考虑边信息的多层贝叶斯需求预测模型

被引:4
作者
邱萍萍 [1 ]
黄晓宇 [1 ]
曾青松 [2 ]
机构
[1] 华南理工大学经济与贸易学院
[2] 广州番禺职业技术学院信息工程学院
基金
国家重点研发计划;
关键词
供应链; 需求预测; 边信息; 非参数模型; 贝叶斯推断;
D O I
10.13196/j.cims.2020.01.020
中图分类号
O224 [最优化的数学理论]; F274 [企业供销管理];
学科分类号
070105 ; 1201 ;
摘要
随着工业互联网经济的发展,需求的不确定性日益增大,为提高需求预测的准确性,提出一个考虑边信息的多层贝叶斯需求预测模型(DFSI)。DFSI模型通过构造隐层的网络结构以实现对客户需求更加精确的刻画,该隐层结构主要包含两组参数:一组用于描述客户需求在时间上固有的连续性特征,另一组则用于融合相关的边信息特征。进一步,以贝叶斯推断为理论基础,以最大化后验概率为目标,推导出了DFSI的优化目标,并基于梯度下降方法设计了相应的求解算法。使用京东商城及某制造企业的真实销售数据对提出的模型进行了检验。结果显示,与常用的需求预测模型相比,DFSI能获得更好的预测结果。
引用
收藏
页码:191 / 201
页数:11
相关论文
共 15 条
[1]   智慧制造云中供应链管理的计划调度技术综述 [J].
肖莹莹 ;
李伯虎 ;
侯宝存 ;
施国强 ;
林廷宇 ;
杨晨 .
计算机集成制造系统, 2016, 22 (07) :1619-1635
[2]  
SSpace: A Toolbox for State Space Modeling[J] . Marco A. Villegas,Diego J. Pedregal.Journal of Statistical Software . 2018 (1)
[3]  
A greedy aggregation–decomposition method for intermittent demand forecasting in fashion retailing[J] . Chongshou Li,Andrew Lim.European Journal of Operational Research . 2018 (3)
[4]  
Multi-step-ahead time series prediction using multiple-output support vector regression[J] . Yukun Bao,Tao Xiong,Zhongyi Hu.Neurocomputing . 2013
[5]   Combining seasonal ARIMA models with computational intelligence techniques for time series forecasting [J].
Khashei, Mehdi ;
Bijari, Mehdi ;
Hejazi, Seyed Reza .
SOFT COMPUTING, 2012, 16 (06) :1091-1105
[6]   The Impact of External Demand Information on Parallel Supply Chains with Interacting Demand [J].
Zhang, Xiaolong ;
Zhao, Yao .
PRODUCTION AND OPERATIONS MANAGEMENT, 2010, 19 (04) :463-479
[7]  
Forecast combinations of computational intelligence and linear models for the NN5 time series forecasting competition[J] . Robert R. Andrawis,Amir F. Atiya,Hisham El-Shishiny.International Journal of Forecasting . 2010 (3)
[8]  
A decision support framework for global supply chain modelling: an assessment of the impact of demand, supply and lead-time uncertainties on performance[J] . Yavuz Acar,Sukran Kadipasaoglu,Peter Schipperijn.International Journal of Production Research . 2010 (11)
[9]  
Gaussian Processes for Machine Learning (GPML) Toolbox[J] . Carl Edward Rasmussen,Hannes Nickisch.Journal of Machine Learning Research . 2010
[10]   Stochastic models underlying Croston's method for intermittent demand forecasting [J].
Shenstone, L ;
Hyndman, RJ .
JOURNAL OF FORECASTING, 2005, 24 (06) :389-402