Feedback Analytics for High-Performance Design Teams

When designing complex systems, engineering design teams navigate the problem-solution space, leading to emergence of shared knowledge, concepts, and prototypes. In this process, feedback is an essential mechanism to improve team performance. Considering that engineering design teams are complex adaptive systems, instantaneous feedback could be used to improve team performance. This study explores feedback analytics based on teams' shared knowledge structures that could be used in near real-time to analyze discussion patterns in high-performance design teams. Data for the study was collected during a design team effectiveness workshop where seven industry engineering design teams were asked to generate conceptual prototypes in response to the same design prompt. Each team proposed two conceptual solutions to the given challenge: a Best Fit solution, which has a greater likelihood of meeting the design requirements, and a Dark Horse solution, which challenges the design constraints to generate a radical design outcome. Transcribed team discussions were used to develop feedback analytics using natural language processing methodologies and knowledge structure models. Primarily, the alignment of team knowledge networks with experts is examined in the feedback analytics when project discussions are oriented towards the Best Fit solution and the Dark Horse solution. The results have implications for educators, and practitioners in terms of extracting intervention strategies to improve team performance during design discussions.