摘要
为了针对性地识别与治理城市道路事故黑点,需要深入挖掘事故特征。文中采用国内某市内城市道路交通事故数据,分别提取年事故数、每公里事故数为指标,利用空间聚类算法对事故点进行聚类,经过数据可视化得到城市道路交通事故风险分布的聚类结果图并进行特征分析。结果表明,该方法可以有效关联事故多发位置鉴别与特征分析,能在路网层级准确识别事故多发区域,为城市道路交通安全管理提供决策指导。
In order to identify and ameliorate the accident black spots targetedly, it is imperative to extract the accident risks distribution apart from its level in the scopes of the urban networks. Based on the data sets of accidents happened in urban roads from a domestic city, accidents-prone spots clustering was completed by means of spatial clustering method with adopting the index of the number of annual accidents and the number of annual accidents per kilometer respectively. The clustering result diagrams depicting the accident risks distribution in the whole urban networks were also obtained to analyze features. The results show that the proposed method can correlate the identification to the features analysis of accident-prone locations, and accurately identify accident-prone locations at the level of road networks. Eventually it will contribute to providing decision guidance for urban road traffic safety management.
作者
樊鹏程
董宪元
FAN Pengcheng;DONG Xianyuan(College of Transportation Engineering, Tongji University, Shanghai 201804, China;Tongji University, The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Shanghai 201804, China;Tongji University, Ministry of Education & Engineering Research Center of Road Traffic Safety and Environment, Ministry of Education, Shanghai 201804, China;Shanghai Key Laboratory of Rail Infrastructure Durability and System Safety, Shanghai 201804, China)
出处
《交通科技》
2019年第4期85-89,共5页
Transportation Science & Technology
关键词
交通事故
空间聚类
事故挖掘
特征分析
可视化
traffic accident
spatial clustering
accident data mining
features analysis
visualization
