高速公路交通事故严重程度与交通流特征的关系研究
Relationships between Crash Severity and Traffic Flow Characteristics on Freeways
侯树展 1孙小端 1贺玉龙 1田启华1
作者信息
- 1. 北京工业大学北京市交通工程重点实验室,北京100124
- 折叠
摘要
为研究交通事故发生前后交通流特征对事故严重程度的影响,以JT高速公路作为研究对象,长期观测和采集交通流及事故数据.将交通事故发生时段的交通流主要衡量指标与事故信息进行数据匹配,形成交通事故与事故小时交通流匹配数据集,并分析流量、速度、大车比例等交通流表征指标与不同等级事故数的分布规律.通过分析发现:在某些流量、速度或大车比例区段,交通事故数及其严重程度处于较高的水平.在此基础上,利用主成分分析(PCA)技术对衡量交通流特征的初始指标进行降维处理,用交通流主成分指标综合反映交通流特征,并建立事故严重程度与交通流主成分指标的统计分析模型.结果表明:交通流主成分指标趋于零的区段的事故严重程度明显高于其他区段.
Abstract
In order to determine the relationships between traffic flow characteristics in the crash hour and crash severity, traffic flow and crash data were obtained from the JT Freeway for seven months. Detailed information about traffic flow characteristics was matched for each crash to build a data set of crashes and traffic flow in the crash hour. The distributing rules of traffic flow and crashes show that the number of crashes and crash severity peaks in some sections of volume, speed or proportion of trucks. On the basis of above analysis, using principal component analysis, a principal component index of traffic flow was presented to stand for the traffic flow characteristics by dimensionality reduction. Statistical analysis models for the relationship between crash severity and traffic flow principal component index were developed for freeway. The study shows crashes are more serious when the value of traffic flow principal component index is close to zero.
关键词
事故严重程度/交通流/事故小时/主成分分析(PCA)/高速公路Key words
crash severity/traffic flow/crash hour/principal component analysis (PCA)/freeway引用本文复制引用
基金项目
国家科技支撑计划(2009BAG13A02)
出版年
2011
NETL
NSTL
维普
万方数据