SUMMARY
The determination of the economic order quantity is important for the rational realization of the logistics process of transport, manipulation and storage in the supply chain. In this paper an expert model for the determination of the economic order quantity has been developed by means of a hybrid method of the artificial intelligence Adaptive neuro-fuzzy inference systems - ANFIS. It has been used for modeling a complex logistics process in which it is difficult to determine interdependence of the presented variables applying classical methods. The hybrid method is applied in order to take advantages of the individual methods of artificial intelligence: fuzzy logic and neural networks. Experts’ experience and information about the company operations for a certain group of items are used to form the model. An analysis of the validity of the model results was performed on the basis of the average relative error. It is shown that the model imitates the work of the expert in the observed company with great accuracy. A sensitivity analysis is also applied which indicates that the model gives valid results. The proposed model is flexible and can be applied to various types of goods in the supply chain management.