Comparative Analysis of MultiCriteria Inventory Classification Models for ABC Analysis

Abstract

ABC analysis is a commonly used inventory classification technique which consists in splitting a large number of inventory items into three categories, A, B and C: category A consists in the most important items, category B consists in the moderately important items and category C consists in the least important ones. Through this classification, inventory items are managed in an efficient way. In this paper, we argue the benefits of cross-fertilization of both Artificial Intelligence (AI) and MultiCriteria Decision Making (MCDM) techniques to carry out the ABC classification of inventory items. For this purpose, we develop some new hybrid inventory classification models based on metaheuristics (AI techniques) to generate the criteria weights and on Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method (MCDM technique) to compute the overall weighted score of each item on which the ABC classification is performed. To evaluate the effectiveness of the proposed classification models with respect to some classification models from the literature, a comparative study — based on a service-cost analysis and three real datasets — is conducted. The computational analysis demonstrates that our proposed hybrid models are competitive and produce satisfactory results. The results have also shown, that our proposed models outperform some existing models from the literature.