Title :
A new selection operator to improve the performance of genetic algorithm for optimization problems
Author :
Ritthipakdee, Amarita ; Thammano, Arit ; Premasathian, N. ; Uyyanonvara, Bunyarit
Author_Institution :
Comput. Intell. Lab., King Mongkut´s Inst. of Technol. Ladkrabang, Bangkok, Thailand
Abstract :
Nature-inspired algorithms, such as Particle swarm optimization (PSO), Ant colony optimization (ACO), and Firefly algorithm, are well known for solving NP-hard optimization problems. They are capable of obtaining optimal solutions in a reasonable time. The algorithm presented in this paper is a combination of a firefly mating concept and genetic algorithm. Genetic algorithm is used as the core of the algorithm while a firefly mating concept is used to compose a new selection operator. The proposed algorithm is tested on four standard benchmark functions. Experimental results have confirmed that the proposed algorithm is not only computationally more efficient than both the original firefly algorithm and the genetic algorithm but also almost always ensure the optimal solutions.
Keywords :
computational complexity; genetic algorithms; ACO; NP-hard optimization problems; PSO; ant colony optimization; firefly algorithm; firefly mating concept; genetic algorithm; nature-inspired algorithms; particle swarm optimization; selection operator; Algorithm design and analysis; Benchmark testing; Biological cells; Genetic algorithms; Optimization; Sociology; Statistics; Firefly Algorithm; Genetic Algorithm; Metaheuristics; Optimization; Swarm Intelligence;
Conference_Titel :
Mechatronics and Automation (ICMA), 2013 IEEE International Conference on
Conference_Location :
Takamatsu
Print_ISBN :
978-1-4673-5557-5
DOI :
10.1109/ICMA.2013.6617947