摘要
针对当前的快递行业,探究运输过程中快递延迟的原因,建立快递延迟预测的指标体系;利用大数据技术采集、存储、汇总相应的货运数据,设计基于大数据技术的快递延迟预测模型;对指标体系所对应的指标值进行量化和归一化的处理,形成相应的训练集,并采用GA(遗传算法)优化BP人工神经网络,进行算法设计;建立基于GA的应用模型,进行仿真训练,确定最优隐层数,同时对比多种预测方法,分析不同方法的预测准确率,并对仿真方法和结果进行分析,结果表明,该模型能对快递延时率进行有效预测,且阐述了该模型在物流企业决策与管理系统中的应用方法和应用前景。
For the current express industry,explore the reasons for express delay forecast in the process of transportation,and establish an index system of express delay rate;Collect,store and summarize the corresponding freight data by using big data technology,and design the express delay rate prediction model based on big data technology;The index values corresponding to the index system are quantified and normalized to form the corresponding training set,and the BP artificial neural network is optimized by GA(Genetic Algorithm);The application model based on GA is established for simulation training to determine the optimal number of hidden layers. At the same time,a variety of prediction methods are compared to analyze the prediction accuracy of different methods. The simulation methods and results show that it can effectively predict the express delay rate. The application method and Prospect of the model in logistics enterprise decision and management system are described.
作者
叶斌
刘杰平
徐金亚
YE Bin;LIU Jieping;XU Jingya(Chengdu Neusoft University,Chengdu 611844,China)
出处
《电子设计工程》
2022年第16期21-26,共6页
Electronic Design Engineering
基金
“西华师范大学——四川省教育厅研究基地:成渝地区教育与经济社会协同发展研究中心”资助项目(CYJXF22032)
四川省高等学校人文社会科学重点研究基地·四川省教育信息化应用与发展研究中心项目(JYXX20-015)
教育部产学合作协同育人项目(201902146043)。
