小批量物料需求周预测的研究

Research on Weekly Forecast of Small Batch Material Demand

在多品种小批量的物料生产中,企业一般事先无法知道物料的实际需求量,而准确的物料需求量预测对于提高企业生产效率具有重要的意义。首先建立了基于需求频数、需求数量、需求趋势和销售单价的物料重要性因子模型,从物料的重要性来筛选企业需要重点关注的物料;随后引入基于需求频数、需求趋势和销售单价的随机森林回归模型和前馈神经网络模型来预测物料的需求量,并通过均方差等指标来评估模型。实验验证表明,该预测模型能为企业小批量物料需求提供一个有参考意义的解决方案。

In the production of multi variety and small batch materials,enterprises generally cannot know the actual demand of materials in advance,and accurate prediction of material demand is of great significance for improving the production efficiency of enterprises.Firstly,the paper establishes a material importance factor model based on demand frequency,demand quantity,demand trend and sales unit price to screen the materials that enterprises need to focus on from the importance of materials;Then the stochastic forest regression model and feedforward neural network model based on demand frequency,demand trend and unit price are introduced to predict the demand of materials,and the model is evaluated by means of indicators such as mean square error.The experimental results show that the prediction model can provide a reference solution for small batch material demand of enterprises.