A systematic review of the crested porcupine optimizer: variants and applications
Abstract
The Crested Porcupine Optimizer, first presented in 2024, is a population-based metaheuristic that has been inspired by the multimodal defense behaviors of crested porcupines. Its unique characteristics in defense-driven movement patterns and explicit repulsion-attraction mechanisms have attracted the attention of many researchers to solve complex nonlinear and high-dimensional optimization problems. This study performs the first systematic review on CPO with the PRISMA 2020 guideline based on peer-review publications from January 2024 to November 2025 across the major academic databases. This paper reviews its biological inspiration, mathematical formulation, and exploration–exploitation balance of CPO with comparative performance evaluation against well-established algorithms such as PSO, GWO, and WOA, as well as major CPO variants, on CEC2017 benchmarks. Existing improvements are classified into initialization strategies, perturbation mechanisms, adaptive parameters, multi-objective extensions, discrete versions, and hybrid frameworks. Analysis of the 117 improvement strategies extracted from the 43 variants shows that perturbation mechanisms account for 38%, population initialization for 27%, and adaptive parameter control for 23%. Application domains are comprehensively reviewed, and machine learning (84 studies) and signal processing with modal analysis (32 studies) represent the most active research area. Although the citations almost tripled from 2024 to 2025, some challenges still persist, which include stagnation in high-dimensional problems, lack of theoretical convergence analysis, and inconsistent evaluation of variants. This work establishes a unified taxonomy, identifies critical research gaps, and proposes possible future directions such as chaos-enhanced exploration, rigorous theoretical analysis, and large-scale distributed implementation for positioning CPO as a competitive bio-inspired optimization paradigm for contemporary computational challenges.