摘要
出行需求预测是交通规划活动中的重要环节,以往的出行需求建模方法依赖于传统的交通数据采集方法,效率低、成本高。笔者提出一种大数据环境下的城市出行需求建模方法,并以南京市为例进行了应用研究,利用移动社交网络数据和手机数据构建了城市常住居民和短期旅居人员的出行生成量和吸引量预测模型。首先基于手机信令数据对城市常住居民和短期旅居人员的出行特征进行了分析,然后将移动社交网络的兴趣点(POI)数据按服务类型划分为不同的类别,将每个类别的POI数量按交通分析区域进行聚合,作为出行生成量和吸引量预测模型的解释变量,同时在解释变量中加入了区域面积、小区内的公路入口数量和公共交通站点数量等建成环境因素;采用探索性回归分析的方法对各种变量组合的性能进行检验,并确定了城市常住居民和短期旅居人员的出行生成量和吸引量预测模型的最佳模型。该研究成果可推广应用于国内大中型城市交通规划项目的出行需求预测环节。
Travel demand forecasting is an important part of transportation planning activities.Traditional traffic data collection methods,which are used by previous travel demand model(TDM)methods,are low efficiency and high cost.This paper introduces an urban TDM method in a big data environment and conducts a case study in Nanjing,builds a travel generation and attraction model for urban residents and short-term travelers by location-based social networking data and cell phone data.Firstly,the paper analyzes the travel characteristics of urban residents and short-term travelers based on cell phone signaling data.Then,the point-of-interest(POI) data of LBSN is divided into different categories by service type,and the number of POIs in each category is aggregated by traffic analysis area as the explanatory variables of the trip generation and attraction model.In addition,the explanatory variables incorporate built environment factors,such as area of the area,the number of highway entrances and the number of public transportation stops in the area.The performance of the various combinations of variables is tested by exploratory regression analysis,and the best model for the trip generation and attraction model of urban residents and short-term travelers is determined.The research results of the paper can be extended and applied to the travel demand forecasting of transportation planning projects of large and medium-sized cities in China.
作者
杨帆
徐光霁
Yang Fan;Xu Guangji(School of Information and Communication Engineeringt Nanjing Institute of Technology,Nanjing 211161,China;School of Transportation,Southeast University,Nanjing 210096,China)
出处
《市政技术》
2022年第3期52-57,共6页
Journal of Municipal Technology
基金
国家自然科学基金(71701044)
南京工程学院高层次引进人才科研启动基金(YKJ202114)
东南大学优秀青年教师科学研究资助项目(2242021R41132)。
关键词
出行需求建模
手机数据
移动社交网络数据
城市交通规划
travel demand model(TDM)
cell phone data
mobile-social-networking data
urban transportation planning
