基于K-Means聚类算法的快消企业库存优化研究——以S公司为例

Research on Inventory Optimization of Fast-Moving Consumer Goods Enterprises Based on K-Means Clustering Algorithm—Taking Company S as an Example

摘要: 在快消行业，库存管理水平犹如企业运营的“晴雨表”，直接牵动着资金周转效率、市场响应速度与最终盈利能力。对于主营日化类快消品的企业而言，其供应链的多层级特征，从海外供应商、国内总仓到区域分仓，再到终端零售网点，进一步放大了库存管理的复杂性。因此，精准平衡供需、优化库存周转，成为日化快消企业提升竞争力的核心课题。本研究以S企业2024年全品类库存数据为基础，先通过传统ABC-XYZ分类法对产品进行初步分类，再选取月均销量、CV变异系数、ABC-XYZ分类构建三维特征空间，运用K-Means聚类算法对产品库存进行更精准地分类。以S企业为案例，根据研究结果，针对快消行业不同类别的产品制定差异化库存优化策略，旨在降低库存成本并减少缺货率，为快消企业的库存管理提供参考。

Abstract: In the fast-moving consumer goods (FMCG) industry, inventory management level is like a “barometer” of business operation, directly affecting the efficiency of capital turnover, market response speed and ultimate profitability. For FMCG enterprises mainly dealing in daily chemical products, the multi-level characteristics of their supply chain, from overseas suppliers, domestic central warehouses to regional distribution centers, and then to terminal retail outlets, further amplify the complexity of inventory management. Therefore, precisely balancing supply and demand and optimizing inventory turnover have become the core issues for FMCG enterprises to enhance their competitiveness. This study, based on the full-category inventory data of S enterprise in 2024, first classifies products through the traditional ABC-XYZ classification method, and then selects average monthly sales volume, CV coefficient of variation, and ABC-XYZ classification to construct a three-dimensional feature space. The K-Means clustering algorithm is then used to classify product inventories more accurately. Taking S enterprise as a case, based on the research results, differentiated inventory optimization strategies are formulated for different types of products in the FMCG industry, aiming to reduce inventory costs and decrease out-of-stock rates, providing a reference for inventory management in FMCG enterprises.