Economic Order Quantity (EOQ) Measurement Using Intelligent Systems Techniques

Abstract

Progressively, Intelligent Systems (IS) techniques have been used to reduce the uncertainty in complex problems. Because Intelligent Systems can be taught and predict the results, operations’ knowledge can be implemented into the systems. This Thesis is related to the application of Intelligent Systems to the solution of real-world problems, and it intends to estimate the Economic Order Quantity (EOQ) by using Intelligent Systems techniques. EOQ is the maximum number of inventories to be ordered at the time to minimize the company’s total costs of ordering, carrying, shortage and holding costs. There are some mathematical methods to calculate EOQ, but there are some disadvantages in using the formulas. Firstly, formulas cannot deal with vague and uncertain data, so if the inputs are not crisp or in linguistic terms, the formulas are not practical. Secondly, in some cases, due to some limitations (such as unpredictable markets, freshness of inventory in food and medicine industry), EOQ cannot be calculated by the formulas. In this situation, Intelligent Systems can help to determine the optimum EOQ. Using formulas, as done by conventional approaches, to calculate the optimum EOQ has some disadvantages; first, in some cases company’s demands/costs/expiry dates vary time to time and are not constant, so there is no constant number to include in the formulas. Moreover, in the case of vague or uncertain data, the formulas are not helpful and practical. Secondly, when EOQ is affected by other factors (such as freshness of inventory in perishable products, fragile/unpredictable markets), there is not a specific formula; therefore, using the formulas is not an option. Third, the conventional EOQ formula does not take into account the inflation rate. In this Thesis in order to reduce the uncertainty and enhance the efficiency, using Fuzzy Inference Systems for EOQ calculations is proposed. In this approach, the EOQ can be calculated in situations with vague data, uncertainty, no formulas, varied inputs, and fusion methods. Two methods of Fuzzy Inference Systems are used in this proposal in order to calculate the optimum EOQ: measuring EOQ by applying Mamdani Fuzzy Inference Systems; and measuring EOQ by applying Sugeno Fuzzy Inference Systems. The proposed nonconventional methodologies for EOQ are based on Intelligent Systems Techniques such as Fuzzy Inference Systems (FIS). The considered FIS has some advantages in EOQ calculations compared to EOQ conventional formulas. These advantages can be listed as: • They can deal with vague and uncertain data. • They can solve problems with inexplicit and incomplete data. • They can cover wide ranges of operating conditions. • They can be customized in linguistic terms. • They are user friendly, simpler and more flexible. This Thesis illustrates the methodologies in detail. The performance of these methods is examined, and the results are analyzed.