MOANA: Multi-objective ant nesting algorithm for optimization problems.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Engineering Applications

Background:

• Multi-objective optimization problems (MOPs) are prevalent in science and engineering.
• Existing evolutionary algorithms face challenges with scalability and solution diversity in MOPs.
• The Ant Nesting Algorithm (ANA) is a metaheuristic, but its multi-objective capabilities are limited.

Purpose of the Study:

• To introduce the Multi-Objective Ant Nesting Algorithm (MOANA) for solving MOPs.
• To enhance exploration-exploitation balance and solution diversity in multi-objective optimization.
• To demonstrate MOANA's effectiveness on benchmark problems and real-world engineering tasks.

Main Methods:

• Developed MOANA by extending the Ant Nesting Algorithm (ANA).
• Incorporated adaptive deposition weight parameters for balancing exploration and exploitation.
• Utilized a polynomial mutation strategy to ensure solution diversity and quality.
• Evaluated performance on ZDT functions and CEC 2019 multi-modal benchmarks.

Main Results:

• MOANA demonstrated superior convergence speed and Pareto front coverage compared to MOPSO, MOFDO, MODA, and NSGA-III.
• The algorithm achieved a broad range of optimal solutions in the welded beam design problem.
• MOANA effectively addressed limitations in scalability and diversity found in traditional evolutionary algorithms.

Conclusions:

• MOANA is a robust and effective algorithm for tackling complex multi-objective optimization tasks.
• Its adaptive mechanisms and mutation strategy contribute to high-quality and diverse solutions.
• MOANA presents a practical tool for decision-making in engineering and other optimization-intensive fields.