摘要
在充分分析轨道交通行程时间组成要素及分布特性的基础上,利用已知AFC数据对行程时间参数进行估计,提出一种基于朴素贝叶斯分类器的轨道交通网络客流分配模型。利用AFC数据估算行程时间组成要素参数,得到OD间每条路径行程时间的均值及方差;将每名乘客的AFC数据作为一个样本,行程时间作为一个特征属性,利用朴素贝叶斯分类器进行概率分类,将每名乘客划分到后验概率最高的某条路径;得到轨道交通OD间每条有效路径的客流。根据广州地铁算例验证及仿真实验结果表明,该方法能较好预测每条有效路径客流分配比例。
Passenger travel time components in urban rail transit and the corresponding distribution characteristicswere fully analyzed and a model of passenger flow assignment in rail transit network based on automatic fare collection（ AFC）and naive Bias classifier was proposed.Firstly,travel time parameters were collected by using AFC data and the mean and variance of each path of travel time between OD were obtained.Secondly,by taking each passenger＇s AFC data as a sample and the travel time as a characteristic property,the probabilistic classification was made using a Naive Bayes classifier.Each passenger was allocated to one path of the highest riding probability.In the end,the passenger flow of each effective path between the rail transits OD was obtained.The example of the Guangzhou subway and the simulation results show that the method can predict the distribution of the passenger flows of each effective path.
作者
翁小雄
汪周盼
黄靖翔
WENG Xiaoxiong;WANG Zhoupan;HUANG Jingxiang(School of Civil Engineering and Transportation,South China University of Technology, Guangdong 510641 ,Guangzhou,P.R.China)
出处
《重庆交通大学学报（自然科学版）》
CAS
北大核心
2018年第6期110-113,120,共5页
Journal of Chongqing Jiaotong University(Natural Science)
基金
国家自然科学基金项目(51108191
51378222)
广东省科技计划项目(2014B090904059
2016A030305001)
关键词
交通工程
城市交通
客流分配
朴素贝叶斯
轨道交通
AFC数据
traffic engineering
urban traffic
passenger flow assignment
Naive Bayes
rail transits AFC data
