中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

Journal of Chongqing Jiaotong University(Natural Science) ›› 2023, Vol. 42 ›› Issue (5): 124-131.DOI: 10.3969/j.issn.1674-0696.2023.05.16

• Transportation Infrastructure Engineering • Previous Articles    

Cause Analysis of Traffic Accidents in Jinan Based on GIS

FENG Haixia,NING Erwei,WANG Qi,LI Jian   

  1. (School of Transportation and Logistics Engineering, Shandong Jiaotong University, Jinan 250399, Shandong, China)
  • Received:2021-12-17 Revised:2022-04-17 Published:2023-07-13

基于GIS的济南市交通事故成因分析

冯海霞,宁二伟,王琦,李健   

  1. (山东交通学院 交通与物流工程学院,山东 济南 250399)
  • 作者简介:冯海霞(1976—),女,山东聊城人,副教授,博士,主要从事定量遥感、智能交通方面的研究。E-mail:547500148@qq.com 通信作者:宁二伟(1998—),男,山西运城人,硕士,主要从事交通规划与管理方面的研究。E-mail:2496526326@qq.com
  • 基金资助:
    山东省重点研发计划项目(公益类)(GG201809190147)

Abstract: In order to reduce traffic accidents and ensure peoples normal life and personal safety, identifying traffic accident-prone areas and analyzing the causes of accidents are the key to reduce accidents. Taking Jinan as the research area, based on the traffic accident data of Jinan in 2018, the spatial characteristics and causes of traffic accidents in Jinan were studied by the methods such as kernel density analysis, accident equivalent combined with buffer zone analysis and geographic weighted regression analysis. Meanwhile, the corresponding improvement measures were proposed. The research results show that the accident sites have obvious aggregation, and six accident-prone areas in Jinan have been identified. The lack of road physical isolation and roadside protection facilities are the main reasons for the accident-prone areas. The number of vehicles, highway mileage, GDP and population have a great impact on the number of regional accidents, and they are positively correlated with the number of accidents, among which the number of vehicles has the greatest impact. The correlation regression coefficient of the GWR model constructed on the basis of the number of vehicles and highway mileage is as high as 0.75, which can provide timely warning and improve corresponding transportation facilities for accident-prone areas in various regions according to the proposed model. This study is of great significance to reduce the occurrence of traffic accidents and improve the level of traffic safety.

Key words: traffic and transportation engineering; traffic accident; kernel density analysis; buffer analysis; equivalent accident number method; geographically weighted regression

摘要: 为了降低交通事故,保障人们的正常生活和人身安全,对交通事故多发区域进行鉴别并对事故成因进行分析是减少事故的关键。以济南市为研究区,以2018年济南市的交通事故数据为基础,采用核密度分析、事故当量结合缓冲区分析、地理加权回归分析等方法,对济南市交通事故的空间特性、成因等进行了研究,并提出了相应的改善措施。研究结果表明:事故发生地点具有明显的聚集性,鉴别出了济南市6个事故多发区域,无道路物理隔离和无路侧防护设施是事故多发的主要原因;车辆数、公路通车里程、GDP、人口等均对区域交通事故数有较大影响,且均与交通事故数呈正相关的关系,其中车辆数的影响最大;构建的基于车辆数和公路通车里程的GWR模型的相关性回归系数高达0.75,可根据模型对各区县的事故多发区域进行及时预警和完善相应的交通设施。研究结果对减少交通事故的发生以及道路交通安全水平的提升具有重要意义。

关键词: 交通运输工程;交通事故;核密度分析;缓冲区分析;当量事故数法;地理加权回归

CLC Number: