%0 Journal Article
%T 基于深度学习方法下的产品订单的数据分析与需求预测<br>Data Analysis and Demand Forecasting for Product Orders Based on Deep Learning Methods
%A 巢逸
%J Statistics and Applications
%P 722-729
%@ 2325-226X
%D 2024
%I Hans Publishing
%R 10.12677/sa.2024.133072
%X 近年来企业外部环境变化不确定，企业供应链面临着更多的挑战。因此，需求预测作为企业供应链中的第一道防线，能够帮助企业更好地制定采购计划和生产计划，减少业务波动对企业的影响。针对销售区域、销售时段与订单需求量的关系分析，本文采取了Prophet模型。该模型能够捕捉到销售区域和销售时段对订单需求量的影响，并对现有的和未来的订单需求量进行准确的预测。针对订单需求量的整体影响，本文采用了GNN模型，考虑了销售区域编码、产品编码、产品大类编码、产品细类编码、销售渠道名称、产品价格等多种因素。该模型能够对订单需求量进行全面准确的预测，分析出影响因素占比大小关系，为企业提供决策支持。综上所述，本文的研究结果能够帮助企业更准确地预测未来的需求量，制定更好的采购计划和生产计划，从而提高企业的运营效率和竞争力。<br />
In recent years, the uncertainty of changes in the external environment has posed greater challenges to corporate supply chains. Therefore, demand forecasting, as the first line of defense in a company’s supply chain, can help enterprises better formulate procurement and production plans, thereby reducing the impact of business fluctuations. This paper employs the Prophet model to analyze the relationship between sales regions, sales periods, and order demand. The model captures the influence of sales regions and periods on order demand and provides accurate forecasts for both current and future order demand. For the overall impact on order demand, this paper utilizes the GNN model, taking into account various factors such as sales region codes, product codes, major product categories, sub-product categories, sales channels, and product prices. This model enables a comprehensive and accurate prediction of order demand, analyzing the relative significance of influencing factors to support corporate decision-making. In conclusion, the research findings of this paper can help enterprises more accurately forecast future demand, develop better procurement and production plans, and thereby improve operational efficiency and competitiveness.
%K 需求预测，Prophet模型，GNN模型，经营策略<br>Demand Forecasting
%K Prophet Model
%K GNN Model
%K Business Strategy
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=89912