适用于一票制公交大数据的系统化处理方法及应用——以银川市为例  

Systematic Processing Method for Big Data of One-vote Bus and Its Application: A Case Study of Yinchuan

作　　者：赵海宾[1] 郭忠[1] 吴洪洋[1] 刘海旭 王子甲 Zhao Hai-bin;Guo Zhong;Wu Hong-yang;Liu Hai-xu;Wang Zi-jia(China Urban Sustainable Transport Research Centre(CUSTReC),China Academy of Transportation Sciences,Beijing 100029,China;Transportation Research Center,Beijing Urban Construction Design&Development Group Co.Limited,Beijing 100032,China;Department of Highway and Railway Engineering,School of Civil and Architectural Engineering,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]交通运输部科学研究院城市交通与轨道交通研究中心,北京100029 [2]北京城建设计发展集团股份有限公司交通研究中心,北京100032 [3]北京交通大学土木建筑工程学院道路与铁道工程系,北京100044

出　　处：《交通运输研究》2020年第4期109-116,共8页Transport Research

基　　金：中央级公益性科研院所基本科研业务费项目(20194807)。

摘　　要：为分析公交乘客出行特征,利用公交IC卡数据、公交GPS数据、车载机数据和单程站点关系表,通过各类数据关联融合,提出适用于一票制公交大数据的系统化处理方法。基于Oracle搭建分析数据库,采用Python语言编写代码,构建了乘客上车站点推断算法、基于出行链的乘客下车站点推断算法、基于概率的乘客下车站点推断算法和乘客换乘站点识别算法4种站点推断算法。基于此,运用银川公交大数据进行客流集散点识别、客流走廊识别,得到公交站点上下客流量分布情况、公交线路客流量分布情况、公交站点的换乘客流量分布情况。研究结果表明,一票制公交大数据系统化处理方法在分析公交乘客出行特征方面具有一定的应用价值。In order to analyze bus passengers′ travel characteristics, a systematic processing method for big data of one-vote bus was put forward by association and fusion of bus IC card data, bus GPS data, on-board machine data and one-way station relation table. By building an analysis database on Oracle and through Python language programming code, four station inference algorithms were established, including passenger boarding station inference algorithm, passenger alighting station inference algorithm based on travel chain, passenger alighting station inference algorithm based on probability,passenger transfering station recognition algorithm. On this basis, Yinchuan bus data was applied to identify its passenger flow distribution center and passenger flow corridor. The distribution of passenger flow at bus stops, the distribution of passenger flow on bus lines, and the distribution of transfering passenger flow at bus stops were obtained. The results show that the systematic processing method for big data of one-vote bus is valuable in analyzing bus passengers′ travel characteristics.

关 键 词：城市公共交通 一票制公交 站点推断 公交IC卡数据 公交GPS数据 

分 类 号：U491.1[交通运输工程—交通运输规划与管理]