基于大数据分析的协议库存需求预测研究

Research on Forecasting Stock Demand Based on Big Data Analysis

摘要: 在物资集约化和供应链管理智能化的背景下，本文基于协议库存需求分析应用系统工程理论和现代需求链管理理论，构建了包括自上而下和自下而上两种预测模型，并根据预测模型建立适合电力系统采购需求特性的采购需求预测体系，并应用大量历史数据针对不同生命周期阶段和不同的需求类型来分析预测效果，分析过程中通过不断调整模型参数以提高预测精准度，在不断完善需求预测体系的过程中，固化自上而下预测模型，最终基于项目全程可视化平台，打造大计划体系。

Abstract: Under the background of material intensification and intelligent supply chain management, this paper builds a set of top-down and bottom-up prediction models based on protocol inventory de-mand analysis, application system engineering theory and modern demand chain management theory. Based on the predictive model, this paper establishes a procurement demand forecasting system suitable for the procurement demand characteristics of the power system, and applies a large amount of historical data to analyze the forecasting effect for different life cycle stages and different demand types. In the analysis process, this paper improves the prediction accuracy by continuously adjusting the model parameters, and in the process of continuously improving the demand forecasting system, solidifies the top-down forecasting model, and finally builds a large planning system based on the project’s full-process visualization platform.