2026 ASEE Annual Conference & Exposition

This complete evidence-based practice paper describes the development of an auto-grading system that allows for mastery-based learning in a large first-year engineering design course. An early skill first-year engineering students develop is the ability to design and model parts using Computer Aided Design (CAD) software. CAD proficiency is foundational to a host of engineering disciplines as it enables engineers to design, fabricate and analyze parts and assemblies. To develop CAD proficiency, many introductory courses require students to practice creating parts that meet certain specifications. As students practice these tasks they learn typical CAD functions such as extrudes, fillets, lofts, holes, and assembling and dimensioning parts. These lessons build off one another: sketches are extruded into bodies, which have features added to them, and parts and drawings are created and assembled together. The sequential nature of CAD therefore means that before moving on to more complex CAD tasks, it is critical that students develop a robust understanding of the fundamental tasks.

Given this sequential and skill-dependent nature of CAD, teaching methods that emphasize mastery and iterative learning can be particularly effective. Mastery-based learning has emerged as an effective method for students to develop skills that build off of one another. In mastery-based learning, students are allowed to correct and resubmit work many times until it is correct. The idea behind this approach is that students must demonstrate mastery over a course topic before they are able to move on to subsequent topics.

Despite clear advantages, mastery learning is usually not utilized in large CAD courses. This is in large part due to logistical and resource constraints that make it difficult to quickly provide detailed feedback on student work. Instead, students often wait weeks to discover that they made an error. As a result, students may fail to fully develop their CAD skills, which will detract from their ability to be an effective engineer in the future.

After identifying these deficiencies in the current state of CAD education and grading, we developed a method to provide students with instantaneous feedback on the quality of the CAD models they have created via automatically graded CAD assignments. To achieve this goal, we developed a software solution that leverages Autodesk Fusion 360's Application Programming Interface (API) to provide immediate feedback to students about their CAD models. The software collects data on part geometry, sketching constraints, and other parameters that an instructor can choose to have graded. This program then parses this data to provides immediate feedback on students' designs and enables students to engage in mastery-based learning as they are allowed multiple submissions with immediate feedback after each attempt. In addition to the API program, we also utilize PrairieLearn, an open-source mastery-based learning platform, to provide feedback and track student submissions.

This approach has been used successfully at a large Midwest University and student feedback has been positive. Overall, the auto-grader program improves the types and frequency of feedback available to students. The program also significantly reduces the time that the course staff allocates for grading, and creates more time to develop course materials and provide additional support for students. Current plans include expansion to more assignments with various submission requirements and formats, and mapping use of the API program to learning outcomes.