基于赋权关联度算法的地铁拥挤踩踏事故风险研究

Research on Crowded Stamped Accident Risk at Subway Based on Empowering Related Degree Method

为研究地铁站内发生拥挤踩踏事故风险,对我国近年地铁拥挤踩踏事故进行统计分析。总结出引发地铁拥挤踩踏事故的主要因素为4方面,13个因素。地铁拥挤踩踏事故易发场所主要为自动扶梯、换乘楼梯、上下车门口处。利用数学建模思路,引入赋权关联度算法,选取地铁实例样本与典型事故案例进行比对,计算关联系数,结果表明,地铁实例样本在自动扶梯处发生拥挤踩踏事故危险程度最高,人群拥挤、跌倒碰撞、电梯故障和应急管理缺陷等因素最可能诱发拥挤踩踏事故。评估结果与地铁实例样本状况基本吻合。

In order to study the subway station crowded stampede accident risk,typical metro crowded stampede accidents happened in recent years in China were statistically analyzed.13 factors,causing the accident and belonging to 4 aspects,were identified.Then empowering related degree calculation method was used.The selected actual metro samples and typical accidents were compared.The correlation coefficients of the actual sample and typical cases were calculated.The results show that points where the accident is apt to occur are escalators,transfer stairs and the platform＇s areas near coach doors,escalators are the most dangerous points,and that crowd pushing-squeezing,fall down,collision,escalators failure,and emergency management deficiencies are most likely to induce the accident.The assessment result basically conforms with reality.