论文部分内容阅读
应用差分自回归移动平均模型(ARIMA)和最小二乘支持向量机(LS-SVM)的组合模型,对某航空公司的月度事故征候万时率进行了预测分析。对2008—2016年某航空公司的事故征候、飞行小时、航空器数量等历史数据建立ARIMA模型,应用SPSS软件进行模型拟合,获得事故征候万时率的线性部分;随后利用LS-SVM分析ARIMA模型的残差,获取非线性部分,最终通过二者之和获得ARIMA+LSSVM组合模型。对2017年1—3月的月度事故征候万时率进行了预测,并用实际数据验证。结果表明:ARIMA(1,1,1)(1,1,1)12模型较好地拟合了事故征候万时率的历史序列,LS-SVM模型对残差的拟合获得了较好的精度;组合模型的短期(3个月)预测值与航空公司事故征候万时率的趋势完全一致,且预测精确度可接受。
Using the combined model of ARIMA and LS-SVM, the forecasting rate of monthly incidents of an airline is predicted. ARIMA model was established based on historical data of an airline’s incidents, flight hours and number of aircraft from 2008 to 2016. SPSS software was used to fit the model and a linear part of the hourly rate of incidents was obtained. Then the ARIMA model was analyzed by LS-SVM Of the residual, access to non-linear part of the final sum of the two obtained ARIMA + LSSVM combination model. The monthly incident rate of January-March 2017 was estimated and verified with actual data. The results show that the ARIMA (1,1,1) (1,1,1) 12 model well matches the historical sequence of the incidents hourly rate, and the LS-SVM model has good fitting to the residuals Accuracy. The short-term (3-month) predicted value of the combined model is exactly the same as that of the airline incident, and the prediction accuracy is acceptable.