ADEPT: Assessing Design Engineering Project Classess with Multi-Disciplinary Teams

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

The objective of this engineering education project is to develop: 1) a physical infrastructure that enables systematic capture of student design team synchronous and asynchronous communications; 2) design webs that integrate and summarize the team communications to provide traceability and accountability for individual contributions to the shared knowledge; 3) assessment methods for learning in multi-disciplinary, project-based engineering design courses; 4) a method to support the facilitated improvement of engineering design courses; and 5) a dataset collected from a testbed multidisciplinary design project course. The scope of this project is unique in that its goal is to capture all of the group interactions of a 25 person multi-disciplinary design team over the course of a semester. Part of the underlying technology that enables the automatic assessment tool is a novel form of automatic multi-document summarization. What is different about this form of summarization is that the resulting representation of communication is organized around individuals rather than around pieces of information. While state-of-the-art summarization technology is dominated by forms of extractive summarization, this application requires the inclusion of insights about individual members of the group, the relationships between them and their ideas, and their evolution over time. The PIs propose to administer many pre- and post-tests to capture as much as is possible the initial and final knowledge states of the students. Multidisciplinary teams are critical in engineering education. This proposal addresses larger issues that are central to learning in teams that are difficult to observe. This proposal suggests a novel approach to analyzing team learning. This research could revolutionize understanding of how multidisciplinary teams integrate their knowledge. This research has the potential to have an impact on teaching engineering design courses at many universities because we will disseminate the technology requirements and the assessment tools. In addition, the PI team will make the dataset of the completely instrumented semester-long course publicly available so that other researchers will be able to use it to answer their own related research questions. This research has the potential to have an impact on teaching engineering design courses at many universities.