A robust and efficient algorithm for numerical optimization problem: DEPSO-Scout: A new hybrid algorithm based on DEPSO and ABC

This paper presents a new hybrid algorithm that well performs in solving the numerical optimization problem. The DEPSO-Scout is inspired by Differential Evolution (DE), Particle Swarm Optimization (PSO) and Artificial Bee Colony (ABC). Strong points of DE and PSO which is solution exploitation and exploration are used in algorithm design. We propose a new idea about how to combine DE and PSO to reduce the chance of getting stuck into local optima. Adaptive weight in PSO is introduced to balance exploration and exploitation phase. Moreover, with the interesting characteristic of Artificial Bee Colony (ABC) in suboptimal avoidance, scout Bee concept is then selected and included to be a part of proposed algorithm. The experimental result on three numerical benchmark functions shows that the proposed DEPSO-Scout outperform DE, PSO, and ABC. The improvement opportunity of DEPSO-Scout is discussed in the last section.