A Novel Hybrid Artificial Bee Colony Algorithm for Numerical Function Optimization

Artificial Bee Colony (ABC) algorithm is very interesting population based swarm optimization technique. This technique is motivated by means of extraordinary nature of honey bees. ABC algorithm commonly used to get to the bottom of nonlinear and complex problems. Comparable to other population based strategies, ABC also has some negative aspects. It is computationally steep because of its sluggish temperament during search process. This paper planned a new approach so as to enhance performance of ABC strategy by combining positive characteristics two most popular modification of ABC algorithm named Gbest-guided ABC (GABC) and Memetic search in ABC (MeABC). It combines best part of both algorithm and balance process of exploration and exploitation in ABC. GABC improves exploitation and MeABC improves exploration. The proposed algorithm is named as Memetic search with Global best solution in ABC (MGABC). It is tested over some impartial test problems with diverse intricacy; results indicate that the anticipated algorithm outperforms the original ABC, GABC and MeABC in most of the experiments. It is also tested over two real world problems namely Welded beam design problem and compression spring problem and results are very impressive for both problems. Keywords— Particle Swarm Optimization, Artificial Bee Colony Algorithm, Swarm Intelligence, Evolutionary Computation, Memetic Algorithm.