MOBRO: multi-objective battle royale optimizer.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• The Battle Royale Optimizer (BRO) is a novel game-based optimization algorithm.
• Existing BRO versions address single-objective problems, leaving a gap for multi-objective applications.
• The no-free-lunch theorem highlights the need for diverse optimization algorithms.

Purpose of the Study:

• To develop and implement a multi-objective version of the Battle Royale Optimizer (MOBRO).
• To evaluate MOBRO's performance on standard multi-objective benchmark datasets.
• To compare MOBRO against state-of-the-art multi-objective optimization algorithms.

Main Methods:

• The proposed Multi-Objective Battle Royale Optimizer (MOBRO) was designed and implemented.
• MOBRO was applied to CEC 2009, CEC 2018, ZDT, and DTLZ benchmark datasets.
• Performance was assessed using inverted generational distance, maximum spread, and spacing metrics.

Main Results:

• MOBRO demonstrated superior performance across most benchmark suites.
• The algorithm exhibited competitive results against other state-of-the-art methods.
• Evaluation focused on convergence, spread, and distribution aspects of optimization.

Conclusions:

• MOBRO successfully extends the game-based optimization approach to multi-objective problems.
• The developed algorithm offers a competitive alternative to existing multi-objective optimization techniques.
• Further research can explore MOBRO's application in diverse scientific and engineering domains.