Improved Distributed Genetic Algorithms Based on Their Methodologies and Processes
In this paper we evaluates the effectiveness of three different distributed genetic algorithms (DGAs). The first one is DGA with Diversity Guided Migration, second one is DGA with Automated Adaptive Migration and the last one is DGA with Bicoded chromosomes and confidence rates. All these algorithms were investigated to improve the overall quality of solutions in the distributed genetic algorithm for different problems. Our comparison between those algorithms depended on some important factors ;like, achieving diversity in selecting individuals, process of replacing the individuals between subpopulations, computational time and memory space. As a result, DGA with Diversity Guided Migration (DGM), was nominated to be better than the other DGAs. It improves the performance for many problems and search spaces.
