Scholarlink

<p>In the global employment landscape, career planning education is crucial for enhancing the competitiveness of higher education students. Within this framework, engagement in collegiate competitions has become a widely adopted method for developing professional competencies, cultivating a passion for learning, and advancing career aspirations. However, the traditional approach to competition selection predominantly relies on personal experience and unsystematic guidance, which often leads to undesirable results, including imbalanced skill development, wasted energy, squandered time, and diminished job-market competitiveness. Since the traditional approach does not take into account skill coverage, time commitment, and energy expenditure when addressing the multi-objective and multi-constraint nature of competition selection. To solve this problem, this study proposes an innovative energy-aware staged-elite multi-objective particle swarm optimisation algorithm (ES-MOPSO) to develop an optimal competition selection system for higher education. In addition, the proposed algorithm features a staged iteration and elite-update strategy to efficiently optimise multiple objectives while avoiding local optima. The developed system integrates energy-aware reward modelling and structured data management to enable optimal, personalised competition selection for students. In the context of student competition selection, experimental validation on representative scenarios demonstrates that the ES-MOPSO outperforms established optimisers, such as MOPSO, SS-MOPSO, and NSGA-II, according to the hypervolume (HV) metric. Moreover, the case study analysis indicates that the proposed system (1) reduces the coefficient of variation (CV) in skill development by over 30%, effectively resolving imbalances in skill growth; (2) decreases fluctuations in energy investment by approximately 10%, facilitating more scientific and efficient time management; and (3) achieves higher competition scores with equivalent energy input or requires less energy to attain comparable scores. Furthermore, scalability experiments demonstrate the system’s robust performance in complex environments.</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

ES-MOPSO: an energy-aware staged-elite MOPSO system for optimal competition selection in higher education

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月23日

查看全文
  • Tingwen Ruan,
  • Weiyu Lu,
  • Yuchen Song,
  • Zheng Jun Chew,
  • Tianle Ma,
  • Feng Liu

摘要

In the global employment landscape, career planning education is crucial for enhancing the competitiveness of higher education students. Within this framework, engagement in collegiate competitions has become a widely adopted method for developing professional competencies, cultivating a passion for learning, and advancing career aspirations. However, the traditional approach to competition selection predominantly relies on personal experience and unsystematic guidance, which often leads to undesirable results, including imbalanced skill development, wasted energy, squandered time, and diminished job-market competitiveness. Since the traditional approach does not take into account skill coverage, time commitment, and energy expenditure when addressing the multi-objective and multi-constraint nature of competition selection. To solve this problem, this study proposes an innovative energy-aware staged-elite multi-objective particle swarm optimisation algorithm (ES-MOPSO) to develop an optimal competition selection system for higher education. In addition, the proposed algorithm features a staged iteration and elite-update strategy to efficiently optimise multiple objectives while avoiding local optima. The developed system integrates energy-aware reward modelling and structured data management to enable optimal, personalised competition selection for students. In the context of student competition selection, experimental validation on representative scenarios demonstrates that the ES-MOPSO outperforms established optimisers, such as MOPSO, SS-MOPSO, and NSGA-II, according to the hypervolume (HV) metric. Moreover, the case study analysis indicates that the proposed system (1) reduces the coefficient of variation (CV) in skill development by over 30%, effectively resolving imbalances in skill growth; (2) decreases fluctuations in energy investment by approximately 10%, facilitating more scientific and efficient time management; and (3) achieves higher competition scores with equivalent energy input or requires less energy to attain comparable scores. Furthermore, scalability experiments demonstrate the system’s robust performance in complex environments.