A Decomposition-Based Multi-Objective Flying Foxes Optimization Algorithm and Its Applications

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Bio-inspired Computing

Background:

• The flying foxes optimization (FFO) algorithm demonstrates effectiveness in single-objective optimization.
• Existing multi-objective optimization algorithms face challenges in complex problem-solving.

Purpose of the Study:

• To adapt the flying fox optimization strategy for multi-objective problems.
• To introduce a novel decomposition-based multi-objective flying foxes optimization algorithm (MOEA/D-FFO).

Main Methods:

• Developed a new offspring generation mechanism to enhance peripheral space exploration.
• Implemented a population updating approach with adjusted neighbor matrices for improved convergence.
• Compared MOEA/D-FFO against established and state-of-the-art algorithms.

Main Results:

• MOEA/D-FFO achieved superior performance, outperforming classical and cutting-edge algorithms in over 11 benchmark tests.
• Experimental results demonstrated high adaptability across different population sizes.
• The algorithm showed significant improvements in exploration and convergence rates.

Conclusions:

• MOEA/D-FFO effectively addresses multi-objective optimization challenges.
• The proposed algorithm offers a promising approach for engineering applications.
• The enhanced population management strategy is key to its success.