Inventory Categorization Using Multiple Criteria Classification

Abstract

Inventory management holds paramount importance in modern business landscapes, where expert resource handling is essential for success. This study investigates the correlation between specific Stock Keeping Units (SKUs) and age categories of items, exploring factors influencing the speed of warehouse movement. Categorization based on product life cycle, pricing, and remaining stock is examined alongside savings levels. Employing multi-criteria classification algorithms, including traditional and machine learning techniques, the research illuminates inventory dynamics. This study compares bi-criteria, multi-criteria of traditional, and machine learning inventory classification methods, providing a comprehensive analysis of inventory categorization strategies. Machine Learning method achieved highest accuracy up to 99% by Decision Tree and 68% by Support Vector Machine. The accuracy score followed by the traditional method by using FSN-fuzzy method accuracy score up to 86.7%. The outcomes of the FSN analysis and fuzzy classification experiment will offer stakeholders valuable insights, potentially sparking innovative ideas for their business.