基于GM(1,1)模型的军用车辆交通事故预测研究  

Forecast of Military Vehicle Traffic Accidents Based on GM(1,1) Model

作　　者：宋传平[1] 王鹏飞[1] 李涛[1] 

机构地区：[1]解放军汽车管理学院车辆管理系

出　　处：《中国安全科学学报》2010年第11期52-55,共4页China Safety Science Journal

摘　　要：在分析军用车辆交通事故预测技术重要性的基础上,根据2005—2009年的军用车辆交通事故原始数据,建立军车交通事故的GM(1,1)灰色预测模型。对军用车辆交通事故起数和死亡人数进行短期预测,并通过实例验证预测模型的适用性。通过模型验证结果可以看出:GM(1,1)模型预测所得数据较为精确。预测结论如下:未来几年,军车交通事故会稳步下降,事故发生次数会快速减少,但死亡人数相对减少较缓。The importance of forecasting technique for military vehicles traffic accidents is analyzed.Based on the primordial data of military vehicle accidents from 2005 to 2009,the GM（1,1） prediction model for military vehicles traffic accidents is first constructed.Then,the number of military vehicles traffic accidents and the death toll in the near future are forecasted with this model,and the applicability of the forecast model is validated.It shows that the GM（1,1）model can accurately forecast the number and the death toll of military vehicle accidents.And in the future years,military vehicle accidents take on the following characteristics：the number of military vehicles accidents will drop steadily,the accident frequency will drop fast,but the death toll will decrease relatively slowly.

关 键 词：军用车辆交通事故 灰色系统 GM(1 1)模型 预测 精度检验 

分 类 号：X928.03[环境科学与工程—安全科学] O159[理学—数学]