融合突变点校正的PELT-GM-SARIMA公路货运周转量组合预测模型

打开文本图片集

中图分类号：U491.1 文献标志码：A 文章编号：1002-4026（2025）05-0093-11

Abstract：Toaddressthelimitedaccuracyof single-modelforecastingandchallenges faced bycombinedmodels in handlingabnormal datafluctuations，this studyproposes anovel forecasting method integrating mutation pointcorectin intoapruned exactlinear time（PELT）-greypredictionmodel（GM）-seasonal autoregressve integrated moving average （SARIM）combined model.This method initiallemploysthePELTalgorithmtodetectfluctuationsin freightturnoverdata andidentifychangepoints.The GreyGM（1，1）modelis thenused tocorect anomaliesatthese change points，enabling thedataset tobeter meet the stationarityandrandomnessrequirementsforthe SARIMA model.Finaly，basedonthe optimized dataset，the SARIMA model is usedtoperform predictions ontherefined data.Using freightturnoverdata from Beijingasacase study，comparativeanalysisof diferenthybridmodels reveals that theproposed model exhibitssuperior performance thanother combined models，with significant reductions in mean squared error and mean absolute error and a coeficientof determination close to1.The PELT-GM-SARIMA model is structuralysimpleandcan better adapt to timeseries data with missing values or frequent anomalies，resulting in more accurate predictions.This study presents a more effective approach for traffc predictions in highway transportation planning and investment decision making.

Key words：transportation economics；PELT-GM-SARIMA model；freight turnover；change-point correction; transportation planning

公路货运周转量是指在一定时期内运输工具实际运送的货物数量与其相应运输距离的乘积之和，能够全面反映运输生产成果。（剩余10465字）