摘要
通过对物料的历史销售数据使用统计分析和主成分、核主成分分析以及画时间序列图的方法得到6种最重要的物料。使用ARIMA模型对6种物料分别构建时间序列预测模型,模型结果均较好,R2均在0.65以上,最高达到0.91,显著性水平均小于等于0.002。根据预测结果建立了3个生产计划模型,发现第三个模型平均生产计划数、平均实际需求数基本不变;平均库存量大大降低;客户满意度达到85%的标准。采用第三个模型为物料制定生产计划,将能有效降低企业的库存管理费用,节约生产成本,并达到较高客户满意度。
Through the statistical analysis of historical sales data of materials and analysis of principal component and kernel principal component with method of drawing time series diagram,six kinds of most important materials are obtained.Using ARIMA model to build time series prediction models for six kinds of materials,the model results are good.R2 is more than 0.65,the highest is 0.91,and the significance level is less than or equal to According to the forecast results,three production planning models are established,and it is found that the average number of production plans and the average actual demand of the third model are basically unchanged;the average stock is greatly reduced;and customer satisfaction reaches the standard of 85%.Using the third model to make the production plan for the material will effectively reduce the stock management cost of the enterprise,save the production cost and achieve higher customer satisfaction.
作者
陈伟
陈丹
CHEN Wei;CHEN Dan(Changjiang Institute of Technology,Wuhan 430212;Graduate School of Baguio University,Baguio 26000,Philippines)
出处
《湖北工业职业技术学院学报》
2023年第3期75-80,共6页
Journal of Hubei Industrial Polytechnic
关键词
动态规划
ARIMA预测
安全库存
主成分分析
物料生产计划模型
dynamic programming
ARIMA forecasting
safety stock
principal component analysis
material production planning model
