Scholarlink

This study proposes an AI-based framework for curriculum gap analysis that aligns academic content with the evolving knowledge and skill demands of Industry 4.0. By applying Sentence-BERT (SBERT), the method calculates semantic similarity between course descriptions and predefined knowledge areas, as well as between related program learning outcomes (PLOs) and targeted professional skills. Two interpretable metrics are introduced: the Knowledge-Aligned Coverage Score (KACS) and the Skill-Aligned Coverage Score (SACS), which quantify curriculum alignment with knowledge domains and skill expectations, respectively. A genetic algorithm is further employed to discover optimal elective course combinations that minimize misalignment with industry needs. Applied to the undergraduate Industrial Engineering curriculum of a public university in Thailand, the results reveal strong alignment in core technical areas but significant gaps in global competencies, communication, and adaptability. The proposed approach offers a scalable, data-driven tool to support evidence-based curriculum design and continuous improvement in response to a rapidly changing industrial environment.

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

AI-Based Curriculum Gap Analysis Between Industrial Engineering Program and Industry 4.0 Needs: A Case Study in Thai Automotive Industry

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rungchat Chompu-inwai,
  • Trasapong Thaiupathump

摘要

This study proposes an AI-based framework for curriculum gap analysis that aligns academic content with the evolving knowledge and skill demands of Industry 4.0. By applying Sentence-BERT (SBERT), the method calculates semantic similarity between course descriptions and predefined knowledge areas, as well as between related program learning outcomes (PLOs) and targeted professional skills. Two interpretable metrics are introduced: the Knowledge-Aligned Coverage Score (KACS) and the Skill-Aligned Coverage Score (SACS), which quantify curriculum alignment with knowledge domains and skill expectations, respectively. A genetic algorithm is further employed to discover optimal elective course combinations that minimize misalignment with industry needs. Applied to the undergraduate Industrial Engineering curriculum of a public university in Thailand, the results reveal strong alignment in core technical areas but significant gaps in global competencies, communication, and adaptability. The proposed approach offers a scalable, data-driven tool to support evidence-based curriculum design and continuous improvement in response to a rapidly changing industrial environment.