A Computational Study of Design Team Robustness Through the Lens of Cognitive Style

Abstract

High-performing design teams are characterized by their ability to maintain performance across a variety of problem types. This is often referred to as robustness, and is usually achieved through careful management of team processes. However, there exists an opportunity to design teams that are likely to be inherently robust by addressing and embracing the individual variability of team members. Cognitive style provides an avenue by which we can compose robust teams based on the problem-solving approach of the individual. In this work, we used the KAI agent-based organizational optimization model (KABOOM) to evaluate the effects of team composition and team structure on the robustness of overall team performance. Teams of homogeneous and heterogeneous KAI styles were tasked to solve a variety of different abstract design problems and evaluated based on their performance with and without sub-teams. Results indicate that there is a significant difference in the distribution of aggregate scores for homogeneous and heterogeneous teams without sub-teams, and heterogeneous teams may be more robust. Sub-teams were found to significantly increase the overall median score and robustness for some teams.