基于组合预测模型的快递需求预测研究

A Research on Express Demand Forecast Based on Combination Forecasting Model

考虑到快递需求受电子商务环境和季节性特征影响,文章将电子商务因素纳入快递需求预测指标体系中,采用灰色关联分析法确定关联度显著指标,采用GM(1,N)模型进行趋势预测,应用时间序列ARIMA模型预测季节性差异,并建立上述两类模型的GM(1,N)ARIMA权重组合模型以提升预测精度,运用组合预测模型对杭州市2021年4季度—2022年4季度的快递需求进行预测。研究结果表明,组合预测模型的预测误差显著优于所选单一模型;电子商务的发展对快递需求的影响将会逐渐增加;快递需求的季节性差异出现新特征。

Considering that the express demand is affected by the e-commerce environment and seasonal characteristics,this paper brings the e-commerce factors into the express demand prediction index system,using the grey correlation analysis method to determine the significant correlation index,the GM(1,N)model to predict the trend and the time series ARIMA model to predict the seasonal difference,and establishing the GM(1,N)of the above two types of models-Arima weight combination model to improve the prediction accuracy.The first mock exam of Hangzhou's express demand in the 4th quarter of 2021 and the 4th quarter of 2022 is carried out by using the combined forecasting model.The results show that the prediction error of the combined forecasting model is significantly better than that of the selected single model,the impact of the development of e-commerce on the demand for express delivery will gradually increase and the seasonal difference of express demand will appear some new characteristics.