Genetic algorithms for parameters estimation applied in inventory management problem

Abstract

Inventory management is one important instrument for the efficiency of modern organizations. In such models, the necessity of parameters estimate is usual, which can be related to some uncertainties. For this purpose, algorithms of approximation are commonly used, once this kind of problem deals with a big solution domain that can make unfeasible in most of the cases. Thus, an interesting alternative to solve problems of this nature is the application of Genetic Algorithm (GA) and Neural Network Techniques (ANN). In this work, it is presented a method based on the Genetic Algorithms, which is an optimization technique inspired by genetics and natural evolution, in the parameters estimation, in order to maximize the inventory management. The GA technique was implemented and tested successfully, and the results obtained are in a good agreement with the results obtained by exhaustive searching algorithm that justifies the application of the GA method in problems of this nature.