短学时地下工程专业《岩土工程勘察》课程的人机协作教学模式探索

Exploration of a Human-AI Collaborative Teaching Mode for the “Geotechnical Investigation” Course in Underground Engineering Majors under Limited Class Hours

摘要: 基于地下工程专业《岩土工程勘察》课程的教学实践，针对该课程在特定培养方案下存在的学时受限、内容繁杂及野外实践教学缺失等问题，本文开展了基于BOPPPS模型、融合真实工程资料复盘与人工智能(AI)辅助的混合式教学探索。研究旨在通过依托真实工程情境，将工程一线原始勘察数据引入课堂，并借助AI工具优化教学流程。对30名学生的教学实践显示：该教学模式有效提升了学习积极性，期末平均成绩达82分；事故案例导入与参与式学习环节获得了85.7%的正面评价。问卷调查同时显示，约25%的学生在岩石宏观辨识与复杂参数取值上仍存在掌握不足。实践表明，AI工具虽能显著提升规范查询与报告编写效率，但无法完全替代实物感知训练；后续教学需进一步完善人机协作下的数据复核机制，兼顾数智化技能与工程伦理培养。

Abstract: Based on the teaching practice of the “Geotechnical Investigation” course for Underground Engineering majors, this paper addresses challenges such as limited teaching hours, complex content, and the absence of field practice within specific training schemes. It explores a blended teaching approach based on the BOPPPS model, integrating the review of authentic engineering data with Artificial Intelligence (AI) assistance. The research aims to return to engineering fundamentals by introducing raw frontline investigation data into the classroom and leveraging AI tools to optimize the teaching process. Teaching practice involving 30 students demonstrates that this mode effectively enhanced learning enthusiasm, resulting in an average final score of 82. Furthermore, the introduction of accident cases and participatory learning segments received 85.7% positive feedback. However, questionnaire surveys indicate that approximately 25% of students still show deficiencies in macroscopic rock identification and the selection of complex parameter values. Practice suggests that while AI tools significantly improve the efficiency of standard code consultation and report writing, they cannot fully replace physical perception training. Future teaching should further refine data verification mechanisms under human-AI collaboration, balancing the cultivation of digital intelligence skills with engineering ethics.