基于荟萃分析的事故修正系数计算方法  

Calculation Method of Crash Modification Factor Based on Meta-analysis Method

作　　者：陆佳欢 羊钊[1] 王一凡 唐荣 LU Jiahuan;YANG Zhao;WANG Yifan;TANG Rong(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)

机构地区：[1]南京航空航天大学民航学院,南京211106

出　　处：《武汉理工大学学报（交通科学与工程版）》2021年第2期237-242,共6页Journal of Wuhan University of Technology（Transportation Science & Engineering）

基　　金：国家自然科学基金项目(51608268,71971112);南京航空航天大学基本科研业务费项目(NS2020046)。

摘　　要：提出一种基于荟萃分析的事故修正系数计算方法.将无信号交叉口直接左转设计改造为远引掉头设计,通过异质性分析,表明数据总体异质性为24.9%,各研究之间存在轻度异质性.荟萃回归结果显示,引入碰撞类型、调查地点、碰撞严重程度以及研究方法后,分别产生50.84%、9.15%、18.43%和8.42%的异质性来源.采用荟萃分析法集成多个研究结果,得到的关于远引掉头设计事故修正系数更准确.A method to estimate the crash modification factors(CMF)based on meta-analysis methods was proposed.The design of direct left turn at unsignalized intersection was transformed into the design of far turn.Heterogeneity analysis shows that the overall heterogeneity of data is 24.9%,and there is slight heterogeneity among the studies.Meta-regression results show that after introducing collision types,investigation locations,collision severity and research methods,50.84%,9.15%,18.43%and 8.42%of heterogeneous sources are generated respectively.Meta-analysis method is used to integrate multiple research results,and the accident correction coefficient of the far-off U-turn design is more accurate.

关 键 词：荟萃分析 事故修正系数 无信号交叉口 直接左转 远引掉头 

分 类 号：U8[交通运输工程]