【摘要】 为解决传统的热点分析方法无法识别相对独立的事故多发点，且识别结果受极值影响较大的问题，提出1种基于泰森多边形的事故多发点识别方法，基于2018年江苏省盐城市交通事故数据，用泰森多边形划分空间统计单元，依据单元面积和其内部事故数量的比值识别事故多发点，补充相对独立的事故多发区域，并用缓冲区修正多边形的形状。研究结果表明：此方法识别结果能有效避免极值的影响，能更准确地识别出事故多发点，更有效地为道路交通安全管理提供依据。更多还原

【Abstract】 In order to solve the problem that the traditional hot spot analysis methods cannot identify the relatively independent accident-prone locations, and the identification results are greatly affected by the extreme values, a method for identifying the accident-prone locations based on Tyson polygon was proposed.Based on the traffic accident data of Yancheng, Jiangsu in 2018,the spatial statistical units were divided by Tyson polygon, and the accident-prone locations were identified based on the ratio of the unit area and the number of internal accidents.The relatively independent accident-prone areas were supplemented, and the shape of polygon was corrected by the buffer zone.The results showed that the identification results of this method could effectively avoid the influence of extreme values, and more accurately identify the accident-prone locations.It can provide a basis for the road traffic safety management more effectively.更多还原