Abstract
The ABC classification represents one of the most frequently used analysis in production and inventory management domains. This analysis is applied to categorize a set of items in three predefined classes A, B and C, where each class follows a specific management and control policies, in order to generate companies financial well-being. This paper introduces a new approach for the multi-criteria inventory classification based on the hybridization of the Differential Evolution algorithm (DE) with the multi-criteria decision making method namely Electre III. The evolutionary algorithm (DE) attends to learn and optimize the Electre III input parameters (criteria weights). The Electre III method generates a ranking score for all the inventory items and an ABC distribution dispatches all these items into three ordered classes A, B, C, forming a complete classification. An inventory cost function is used thereafter to evaluate each established classification. This function is based on different inventory costs and service level measurement and also represents the objective function of our model, which consists of minimizing the inventory cost. The highlight of our proposed hybridization approach DE-Electre III is the exploitation of the robustness and efficiency of used techniques. Based on generated results, our model provided encouraging results in the ABC MCIC problem.