An improved SABO model for solving architectural engineering design problems

This study proposes an enhanced version of the Subtraction Average-Based Optimizer (SABO) to address the slow convergence, local optima, and imbalance between exploration and exploitation inherent in the original algorithm.The enhancement incorporates chaotic mapping for uniform population initialization and introduces a sinusoidal function to enhance global search efficiency.The enhanced SABO algorithm is evaluated using 23 benchmark functions from IEEE CEC2005, and it outperforms several state-of-the-art algorithms in terms of convergence speed and accuracy.Compared to existing swarm intelligence algorithms, the enhanced SABO algorithm consistently demonstrates superior performance across all 23 benchmark functions.The practical applicability of the enhanced SABO algorithm is further validated through its application to a real-world engineering design problem.Experimental and statistical analyses confirm that the enhanced SABO algorithm achieves rapid convergence, demonstrates superior optimization capability, and exhibits strong effectiveness in addressing complex multimodal functions.Furthermore, the enhanced SABO algorithm effectively maintains a balance between exploration and exploitation, significantly enhancing solution quality in multi-objective optimization problems.