2026 ASEE Annual Conference & Exposition

Machine learning (ML) is a cornerstone of artificial intelligence, yet its instruction in engineering curricula often remains confined to standalone courses designed for computer science (CS) majors, typically requiring substantial programming experience. This paper presents a practical approach to integrating ML into existing engineering courses for non-CS majors by expanding a traditional Design of Experiments (DOE) laboratory module. In this expanded DOE lab, students first conduct a full factorial DOE using a catapult experiment. In the second part of the lab, the collective data sets from all groups are shared with every student, serving as the foundation for constructing ML regression and classification models, introducing students to core ML concepts through hands-on labs.
This approach introduces several innovations to engineering education. Students engage in experiential learning by interacting with a catapult and collecting original data, rather than relying on generic online sources. ML is integrated by connecting it to DOE, a standard curriculum topic, allowing seamless inclusion without major course changes. DOE is highlighted for situations with limited or costly data, while ML is employed for large datasets, providing students with a thorough overview of statistical and predictive methods in engineering.
The objective of this expanded DOE lab is to introduce fundamental machine learning concepts to engineering majors, enabling them to collaborate effectively with data scientists on interdisciplinary projects. The instructional materials, including laboratory guides, source code, and assessment tools, are designed for straightforward adaptation into any course covering experimental methods and data analysis. Student learning outcomes were evaluated through project presentations, quizzes, and reflective surveys over two semesters. Results indicate increased engagement and improved comprehension of both experimental design and machine learning principles.