Evaluating the effectiveness of a hybrid teaching approach in an undergraduate electric and hybrid vehicle course

Purpose This study aimed to evaluate the effectiveness of a domain-specific hybrid teaching model for the undergraduate Electric and Hybrid Vehicles course. The model integrates traditional lectures, flipped classroom strategies, and project-based learning to improve student engagement, learning outcomes, and practical application skills. It also examines the potential of AI-based predictive analytics in monitoring and enhancing student performance. The broader purpose is to align teaching methods with NBA and ABET accreditation standards and explore opportunities for scalable implementation in engineering education. Design/methodology/approach A quasi-experimental mixed-methods design was used. The experimental group (n = 30) was taught using the hybrid model comprising traditional lectures (20%), flipped classroom strategies (50%), and project-based learning (30%). The control group (n = 30) received conventional instruction. Pre- and post-test scores measured knowledge gain, while project outcomes assessed practical learning. Student engagement was predicted using an AI-driven decision tree classifier trained on Moodle activity data. Quantitative analysis was supported with qualitative feedback to assess satisfaction levels and perceived learning improvements. Findings The experimental group demonstrated a 41.67% improvement in post-test scores compared to a 20.69% gain for the control group, with large effect sizes indicating strong impact. The AI-driven classifier achieved 87% accuracy in predicting student engagement, enabling targeted interventions for at-risk learners. Student satisfaction levels were high at 88%, and project outcomes showed marked improvement in creativity, technical quality, and teamwork. These findings confirm the hybrid model's effectiveness in enhancing conceptual understanding, practical application, and learner engagement beyond what conventional teaching methods deliver. Research limitations/implications The study was conducted with a relatively small sample size (n = 60) from a single institution, which may limit generalizability. The focus was restricted to one specialized course, so further research is required to validate the model across multiple domains and disciplines. Long-term retention of knowledge and transferability to professional contexts were not measured. Future studies could expand the sample, apply advanced AI methods for predictive analytics, and investigate scalability in diverse institutional settings. These findings imply the potential for hybrid pedagogy and AI-based learning analytics in engineering education. Practical implications The hybrid model provides engineering educators with a structured, evidence-based teaching approach that enhances both theoretical knowledge and applied skills. The integration of AI-driven engagement monitoring enables proactive support for students, reducing dropout risks and improving academic outcomes. Institutions can adopt this framework to align with NBA and ABET accreditation requirements while improving student readiness for industry challenges. The approach is scalable, making it feasible for wider implementation across technical courses. Project-based components foster innovation, teamwork, and problem-solving skills, directly benefiting professional preparedness. Social implications By enhancing engineering education quality C, the hybrid model contributes To Producing More industry-ready graduates who are skilled in emerging areas like Electric and Hybrid Vehicles. The integration of AI for learning analytics demonstrates how technology can personalize education and support equity by identifying and assisting struggling learners early. Improved student satisfaction and engagement can foster a culture of lifelong learning. Broader adoption of such models can strengthen the technical workforce C contributing to sustainable mobility solutions and addressing global challenges in clean energy and transportation. Originality/value This study is among the first to design and test a domain-specific hybrid teaching model in Electric and Hybrid Vehicle education, combining flipped classroom, project-based learning, and lectures in a structured ratio. The integration of AI-driven predictive analytics adds a novel dimension by enabling personalized interventions based on student engagement data. The research provides empirical evidence of significant learning gains, improved project outcomes, and alignment with international accreditation standards. Its value lies in demonstrating a scalable, replicable, and innovative framework for modern engineering education.