基于灰色马尔可夫链的铁路事故死亡人数预测模型  

Forecasting the death toll of railway accidents based on grey Markov chain

作　　者：房新 谈敏佳 吴啸宇 何君 FANG Xin;TAN Minjia;WU Xiaoyu;HE Jun(School of Automation,Nanjing University of Science and Technology,Nanjing 210094,China)

机构地区：[1]南京理工大学自动化学院,南京210094

出　　处：《交通科技与经济》2022年第4期37-42,共6页Technology & Economy in Areas of Communications

基　　金：江苏省“六大人才高峰”高层次人才项目(JXQC-021);河南省重点科技攻关项目(182102310004);教育部人文社科科学研究项目(18YJAZH028)。

摘　　要：为反映铁路运行的安全状况和未来趋势,并考虑到铁路事故发生的非线性、动态性和随机性,使用灰色马尔可夫模型对铁路事故死亡人数进行预测。采用2010—2020年全国铁路事故死亡人数数据进行建模和验证,对2021—2023年铁路事故死亡人数进行预测,判断未来3年内的铁路运行安全状况和趋势。应用结果表明:灰色马尔可夫模型可以提高灰色GM(1,1)模型的预测精度,经过修正后,2019年的预测值精度提高22%,2020年的预测值精度提高50%。预测2021—2023年中国铁路事故死亡人数下降率将维持在8%左右,铁路运行状态良好,但可通过扩大监管措施和加大惩治力度等行为提升安全运营状况。In order to reflect the safety status and future trend of railway operation,and considering the nonlinearity,dynamics and randomness of railway accidents,the grey Markov model is used to predict the number of railway accident deaths.Using the national railway accident death toll data from 2010 to 2020 for modeling and verification,predict the railway accident death toll from 2021 to 2023,and judge the railway operation safety status and trend in the next three years.The application results show that the grey Markov model can improve the prediction accuracy of the grey GM(1,1)model.After correction,the prediction accuracy in 2019 is increased by 22%,and the prediction accuracy in 2020 is increased by 50%.From 2021 to 2023,the decline rate of China's railway accident fatalities will remain at about 8%.The railway operation is in good condition,but the safe operation can be improved by expanding regulatory measures and increasing punishment.

关 键 词：铁路事故死亡人数 灰色理论 灰色马尔可夫模型 短期预测 安全措施 

分 类 号：U491.2[交通运输工程—交通运输规划与管理]