基于ARIMA-LOESS模型的危化品道路运输事故起数预测

文章针对危化品道路运输事故预测问题,运用差分自回归移动平均模型(Autoregressive Integrated Moving Average,ARIMA)与局部加权回归模型(Locally Estimated Scatterplot Smoothing,LOESS)的组合模型,对我国危化品道路运输事故发生起数进行预测。首先,基于2011—2018年我国发生的危化品道路运输事故数据建立ARIMA模型,利用SPSS软件进行模型拟合预测,获取危化品道路运输事故起数的线性部分;其次,应用MATLAB建立LOESS回归模型,对ARIMA模型预测偏差进行残差优化,获取危化品道路运输事故起数的非线性部分;最后,建立ARIMA-LOESS组合模型,利用组合模型对危化品道路运输事故发生起数进行预测,并根据真实数据对预测结果进行对比验证。结果表明:ARIMA-LOESS组合预测模型可较好拟合危化品道路运输事故数据序列,并修正单一模型的误差,获取较高的预测精度。该研究可为危化品道路运输安全与运行的趋势分析与判断提供更加可靠的数据依据,也可为危化品道路运输事故防控方案提供帮助。

Road Transport Accident Prediction of Hazardous Chemicals Based on ARIMA-LOESS Model

Aiming at the prediction problem of hazardous chemicals road transportation accidents,this paper applied the combined model of differential Autoregressive Integrated Moving Average(ARIMA)and Locally Estimated Scatterplot Smoothing(LOESS)to predict the number of hazardous chemicals road transportation accidents in China.First,based on the number of hazardous chemicals road transport accidents in China over the years 2011-2018,an ARIMA model was established,and SPSS software was used to fit the model to obtain the linear part of the number of hazardous chemical road transport accidents.Second,MATLAB was used to establish LOESS regression model to optimize residuals of ARIMA model prediction errors,and to obtain the non-linear part of the number of hazardous chemicals road transportation accidents.Finally,a ARIMA-LOESS combination model was established and used to perform the number of hazardous chemicals road transportation accidents and the prediction was verified with actual data.The results showed that the combined model fitted the historical sequence of the number of hazardous chemicals road transport accidents well,corrected the error of the single model,and obtained high prediction accuracy.This research can provide more reliable data basis for the analysis and judgment of the trend and safety of road transportation safety and operation of hazardous chemicals,and can also help the prevention and control plan for road transportation accidents of hazardous chemicals.