Harnessing Generative AI for Adaptive Learning in Mechanics: A Competency-Based Approach

<p dir="ltr">Abstract </p><p dir="ltr">Introduction: The increasingly diverse learning needs of today's students pose significant challenges to educators, particularly when teaching abstract concepts requiring visualization. Generative AI offers an innovative solution, enabling personalized, competency-based education (CBE) tailored to individual learners. This study explores the application of Microsoft Copilot Studio to create an adaptive learning platform for Mechanics 1, a foundational module at Singapore Polytechnic. The focus is on the topic of moments, a concept that about 60% of the 404 enrolled students struggle with due to difficulties in visualizing the moment arm and force (Mayer, 2009). </p><p dir="ltr">Methods: The adaptive platform dynamically adjusts question difficulty based on performance, using AI-driven nodes to scaffold learning effectively. Students begin with conversational AI interactions that assess their learning profiles and preferences. If a student encounters difficulties, targeted interventions, such as videos, images, or audio resources, are automatically provided. These resources are stored in a shared repository and selected based on predefined AI training, reducing educators' cognitive workload (Siemens, 2013). The platform continues to engage students until they indicate readiness to conclude. A detailed learner profile is generated for analysis, enabling refinement of teaching strategies and resource optimization. </p><p dir="ltr">Results: Preliminary results indicate a 25% improvement in student performance on Mechanics 1 assessments and a 35% increase in satisfaction levels, as captured in end-semester surveys. Additionally, educators reported a 30% reduction in the time spent curating and delivering individualized materials. By integrating conversational AI, the platform bridges generational gaps, presenting concepts in relatable terms and fostering student engagement. </p><p dir="ltr">Conclusion: The initial success of this initiative underscores the transformative potential of generative AI in competency-based education. While the current focus is on moments in Mechanics 1, the platform's scalability enables its application to other complex topics, such as thermodynamics, biology, psychology, economy and/or any other abstract topics that are challenging to conceptualize. This aligns with the SkillsFuture movement's goals of fostering lifelong learning and preparing a future-ready workforce (SkillsFuture Singapore, 2023). Future research will explore broader cross-disciplinary applications and evaluate longitudinal impacts on learning outcomes, advancing education towards a more inclusive and adaptive framework.</p>