Need: The research on design thinking aims to understand how expert designers think in order to develop new methods and tools for supporting their work more effectively, with an educational goal to establish good models of design thinking for students to follow. But design thinking is a moving target that advances with engineering technology. Generative design is a transformative design technology that uses open-ended artificial intelligence (AI) algorithms to arrive at solutions for engineering problems. It allows exploration of a wider variety of potential solutions, with the goal of arriving at an optimal solution in partnership with the human engineer. As leading companies such as Autodesk and PTC launched generative design software and industries embraced the technologies, a new era of engineering is forthcoming. This paradigm shift in design methods entails a fundamental change of mindset for design thinking that must be addressed in the engineering education of the future workforce.
Guiding Questions: Research questions from three perspectives will drive the project: 1) Theoretical perspective: What are the essential elements of generative design thinking that students must acquire so they can work effectively at the human-technology frontier in engineering? 2) Practical perspective: To what extent and in what ways can the curriculum and materials support the learning of generative design as indicated by students’ gains in generative design thinking? 3) Affective perspective: To what extent and in what ways can AI affect the professional formation of engineers as indicated by the changes in students’ interest and self efficacy in engineering?
Outcomes: To answer these questions, interdisciplinary research that integrates the perspectives and knowledge in engineering design, computer science, learning science, and workforce development is being conducted. So far, we have achieved major outcomes in four aspects: 1) the development of the open-source generative design software, Aladdin, and data-driven generative design approaches based on a novel target-embedding variational autoencoder (TEVAE) neural network architecture; 2) the research on an operational definition of generative design thinking (GDT); 3) the research on understanding the impact of students’ cognitive competencies in systems thinking (one of the most important dimensions of GDT) on their design process and design solutions; and, 4) the development of the instructional modules that support project-based learning of generative design.
Broader Impacts: The products of this project are expected to equip students with essential skills and mindsets needed to master using AI approaches in contemporary engineering practices. During Year 2, 202 students completed the introductory generative design module. In the next step, the project will involve more than 1,000 students at 13 institutions around the country. The materials developed by the project will be open-source, including an open-source tool for teaching and learning generative design and a set of project-based learning modules that guide the use of the tool to solve authentic problems in product design, architectural engineering, and energy systems design. With these impacts, this project will contribute to the core research on the Future of Work at the Human-Technology Frontier, one of NSF’s 10 Big Ideas, from the field of engineering.
Onan Demirel, Oregon State University, Corvallis, OR; Charles Xie, Institute for Future Intellegence, Boston, MA; Molly Goldstein, University of Illinois Urbana-Champaign, Champaign, IL; Darya Zabelina, The University of Arkansas, Fayetteville, AR