A guided population archive whale optimization algorithm for solving multiobjective optimization problems

Abstract

This paper proposes a new multiobjective algorithm by extending the recently Whale Optimization Algorithm (WOA) to solve multiobjective optimization problems. Our algorithm which we shall call Guided Population Archive Whale Optimization Algorithm (GPAWOA) is based on Pareto dominance and uses an external archive to store the non-dominated solutions found during the optimization process. The leaders are selected from the archive to guide the population towards promising regions of the search space, also, the mechanism of crowding distance is incorporated into the standard WOA to maintain the diversity. Our proposed algorithm is evaluated on twelve benchmark functions and is applied to four multi-objective engineering design problems: 4-bar truss design, gear train problem, disk brake design and welded beam design and compared against three well-known algorithms: Multiobjective Evolutionary Algorithm Based on Decomposition (MOEA/D), Multi-Objective Grey Wolf Optimizer (MOGWO), and Multi-Objective Particle Swarm Optimization (MOPSO). The experimental results indicate that the proposed GPAWOA is highly competitive and outperforms the selected state-of-the-art multiobjective optimization algorithms, being able to provide an excellent approximation of Pareto front in terms of convergence and diversity.