Asee peer logo

Developing Instructional Design Agents to Support Novice and K-12 Design Education

Download Paper |


2019 ASEE Annual Conference & Exposition


Tampa, Florida

Publication Date

June 15, 2019

Start Date

June 15, 2019

End Date

October 19, 2019

Conference Session

Design in Engineering Education Division: Capstone Design Practices

Tagged Division

Design in Engineering Education

Page Count




Permanent URL

Download Count


Request a correction

Paper Authors


Corey T. Schimpf Concord Consoritum Orcid 16x16

visit author page

Corey Schimpf is a Learning Analytics Scientist with interest in design research, learning analytics, research methods and under-representation in engineering, A major strand of his work focuses on developing and analyzing learning analytics that model students’ cognitive states or strategies through fine-grained computer-logged data from open-ended technology-centered science and engineering projects. His dissertation research explored the use of Minecraft to teach early engineering college students about the design process.

visit author page

author page

Xudong Huang Concord Consortium

author page

Charles Xie

author page

Zhenghui Sha University of Arkansas


Joyce E. Massicotte Concord Consortium

visit author page

Joyce Massicotte serves as a Project Manager at the Concord Consortium. She holds a B.A. in Environmental Studies from San Francisco State University and an M.S. in Resource Administration and Management from the University of New Hampshire (UNH). Prior to joining the Concord Consortium, Joyce worked as Senior Manager of Program Development at Next Step Living, Program Manager at the UNH Office of Sustainability, Program Manager and Technical Support at, and as an environmental educator. Joyce has a special interest in clean energy and creating real-world, authentic activities in the classroom to educate, engage, and empower students to build a brighter, more sustainable future today.

visit author page

Download Paper |


There is a growing interest from teachers in incorporating more engineering design into K-12 classrooms. This has been bolstered by initiatives such as the Next Generation Science Standards that push for greater integration of engineering and other STEM topics into K-12. However, there are several challenges to implementing design projects in these settings. Many K-12 teachers have limited or no experience in teaching engineering or design. Even those who have some experience teaching design, trying to oversee an entire classroom of students navigating open, complex design problems can prove to be a daunting task. This is exacerbated further for teachers with no design teaching experience. As a result, many may avoid trying design projects due to time commitments and a steep learning curves.

To encourage broader adoption of design projects in K-12 classrooms we propose virtual design agents, whose behavior is structured by one or more artificial intelligence (AI) algorithms and who are embedded in design platforms such as computer-aided-design (CAD), to assist teachers. These design agents can engage in non-trivial design practices and strategies to help teachers in supporting and managing students’ design process as well as facilitate and scaffold students’ learning experience. By embodying design practices these agents encode some basic design intelligence such as being able to generate new designs or taking an existing design and searching for an improved design. When employing these design agents in the classroom, there are three pedagogical decisions to consider: the role/mode of the agent, the role of the student and the interaction between them both. This tripartite model allows for the flexible creation of design scenarios to introduce and immerse students in design projects. Here mode refers to how many design agents are active in the project. For example, one design scenario could involve multiple agents who represent junior members of a design team with the student acting as the lead designer. The student may take the lead with creating an initial design and seek new designs from their design agent “team” to ideate and diverge in exploring design solutions.

In this work, we ground the design agent and the tripartite model in our specific curriculum and platform where students design a solar farm in the CAD platform Energy3D. Design agents in Energy3D rely on several genetic algorithms, a common set of AI algorithms which involve creating several populations of design artifacts and keeping those that perform better than others, which enables them to create new designs and improve students’ designs. We discuss several possible design scenarios that can emerge from different model arrangements for this challenge and then report the results of a pilot study where students acted as mentees to an expert design agent who helped them learn about the performance of solar panel racks under several conditions.

Design agents hold promise for augmenting teachers’ ability to run design projects and supporting novice students in engaging in such projects. In this way, they may represent a powerful approach for promoting greater adoption of design projects in K-12 settings.

Schimpf, C. T., & Huang, X., & Xie, C., & Sha, Z., & Massicotte, J. E. (2019, June), Developing Instructional Design Agents to Support Novice and K-12 Design Education Paper presented at 2019 ASEE Annual Conference & Exposition , Tampa, Florida. 10.18260/1-2--32640

ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2019 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015