Investigating Effect of Organization Structure and Cognitive Profiles on Engineering Team Performance Using Agent Based Models and Graph Theory

Abstract

Engineering organizations have to maximize return on investment (ROI) on the products they design and market. Involvement of human decision makers complicates the design process because not only their technical skills affect the results, but also their other cognitive skills in addition to their role and seniority in the organization. These cannot be studies easily within existing organizations for logistical reasons, while at the same time human-subject research is time consuming and difficult. An agent-based model, building upon previous research, is employed in this work which addresses both the above issues. The contributions of this work include an agent-based framework for modeling and simulating engineering design as informed by the organization structure which is represented by a network graph. In particular, the model uses a scale-free network using the Barabasi-Albert algorithm. Using the undirected graph thus generated, we use the eigenvector centrality to depict the realtive influence of a team-member on design decisions. This comprehensive model allowed us to investigate a wide variety of problem settings and the effect of cognitive profiles on the engineering design process. Individuals' intrinsic and organization-induced cognitive skills were incorporated into the model. The model also allowed for the individuals' cognitive profiles to affect and be affected by their experience in the team design effort. Based on our results, we were able to show that organizations that promote collaborative work, particularly with the aim of engaging skilled individuals, are more likely to come up with better designs and/or converge to designs faster