A New Model for Multi-criteria ABC Inventory Classification: PROAFTN Method

ABC analysis is one of the most widely used techniques in inventory management to classify items into three predefined and ordered categories: A (very important items), B (moderately important items) and C (relatively unimportant items). In the literature, most of existing classification models tackled the ABC inventory classification problem as a ranking problem, i.e. a set of inventory items are ranked in a deceasing order based on their performance expressed by an overall weighted score. In this paper, the ABC inventory classification problem will be tackled as an assignment problem, i.e. an inventory item will be classified to the category with which it has the most similar characteristics. For this purpose, the PROAFTN method will be used to classify each inventory item into a specific category. Since the application of PROAFTN method requires the knowledge of some parameter values (e.g. prototypes pessimistic intervals and discrimination thresholds), the Chebyshev's theorem is used for their estimation. To determine the performance of the PROAFTN model with respect to some existing models, a benchmark data set from the literature is used.