By Jane Stoyle Welch and Nicole Capozziello
Published October 2, 2020
When the Department of Materials Design and Innovation launched in 2015, its mission was to take materials science education to the next level. In order to do this, faculty took the unique approach of teaching materials science from a data science and informatics perspective.
A fundamental challenge in teaching a subject like materials science is that students need to absorb and rapidly connect concepts that link materials behavior at the engineering scale with molecular scale phenomena. The pedagogical framework of materials science and engineering requires that the perspectives of chemistry, physics and engineering all converge.
“By using the lens of data science and informatics to view materials science, the unique MDI curriculum can achieve a multidisciplinary/multi-scale convergence to train a new genre of materials scientists,” says Krishna Rajan, SUNY Distinguished Professor and Erich Bloch Chair of the Department of Materials Design and Innovation. “Students can learn about different materials systems by revealing the information and relationships encoded in the data, and can then use data to accelerate the discovery of new materials, and in the process accelerate their own learning process that provides them the critical skills to address societal needs in a more effective, innovative and accelerated way.”
To better build the interconnections between diverse but foundational concepts, faculty in the department analyzed the curriculum using network representations and relevant concepts from graph theory.
Over a two-year period, they analyzed eight core courses of the MDI master’s curriculum. Two of the courses focused on data science and informatics concepts, and the other presented a blend of traditional materials science and data science. They created a comprehensive database that evaluated syllabi, homework assignments, and numerous exam questions.
From the over 200 specific topics addressed in the courses, they condensed the topic list to 10-15 topics for analysis. Finally, they identified how each of these topics connected to other topics across the curriculum, a process that involved active discussions among teaching faculty and peer-auditing each other’s lectures.
“By representing the curriculum as a mathematical network, we have shown how materials science and data science converge in our newly-developed curriculum,” says Erik Einarsson, associate professor in the Department of Materials Design and Innovation who also holds a joint appointment in the Department of Electrical Engineering. “As a result of letting data take center stage, we believe we are better preparing students to be successful scientists and researchers.”
This type of data driven network-based approach could potentially enable personalized education, by showing students exactly what topics will be covered in a curriculum.
Detailed findings and analysis of the network-based approach to curriculum development are provided in “Data-driven visualization schema of a materials informatics curriculum: Convergence of materials science and information science,” published earlier this year in MRS Advances. In addition to Einarsson and Rajan, co-authors include Olga Wodo, Prathima Nalam, Scott Broderick, Kristofer Reyes and E. Bruce Pitman, all from the Department of Materials Design and Innovation.
The Department of Materials Design and Innovation is a joint program between the University at Buffalo’s College of Arts and Sciences and School of Engineering and Applied Sciences.