Multi-objective particle swarm optimization with random immigrants

• Published in: Complex & intelligent systems Vol. 6; no. 3; pp. 635 - 650
• Main Authors: Ünal, Ali Nadi, Kayakutlu, Gülgün
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.10.2020; Springer Nature B.V
• Subjects: Algorithms; Archiving; Complexity; Computational Intelligence; Data Structures and Information Theory; Engineering; Error analysis; Heuristic methods; Multiple objective analysis; Mutation; Original Article; Particle swarm optimization; Performance measurement; Multi-objective optimization; Metaheuristics; Particle swarm optimization; Random immigrants
• ISSN: 2199-4536; 2198-6053
• DOI: 10.1007/s40747-020-00159-y

Complex problems of the current business world need new approaches and new computational algorithms for solution. Majority of the issues need analysis from different angles, and hence, multi-objective solutions are more widely used. One of the recently well-accepted computational algorithms is Multi-objective Particle Swarm Optimization (MOPSO). This is an easily implemented and high time performance nature-inspired approach; however, the best solutions are not found for archiving, solution updating, and fast convergence problems faced in certain cases. This study investigates the previously proposed solutions for creating diversity in using MOPSO and proposes using random immigrants approach. Application of the proposed solution is tested in four different sets using Generational Distance, Spacing, Error Ratio, and Run Time performance measures. The achieved results are statistically tested against mutation-based diversity for all four performance metrics. Advantages of this new approach will support the metaheuristic researchers.