MaOSSA: A new high-efficiency many-objective salp swarm algorithm with information feedback mechanism for industrial engineering problems

• Published in: Results in engineering Vol. 25; p. 104372
• Main Authors: Aljaidi, Mohammad, Ramesh, Janjhyam Venkata Naga, Kiran, Ajmeera, Jangir, Pradeep, Arpita, Pandya, Sundaram B., Fendzi Mbasso, Wulfran, Abualigah, Laith, Alkoradees, Ali Fayez, Khishe, Mohammad
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.03.2025; Elsevier
• Subjects: Engineering design problems; Many-objective optimization; Many-objective salp swarm algorithm; Multi-objective optimization; Reference point; Many-objective salp swarm algorithm; Multi-objective optimization; Engineering design problems; Many-objective optimization; Reference point
• ISSN: 2590-1230; 2590-1230
• DOI: 10.1016/j.rineng.2025.104372

•The proposed MaOSSA combines Many-Objective Salp Swarm Algorithm to achieve optimal convergence and diversity in multi-objective problems.•The algorithm benefits from IFM to enhance decision-making processes which results in improved Pareto front convergence and diversity.•The testing evaluated MaOSSA on DTLZ1-DTLZ7 and five real-world MaOPs with 5-15 objectives against leading algorithms.•Superior Performance: Outperforms MaOSCA, MaOPSO, NSGA-III, and MaOMFO in GD, IGD, SD, SP, HV, and RT metrics.•The algorithm guarantees an even distribution of solutions throughout the objective space thus preventing clustering in MaOPs.•The method demonstrates practical engineering value through its application to car cab and truss structure and antenna design projects.•The development team plans to combine constraint-handling methods in power system optimization and parameter estimation tasks for future work. The pursuit of convergence in multi-objective optimization usually results in population clustering that produces suboptimal outcomes for both convergence and diversity performance. This paper introduces MaOSSA as a new Many-Objective Salp Swarm Algorithm which combines reference point strategies with niche preservation and Information Feedback Mechanism (IFM). The strategy enables control of convergence and diversity while simultaneously adapting to alterations in the Pareto front. The algorithm achieves personal diversity through its edge individual preservation strategy and density estimation method which maintains uniform population diversity. The evaluation of MaOSSA included DTLZ1-DTLZ7 benchmark problems and five real-world engineering design problems (RWMaOP1–RWMaOP5) that contained 5 to 15 objectives. The performance evaluation between MaOSCA, MaOPSO, NSGA-III, and MaOMFO algorithms showed that MaOSSA delivered superior outcomes regarding Generational Distance (GD), Inverted Generational Distance (IGD), Spacing (SP), Spread (SD), Hypervolume (HV), and Runtime (RT). The experimental outcomes show MaOSSA delivers superior performance than current methods by achieving optimal convergence-diversity balance which establishes it as an efficient solution for many-objective optimization tasks.