Challenges with Regard to Learning Analytics Metrics in Complex, Multidimensional Courses with Use of Artificial Intelligence Tools

The growing significance of new engineering methods, such as Model-Based Systems Engineering (MBSE), necessitates that engineering curricula evolve to prepare future engineers effectively. A promising approach to teaching these methods involves combining seminars and case studies within engineering design projects. To succeed, students must grasp interdisciplinary concepts and process methods, but lecturers face challenges in defining appropriate tasks, tracking individual progress, and fostering self-regulated learning. Traditional assessment tools, like standard questionnaires, have proven inadequate in monitoring student learning progress. Instead, structured interviews at key milestones were used to compare solutions and methods between students and lecturers, yielding valuable insights but proving time-consuming and limited in scope. This raises the question of how to efficiently collect meaningful learning analytics data in such courses.