TY - JOUR
T1 - Multi-Objective Volleyball Premier League algorithm
AU - Moghdani, Reza
AU - Salimifard, Khodakaram
AU - Demir, Emrah
AU - Benyettou, Abdelkader
N1 - Funding Information:
We would like to thank Professor Ponnuthurai N. Suganthan and Dr. Seyyed Mostafa Kalami Haris for giving MO test functions and providing MOPSO and MOEA/D codes. This research is also supported by CIIORG (Computational Intelligence and Intelligent Optimization Research Group) of the Persian Gulf University .
Funding Information:
We would like to thank Professor Ponnuthurai N. Suganthan and Dr. Seyyed Mostafa Kalami Haris for giving MO test functions and providing MOPSO and MOEA/D codes. This research is also supported by CIIORG (Computational Intelligence and Intelligent Optimization Research Group) of the Persian Gulf University.
Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2020/5/21
Y1 - 2020/5/21
N2 - This paper proposes a novel optimization algorithm called the Multi-Objective Volleyball Premier League (MOVPL) algorithm for solving global optimization problems with multiple objective functions. The algorithm is inspired by the teams competing in a volleyball premier league. The strong point of this study lies in extending the multi-objective version of the Volleyball Premier League algorithm (VPL), which is recently used in such scientific researches, with incorporating the well-known approaches including archive set and leader selection strategy to obtain optimal solutions for a given problem with multiple contradicted objectives. To analyze the performance of the algorithm, ten multi-objective benchmark problems with complex objectives are solved and compared with two well-known multi-objective algorithms, namely Multi-Objective Particle Swarm Optimization (MOPSO) and Multi-Objective Evolutionary Algorithm Based on Decomposition (MOEA/D). Computational experiments highlight that the MOVPL outperforms the two state-of-the-art algorithms on multi-objective benchmark problems. In addition, the MOVPL algorithm has provided promising results on well-known engineering design optimization problems.
AB - This paper proposes a novel optimization algorithm called the Multi-Objective Volleyball Premier League (MOVPL) algorithm for solving global optimization problems with multiple objective functions. The algorithm is inspired by the teams competing in a volleyball premier league. The strong point of this study lies in extending the multi-objective version of the Volleyball Premier League algorithm (VPL), which is recently used in such scientific researches, with incorporating the well-known approaches including archive set and leader selection strategy to obtain optimal solutions for a given problem with multiple contradicted objectives. To analyze the performance of the algorithm, ten multi-objective benchmark problems with complex objectives are solved and compared with two well-known multi-objective algorithms, namely Multi-Objective Particle Swarm Optimization (MOPSO) and Multi-Objective Evolutionary Algorithm Based on Decomposition (MOEA/D). Computational experiments highlight that the MOVPL outperforms the two state-of-the-art algorithms on multi-objective benchmark problems. In addition, the MOVPL algorithm has provided promising results on well-known engineering design optimization problems.
KW - Engineering design optimization problems
KW - Global optimization
KW - Multi-objective evolutionary algorithm
KW - Pareto solution
UR - http://www.scopus.com/inward/record.url?scp=85082876945&partnerID=8YFLogxK
U2 - 10.1016/j.knosys.2020.105781
DO - 10.1016/j.knosys.2020.105781
M3 - Article
AN - SCOPUS:85082876945
SN - 0950-7051
VL - 196
JO - Knowledge-Based Systems
JF - Knowledge-Based Systems
M1 - 105781
ER -