A Sustainable Framework for Teaching Quality Assurance in Engineering Education: A Case Study of AI-Enhanced Multi-Agent Interactive Classrooms

Traditional engineering education often relies on delayed summative assessments, struggling to provide the real-time feedback required for sustainable teaching quality improvement. To address this gap, this study proposes a sustainable teaching quality assurance framework leveraging OpenMAIC, an AI-enhanced Multi-Agent Interactive Classroom. A comparative case study was conducted in an undergraduate “Electronic Technology” course. The experimental group utilized the AI-driven curriculum, where an “AI Professor” and “AI Peers” engaged students in heuristic dialogues while generating dynamic HTML5 virtual simulations for immersive circuit experiments. We evaluated the framework’s effectiveness by analyzing fine-grained learning-unit analysis interaction logs, milestone scores, and stakeholder surveys. The results revealed that the multi-agent pedagogical model enabled real-time detection of potential misconceptions based on interaction behaviors, identifying learning bottlenecks weeks earlier than traditional exams. Furthermore, the experimental group significantly outperformed the control group in complex project-based learning tasks, alongside reporting reduced learning anxiety. Instructors also experienced a substantial decrease in repetitive administrative workloads. This study concludes that integrating multi-agent AI and immersive virtual simulations not only enhances continuous process assessment but also establishes a brand-new scalable and continuously improvable teaching paradigm for future digital engineering education.