SUMMARY

Many sources advocate for more opportunities for engineering students to practice their design and creativity skills with problems throughout the curriculum. The large time input required for providing formative feedback to problems or the lengthy distribution process for hands-on activities generally prohibits their extensive usage, in favor of standard homework problem sets This thesis presents a study on the usage of Sketch Mechanix, an AI-augmented sketch recognition homework tool used to distribute traditional homework problem sets, as well as an open-ended design problem, within introductory statics and dynamics classes at multiple universities across the United States.
This tool has been developed to reinforce the conceptualization, development, and solution of engineering systems through idealized free-body diagrams. Mechanix provides the user with a drawing space and employs sketch recognition algorithms to automatically detect general body FBDs and planar 2d trusses. The application provides tutoring in order to correct mistakes made during the solution procses. Students at the participating universities were divided into control groups and Mechanix groups and asked to complete a series of homework problems. The control group uses only the traditional homework system assigned to that class section, while the Mechanix group completes a set of homework problems using the Mechanix application in addition to the standard homework system. Students were also asked to complete an open-ended design problem using assigned condition. Results suggest that the usage of the Mechanix application accelerates student understanding of mechanics concepts. Further results from the truss design problem suggests the application allows participants to explore more of the solution space, resulting in a final solution that exceeds the design criteria.