基于灰色-马尔可夫模型的道路交通事故预测

Prediction of Road Traffic Accidents Based on the Grey-Markov Model

根据攀枝花市2010~2017年的交通事故统计数据,分别利用灰色系统理论和灰色马尔可夫链预测法构建攀枝花市交通事故的预测模型,再用两种预测模型分别对2018~2019年攀枝花市发生的交通事故数、死亡人数、受伤人数及财产损失数进行预测,然后将GM(1,1)预测模型与灰色马尔可夫链预测模型的预测结果进行对比分析。实例结果表明,灰色马尔可夫预测模型能有效克服随机波动性数据对道路交通事故四项绝对指标预测精度的影响,预测精度较好。

According to the traffic accident statistics of Panzhihua City from 2010 to 2017,using grey system theory and Grey-Markov chain prediction method respectively,to build a prediction model of traffic accidents.Then two prediction models are used to predict the number of traffic accidents,deaths,injuries and property losses in Panzhihua City from 2018 to 2019.Then,the prediction results of GM(1,1)prediction model and Grey-Markov chain prediction model are compared and analyzed.The example results show that Grey-Markov prediction model can effectively overcome the influence of random fluctuation data on the prediction accuracy of four absolute indexes of road traffic accidents.The prediction accuracy of Grey-Markov prediction model is better than the prediction accuracy of grey prediction model.