Singer Optimization Algorithm: An Effective Human-Based Approach for Solving Optimization Tasks

The Singer Optimization Algorithm (SOA) is a novel human-based metaheuristic inspired by the learning and progression of singers.Unlike traditional metaheuristics, SOA operates without requiring parameter tuning, enhancing adaptability across diverse optimization tasks.The algorithm models three key learning phases: (i) Exploration (Beginner Stage), where diverse solutions are generated to cover the search space broadly, (ii) Imitation (Intermediate Stage), where promising solutions refine their performance by mimicking the best candidates while maintaining individuality, and (iii) Creation (Advanced Stage), where novel solutions emerge through adaptive perturbations, ensuring continuous improvement and avoiding premature convergence.The mathematical foundation of SOA enables dynamic position updates without external control parameters, making it an efficient, self-adaptive optimizer.SOA's performance is assessed on 23 benchmark functions, spanning unimodal, high-dimensional multimodal, and fixed-dimensional multimodal categories.Comparative analyses against nine recent metaheuristics-SFOA, CFOA, COA, POA, OOA, MPA, RSA, AVOA, and WSO-demonstrate that SOA consistently ranks among the top performers.In unimodal functions, SOA exhibits superior exploitation capabilities, achieving optimal solutions with high precision.For multimodal functions, SOA efficiently balances exploration and exploitation, leading to faster convergence and enhanced robustness.Convergence analysis and statistical evaluations confirm SOA's efficiency in solving complex optimization problems while maintaining solution diversity.The algorithm's ability to operate without tunable parameters provides a major advantage over conventional approaches, making it a highly competitive alternative for real-world optimization applications.