摘要
基于出行行为视角深入分析公共交通出行者的行为模式及特征对靶向改善公共交通服务水平具有重要意义.研究通过RP调查获取出行者个体特征信息,并在分析海量智能卡交易数据的基础上,结合关联匹配方法提取公共交通通勤出行链;基于探索性因子分析筛选出行天数、日均出行频次和出行完整度以及个体社会经济属性等9个指标刻画乘客公共交通使用行为;在对连续性变量离散化的基础上,利用DBSCAN算法构建乘客公共交通使用行为辨识模型.结果表明:构建的聚类算法可有效识别公共交通使用行为类别;调查群体被划分为公共交通高、中、低3类使用度群组,占比分别为54.2%、33.7%和12.1%,并将第3类人群视为公共交通使用行为改善潜力最大群体,未来应结合交通限制政策与服务水平2个维度改善此类公共交通乘客的使用行为.
Specifically analysis of the behavior patterns and characteristics of public transport travelers from the perspective of travel behavior is of great significance for targeted improvement of public transport service level.The individual characteristics of travelers were obtained by the RP survey,and the commuting travel chain of public transportation is extracted by the association matching method based on the analysis of massive smart card transaction data.The 9 indicators,including travel days,daily travel frequency,travel integrity and individual socio-economic attributes,are selected to describe passenger public transport usage behaviour based on exploratory factor analysis.On the basis of discretization of continuous variables,DBSCAN algorithm is used to construct the identification model of passenger public transportation usage behaviour.The results show that the constructed clustering algorithm can effectively identify the categories of public transport usage behaviour.The respondents are divided into three categories:high,medium and low usage groups of public transport,accounting for 54.2%,33.7%and 12.1%respectively,and the third group is regarded as the group with the greatest potential to improve public transport behavior.It is necessary to improve the usage bahaviour of such public transport passengers from two dimensions of traffic restriction policy and service level.
作者
胡松
杨贝
翁剑成
王海鹏
常征
HU Song;YANG Bei;WENG Jiancheng;WANG Haipeng;CHANG Zheng(Research Institute of Highway Ministry of Transport,Beijing 100088,China;Zhonglu Gongke(Beijing)Consulting Co.,Ltd.,Beijing 100088,China;Faculty of Urban Construction,Beijing University of Technology,Beijing 100124,China)
出处
《交通工程》
2023年第5期71-76,共6页
Journal of Transportation Engineering
基金
交通运输部公路科学研究所(院)交通强国试点项目(QG2022-2-8-4)
国家自然科学基金(52072011).
