MOJMA: A novel multi-objective optimization algorithm based Java Macaque Behavior Model

Abstract

We introduce the Multi-objective Java Macaque Algorithm for tackling complex multi-objective optimization (MOP) problems. Inspired by the natural behavior of Java Macaque monkeys, the algorithm employs a unique selection strategy based on social hierarchy, with multiple search agents organized into multi-group populations. It includes male replacement strategies and a learning process to balance intensification and diversification. Multiple decision-making parameters manage trade-offs between potential solutions. Experimental results on real-time MOP problems, including discrete and continuous optimization, demonstrate the algorithm's effectiveness with a 0.9% convergence rate, outperforming the MEDA/D algorithm's 0.98%. This novel approach shows promise for addressing MOP complexities in practical applications.

Keywords

• multi-objective optimization
• java macaque algorithm
• nature-inspired optimization
• continuous optimization problem
• discrete optimization problem