A phenotypic analysis of three population-based metaheuristics

Abstract

Metaheuristics are used as very good optimization methods and they imitate natural, biologic, social and cultural process. In this work, we evaluate and compare three different metaheuristics which are population-based: Genetic Algorithms, CHC and Scatter Search. They work with a set of solutions in contrast to trajectory-based metaheuristics which use an only solution. From a comparative analysis, we can infer that Genetic Algorithms and CHC algorithms can solve satisfactorily problems with a growing complexity. While Scatter Search provides high quality solutions but its computational effort is very high too.