2025 ASEE Annual Conference & Exposition

Here we present a novel project-based materials science course that integrates experimental design, computational modeling, and peer-reviewed publication using AI-powered tools like Mathematica. This course was modified from a traditional materials science course to further develop critical engineering skills specifically programming, modeling, and the use of AI and machine learning tools. By combining hands-on experiments with computational modeling, students gain a comprehensive understanding of material behavior while also developing crucial skills in artificial intelligence and simulation techniques.
The project that is at the heart of the course requires students not only to design and test an experimental system but also to develop an interactive model/simulation in Mathematica that will be submitted and reviewed for peer review publication. In the development of this model students learn and utilize AI and machine learning tools.
The student teams are intentionally composed to reflect diverse backgrounds, experiences, and perspectives. Each team is tasked with designing and performing a series of experiments related to core materials science topics such as mechanical properties, phase transitions, and microstructural analysis. Simultaneously, students use Mathematica’s AI-driven tools to model their experiments, analyze data, and simulate outcomes. The combination of experimental and computational work enhances students' understanding of both theoretical and practical aspects of materials science, fostering deeper learning and critical thinking.
One of the key innovations in this course is the integration of AI-driven modeling tools with open educational resources (OER). The use of OER allows students to access high-quality learning materials without the financial burden of expensive textbooks or software licenses. Additionally, students are encouraged to submit their computational models and simulations for peer review through the Mathematica Demonstrations platform. This submission process not only provides students with professional feedback but also familiarizes them with the process of scholarly publication, an essential skill for future engineers and researchers.
The project-based nature of the course has led to several positive educational outcomes. First, students have reported increased engagement and motivation due to the hands-on, real-world relevance of their projects. Second, the course’s focus on teamwork and collaboration, particularly in diverse groups, has enhanced equity by ensuring that all students, regardless of background, can contribute meaningfully to the learning process. Third, the integration of AI tools and computational modeling has prepared students for the growing importance of data science and artificial intelligence in modern engineering practice.
Our preliminary assessment of the course outcomes suggests that this approach has led to significant improvements in student performance, particularly for underrepresented groups in engineering. Students from diverse backgrounds have not only participated more actively in the course but have also demonstrated higher levels of confidence in their ability to conduct both experimental and computational research. The course's structure has also fostered a more inclusive learning environment where peer feedback and collaboration are central to the educational experience.
In conclusion, this project-based materials science course represents a promising model for integrating experimental work, AI-driven modeling, and open educational resources to create a more engaging, equitable, and effective learning experience for engineering students. We believe that this approach can be adapted and scaled to other engineering disciplines, helping to advance both student learning outcomes and equity in engineering education. Future work will focus on further refining the course structure and assessing long-term impacts on student retention, particularly for historically underrepresented populations in STEM fields.