Sistemática de seleção de variáveis para classificação de produtos em categorias de modelos de reposição

Abstract

Companies integrated in supply chains seek initiatives to improve the overall performance of their chains. Vendor Managed Inventory (VMI) enables better results when it comes to managing and balancing stocks along the chain. For that matter, VMI frameworks must rely on well-defined parameters and algorithms aimed at allocating products to replenishment local characteristics. This paper presents a method to classify products in replenishment categories based on Principal Component Analysis (PCA) along with two classification algorithms: k-Nearest Neighbor (KNN) and Linear Discriminant Analysis (LDA). The model seeks to identify the most relevant variables for assigning products to the most appropriate replenishment model. When applied to a real situation, the proposed method yielded 90% classification accuracy, retaining 55% of the original variables on average.