Research on Demand Forecasting of Retail Supply Chain Emergency Logistics Based on NRS-GA-SVM
作者单位：School of Computer and Information Engineering Beijing Technology and Business University
会议名称：《第30届中国控制与决策会议》
会议日期：2018年
学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)]
基 金：supported by Beijing Natural Science Foundation(No.9162002,9102005) the Humanities and Social Sciences Foundation Project of Ministry of Education of China(No.09YJA630003)
关 键 词：Retail Supply Chain Emergency Logistics Demand Dynamic Forecasting Model Machine Learning NRS-GA-SVM Algorithm
摘 要：Aiming at the characteristics of high dimension, dynamic condition and parameter self-adaptation for demand forecasting, the demand forecasting model of retail supply chain emergency logistics was created based on NRS-GASVM algorithm. The sample attribute index reduction model for emergency logistics demand forecasting was established based on NRS algorithm. The continuous data processing method was adopted. The key influencing factors were extracted more accurately. The dynamic demand forecasting model of emergency logistics was established based on nonlinear support vector machine regression theory and parameter optimization machine learning algorithm in order to get the optimal prediction effect. The numerical experiment results show that preprocessing the indexes with NRS and optimizing parameters with GA can not only improve the accuracy of the emergency logistics demand forecasting results,can but also reduce the execution time of the forecasting model, which promotes the emergency safeguard capability of retail supply chain emergency and verifies the feasibility of emergency logistics demand forecasting model for retail supply chain.