基于GM(1,1) Markov的危化品道路运输事故与交通事故预测及关系研究

为准确预测我国危化品道路运输及交通2类事故数量趋势，探究其内在联系，在单一的灰色GM（1,1）模型基础上与马尔科夫过程组合形成灰色GM（1,1）—马尔科夫预测模型，以2013—2017年2类事故数量的原始序列探讨了该组合预测模型的实际应用，采取平均相对误差、均方差比值、小误差概率对模型进行精度检验。研究结果表明:在组合预测模型较优情况的研究中，2类事故数量历年来波动性相似，因危险化学品自身的性质、包装和装卸使得2类事故量变化频率存在偏差；2018—2019年的危化品道路运输事故分别为485起和480起，交通事故分别为225 294起和234 454起。

Study on prediction and relationship of road transportation accidents of dangerous chemicals and traffic accidents based on GM(1,1)-Markov model

In order to predict the trend of the numbers of road transportation accidents of dangerous chemicals and traffic accidents in our country accurately, and explore their intrinsic connection, a grey GM(1,1)-Markov prediction model was established based on the single grey GM(1,1) model combining with the Markov process. The actual application of this combined prediction model was discussed by using the original sequence about the numbers of two types of accidents from 2013 to 2017, and the accuracy of this model was evaluated by the average relative error, mean variance ratio and probability of small error. The results showed that in the research on the better situation of the combined prediction model, the volatility of the numbers of two types of accidents over the years was similar, and the variation frequency of the numbers of two types of accidents had the deviation due to the own characteristics, packaging and handling of dangerous chemicals. The number of road transportation accidents of dangerous chemicals and traffic accidents in 2018 and 2019 was 485 and 480, 225294 and 234454, respectively.