An Improved Tabu Search (ITS) Algorithm Based on Open Cover Theory for Global Extremums

Abstract

In this study, a new improved Tabu Search (ITS) algorithm with an open-cover approach is presented for solving global optimization of multimodal functions which have continuous or discrete variables. The method uses open sets covering the wide domain of possible solutions which are constructed by a specific metric. Instead of dealing with individual elements, these special open sets are considered. To demonstrate the speed and memory effectiveness of ITS applied to continuous global optimization are tested in detail by using classical multimodal functions for which minima are known. It has been point out that, ITS collects both the advantages of Tabu Search and Genetic algorithms together. So, the speed, flexibility, applicability have been improved.