyze the influence of built environment on the pedestrian access ratio of rail transit stations during morning peak,a gradient boosted decision tree(GBDT)model was proposed to study the nonlinear relationship between built environment and pedestrian access ratio.Taking Nanjing metro as an example,POI data of the open platform combined with multi-source data such as road network planning data and access survey data were obtained,then built environment indexes within the coverage area of 800 m surrounding the station were selected and calculated.GBDT algorithm was used for regression analysis to obtain the relative importance degree and partial dependence graph of each index.The results indicate that job density,road network density,land use mixed entropy index,coverage rate in directions of station entrance and exit and bus station density have a significant positive impact on the proportion of pedestrian access in and out of the station during morning peak,and walking accessibility has a two-way effect,while the impact of household density,quantity of entrance and exit are negative.
关键词
建成环境/步行接驳比例/POI数据/GBDT模型/非线性关系Key words
built environment/pedestrian access ratio/POI data/GBDT model/non-linear relationship引用本文复制引用
基金项目
江苏省高等学校基础科学(自然科学)面上项目(21KJB580019)
东南大学成贤学院青年教师科研发展基金项目(z0030)
出版年
2024
NETL
NSTL
维普
万方数据