652 Experimental Analysis of the Applicability of a Multi-Item Constrained EOQ Calculation Algorithm S. Miranda, M. Fera, R. Iannone and S. Riemma Department of Industrial Engineering University of Salerno Fisciano (SA) 84084, Italy Abstract The present paper proposes the results of an experimental campaign targeted to investigate the applicability in real contexts of a new approach for the calculation of the Economic Order Quantity (EOQ) when a multitude of item have to be stored in a limited space. The approach, already presented in the literature in its basic principles, has been modified in order to make it more suitable for real needs. Subsequently, the so obtained algorithm has been tested, in simulated scenarios, with the purpose of identifying the regulation parameters with significant influence on the final results, their best values and the average level of performances the approach is able to guarantee, in terms of total cost of stock and mean saturation of the warehouse. The outcomes of this study confirm the general validity of the approach even if some more deepen investigations are still necessary before a possible implementation in a real stock management application. Keywords Inventory Management, Constrained EOQ, Dynamic calculation 1. Introduction and Literary Review The problem of the Economic Order Quantity (EOQ) calculation under constraint is one of the basic problems in the classical theory of the inventory management. Typical constraints are a limited availability of space in the warehouse or a maximum budget for the purchases. Among the possible approaches to the solution of the constrained EOQ calculation, the most known and widespread is surely given by the application of the Lagrange multipliers method. This because it can assure easiness of implementation with a stationary approach targeted to avoid that all products, when eventually reach a simultaneous peak of stock, do not exceed the space constraint (Hadley and Whitin 1963, Parsons 1966, Phillips et al. 1973). Another classical approach is named “fixed-cycle method”. In this procedure a fixed cycle time for all items is assumed and the orders for the items have to be phased within the cycle. In this way situations where peaks of stock are reached simultaneously are avoided. The main problem of this approach is to decide on the joint ordering cycle and the phasing within the cycle, so as to minimize total cost while satisfying the capacity constraint throughout the cycle. (Parsons 1966, Goyal 1976, Zoller 1977). The two classical methods have been compared by Rosenblatt (1981), evidencing better results but a more difficult implementation of the fixed-cycle approach. A third approach to the problem, called “basic cycle method”, has been proposed by Goyal and Belton (1979), Kaspi and Rosenblatt (1983) and Roundy(1985) and consists in the calculation of individual cycle times which have to be integer multipliers of a basic cycle time. Many solution procedures and applications were developed through this approach also for the unconstrained problem, with economic advantages due to the joint replenishment. Rosenblatt (1985) provided a comparative study of the three classical methodologies. In this study the basic cycle approach has resulted computationally very efficient as compared with the fixed cycle approach but no general conclusions were furnished about the quality of the solutions, being them usually data dependent. For this reason the author suggested to develop some other more sophisticated heuristic algorithms to the basic cycle approach in order to conciliate best efficiency with best solutions to the problem. The main limit of the three above described methodologies is that generate stationary ordering policies which are not very suitable to manage a strongly dynamic resource as the warehouse is. In fact, due to the strict connection of the Proceedings of the 2014 International Conference on Industrial Engineering and Operations Management Bali, Indonesia, January 7 – 9, 2014