Scholarlink

<p>This paper proposes a new meta-heuristic algorithm, the Musk Ox Optimizer (MO). The MO mimics the social behavior of musk ox populations, which make migratory, foraging, and defensive choices based on signals sent by “guards” within the population. These behaviors can be translated into mathematical models and eventually an algorithm that effectively address various optimization challenges in different search spaces. To test the MO, 23 benchmark test functions and CEC 2021 are selected. Furthermore, Wilson sign rank tests are used to judge the statistical significance of the results. The findings indicate that the MO outperforms other algorithms in the literature when applied to the two types of test functions. Moreover, the results derived from practical applications also elucidate the advantages that MO holds in resolving problems within unknown search spaces. The source code of MO is publicly available at <a href="https://www.mathworks.com/matlabcentral/fileexchange/177574-musk-ox-optimizer-mo">https://www.mathworks.com/matlabcentral/fileexchange/177574-musk-ox-optimizer-mo</a>.</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

Musk ox optimizer (MO): a novel optimization algorithm and its application

  • 期刊论文

  • 发表日期：

    2025年10月17日

查看全文
  • Yongliang Yuan,
  • Guangya Chong,
  • Jianji Ren,
  • Wu Zhao,
  • Yanan Li,
  • Zhenxi Wang,
  • Seyedali Mirjalili

摘要

This paper proposes a new meta-heuristic algorithm, the Musk Ox Optimizer (MO). The MO mimics the social behavior of musk ox populations, which make migratory, foraging, and defensive choices based on signals sent by “guards” within the population. These behaviors can be translated into mathematical models and eventually an algorithm that effectively address various optimization challenges in different search spaces. To test the MO, 23 benchmark test functions and CEC 2021 are selected. Furthermore, Wilson sign rank tests are used to judge the statistical significance of the results. The findings indicate that the MO outperforms other algorithms in the literature when applied to the two types of test functions. Moreover, the results derived from practical applications also elucidate the advantages that MO holds in resolving problems within unknown search spaces. The source code of MO is publicly available at https://www.mathworks.com/matlabcentral/fileexchange/177574-musk-ox-optimizer-mo.