Scholarlink

<p>The Coyote Optimization Algorithm (COA) is a nature-inspired method based on the behavior of the social dynamics and environmental adaptability of the <i>Canis latrans</i> species, which lives largely in North America. Since its introduction, the method has been extensively used to solve optimization problems related to energy-power systems, machine learning, control systems, and hyperparameter tuning. This paper focuses on a comprehensive literature review, as well as a bibliometric analysis of the COA algorithm, considering studies related to COA since its proposal in 2018 until 2025. Assuming four research questions: <i>i</i>) What are the most influential and higher-cited studies, and the country-wise analysis? <i>ii</i>) What are the versions proposed to improve the COA? <i>iii</i>) What are the current trends of applications of COA across various domains? <i>iv</i>) What are the future directions and challenges of COA? This study synthesizes a wide range of research efforts to emphasize the variants of COA, their principles, and achievements. It encompasses a broad spectrum of applications, including single-objective, multi-objective, and engineering optimization problems. The studies are ranked using the Ordinatio scoring method, and the most relevant studies are identified and described. Furthermore, this study presents numerical experiments where COA and its variants are evaluated to solve the optimization problems proposed over time in multiple IEEE Congresses on Evolutionary Computation benchmarks (2006, 2007, 2010, 2013, 2014, 2015, 2017).</p> Graphic Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Comprehensive Review of the Coyote Optimization Algorithm: Evolution, Applications, and Future Directions

  • 期刊论文

  • 发表日期：

    2025年10月3日

查看全文
  • Matheus Henrique Dal Molin Ribeiro,
  • Victor Henrique Alves Ribeiro,
  • Lucas Camilotti,
  • Eduardo Luiz Alba,
  • Vinicius Bortolini,
  • Roberto Zanetti Freire,
  • Eduardo de Freitas Rocha Loures,
  • Viviana Cocco Mariani,
  • Leandro dos Santos Coelho

摘要

The Coyote Optimization Algorithm (COA) is a nature-inspired method based on the behavior of the social dynamics and environmental adaptability of the Canis latrans species, which lives largely in North America. Since its introduction, the method has been extensively used to solve optimization problems related to energy-power systems, machine learning, control systems, and hyperparameter tuning. This paper focuses on a comprehensive literature review, as well as a bibliometric analysis of the COA algorithm, considering studies related to COA since its proposal in 2018 until 2025. Assuming four research questions: i) What are the most influential and higher-cited studies, and the country-wise analysis? ii) What are the versions proposed to improve the COA? iii) What are the current trends of applications of COA across various domains? iv) What are the future directions and challenges of COA? This study synthesizes a wide range of research efforts to emphasize the variants of COA, their principles, and achievements. It encompasses a broad spectrum of applications, including single-objective, multi-objective, and engineering optimization problems. The studies are ranked using the Ordinatio scoring method, and the most relevant studies are identified and described. Furthermore, this study presents numerical experiments where COA and its variants are evaluated to solve the optimization problems proposed over time in multiple IEEE Congresses on Evolutionary Computation benchmarks (2006, 2007, 2010, 2013, 2014, 2015, 2017).

Graphic Abstract