Pipefish Optimization Algorithm (POA): A Nature-inspired Metaheuristic for Robust and Adaptive Global Optimization

Abstract

Pipefish Optimization Algorithm (POA), a novel nature-inspired metaheuristic designed to efficiently solve complex numerical optimization problems, is presented in this paper. POA draws its inspiration from the unique reproductive and behavioral strategies of pipefish, including mutual mate choice, male pregnancy, selective paternal investment, and natural mortality. These biological mechanisms are systematically mapped into computational operators to achieve a robust balance between exploration and exploitation. The mutual mate choice mechanism facilitates broad and diverse exploration of the search space, preventing premature convergence, while male pregnancy and incremental offspring refinement guide local exploitation toward high-quality solutions. Selective resource allocation and elitist survival ensure that only the most promising candidates progress, enhancing convergence reliability. The algorithm’s performance has been rigorously evaluated on the CEC 2017 benchmark suite, which comprises unimodal, multimodal, hybrid, and composition functions. Comparative analysis against nine state-of-the-art metaheuristics—including MOA, WaOA, AOA, GWO, LSA, SWO, TLBO, BaOA, and WSO—demonstrates that POA consistently achieves top-ranked results across the majority of test functions. In unimodal landscapes, POA precisely converges to global optimum, while in multimodal and complex composition functions, it effectively avoids local optima and maintains robust solution quality. Statistical assessments and boxplot visualizations confirm the algorithm’s high reliability, low variance, and stable convergence behavior. The results indicate that POA is a versatile and scalable optimizer, capable of addressing diverse real-world optimization problems, including engineering design, resource allocation, and parameter tuning tasks. Its biologically grounded design, adaptive search mechanisms, and consistent performance make POA a promising tool for both theoretical research and practical applications in high-dimensional and complex optimization scenarios.