高速公路月度交通量ARIMA预测模型  

ARIMA model of expressway traffic volume monthly forecasting

作　　者：芮少权 匡安乐[2] 

机构地区：[1]陕西西禹高速公路有限公司,陕西西安710016 [2]长安大学经济与管理学院,陕西西安710064

出　　处：《长安大学学报（自然科学版）》2010年第4期82-85,91,共5页Journal of Chang’an University(Natural Science Edition)

基　　金：国家自然科学基金项目(50808022)

摘　　要：为避免传统预测方法误差率较大的缺陷,在考虑高速公路月度交通量季节性周期特点的基础上,构建了ARIMA预测模型,并对ARIMA模型识别、模型检验和模型预测进行了系统分析,并应用于某高速公路进行月度交通量预测。应用结果表明:模型预测综合误差率为5.45%,低于灰色模型35.43%的误差率,低于三次指数平滑法的5.65%误差率;ARIMA预测模型能更好地适应于高速公路月度交通量预测。In order to avoid the defect of large error rate of traditional forecasting methods,taking into account the seasonal cycle characteristics of expressway month traffic volume,an ARIMA prediction model was constructed,and the ARIMA model identification,model checking and model predictions were analyzed.The model is applied to an existed expressway to predict its month traffic volume.The results show that： the comprehensive error rate is 5.45%,better than the gray model of 35.43%,lower than the triple exponential smoothing of 5.65%;the ARIMA forecasting model can better meet the forecast of expressway month traffic volume.3 tabs,4 figs,9 refs.

关 键 词：交通工程 高速公路 月度交通量 预测 ARIMA模型 

分 类 号：U491[交通运输工程—交通运输规划与管理]