عنوان مقاله [English]
One of the important techniques in the field of inventory management is inventory classification. The classification method based on ABC analysis is a well-known method which sets the items in a different class, according to their importance and values. In this paper, a three-level supply chain network for a bi-objective optimization model is proposed to improve the quality of inventory grouping based on the ABC technique. The first objective maximizes the net profit of the items in the central stock (level 2) and the second objective function maximizes the net profit of items in different wards (level 3) in uncertain conditions. The proposed model simultaneously optimizes the service level, the number of inventory groups, and the assigned items to each group based on the set of constraints, inventory shortages, ordering cost, and purchasing cost. To solve this model, the heuristic and exact methods such as LP-metric and modified ε-constraint are applied. Then for comparing those methods, the statistical hypothesis test is used. In conjunction with, the AHP technique is exerted to choose the most efficient solving method. The results show that ε-constraint has the best performance. Ultimately, the proposed model has been implemented in one set of numerical examples to show its applicability in practice.