摘要
基于我国西部M市2023—2024年道路交通事故数据,围绕“人-车-路-环境-管理”五大致因维度构建系统性变量体系,引入事故参与方数量作为致因复杂度代理指标,利用TF-IDF加权机制优化Apriori算法,通过对伤亡事故复合致因模式的挖掘,提升低频高价值致因的识别能力。研究结果显示:车辆违法(不按规定让行、违反信号灯)与行人违法穿行是导致伤亡事故的核心因素;低能见度、交通设施效能不足加剧了事故风险;“刮撞行人”的事故形态在人员受伤中高频出现;死亡事故呈现人、车、路、环境、管理多因素交织的复合致因,可为交通事故预防与差异化治理提供理论与方法参考。
Based on the road traffic accident data of City M in western China from 2023—2024,a systematic variable system was constructed around the five causal dimensions of"human-vehicle-road-environment-management".The number of accident participants was introduced as a proxy indicator for causal complexity,and the Apriori algorithm was optimized using the TF-IDF weighting mechanism.Through the mining of the compound causal patterns of casualty accidents,the ability to identify low-frequency and high-value causes was improved.The research results showed that:vehicle violations(failure to yield as required,running red lights)and pedestrian illegal crossing were the core factors leading to casualty accidents;low visibility and insufficient efficiency of traffic facilities exacerbated accident risks;the accident pattern of"scraping and hitting pedestrians"occurred frequently in personal injury cases;fatal accidents presented compound causes involving the interweaving of multiple factors including human,vehicle,road,environment and management.The research results can provide theoretical and methodological references for traffic accident prevention and differentiated governance.
作者
熊杰
李慧
XIONG Jie;LI Hui(School of Automotive and Transportation Engineering,Xihua University,Chengdu 611730,Sichuan,China;Sichuan Xihua Transportation Judicial Appraisal Center,Chengdu 611732,Sichuan,China)
出处
《农业装备与车辆工程》
2025年第10期113-119,共7页
Agricultural Equipment & Vehicle Engineering
