Kakapo Optimization Algorithm (KOA): A Novel Bio-inspired Metaheuristic for Optimization Applications

Abstract

Kakapo Optimization Algorithm (KOA), a novel bio-inspired metaheuristic grounded in the unique ecological and behavioural characteristics of the endangered kakapo parrot is presented. Unlike flighted birds, kakapos rely on nocturnal roaming, camouflage, booming calls, and cautious foraging strategies for survival. These natural mechanisms are systematically translated into mathematical operators to achieve the critical optimization balance between global exploration and local exploitation. The exploration process is modelled on the kakapo’s zigzag nocturnal navigation, ensuring broad coverage of the search space and preventing premature convergence. The exploitation phase is inspired by freezing and camouflage behaviours, introducing controlled local refinements to intensify promising regions while maintaining solution stability. A survival-based elitist selection further preserves high-quality solutions, sustaining convergence reliability across iterations. KOA has been extensively validated on the CEC 2017 benchmark suite, encompassing unimodal, multimodal, hybrid, and composition functions. Comparative analysis against recent metaheuristic algorithms demonstrates KOA’s robust adaptability, consistent accuracy, and high stability, as confirmed through statistical metrics and boxplot visualizations. Results indicate that KOA effectively converges toward global optimum with low variance, outperforming or matching state-of-the-art algorithms across diverse problem landscapes. The biologically grounded design of KOA, its strong exploratory and exploitative dynamics, and its reliable convergence highlight its potential for real-world applications such as engineering design, scheduling, energy optimization, and machine learning hyperparameter tuning.

Keywords: Kakapo Optimization Algorithm, Metaheuristic, Nature-Inspired Computing, Exploration And Exploitation, Global Optimization, CEC 2017 Benchmarks.