Team Collaboration as a System of Systems Agent-Based Model

There is a current need to study and understand the behaviors and characteristics of systems of systems. Studying a single system is relatively straightforward when compared to studying a system of systems. A system of systems has unique characteristics that distinguish it from a single system. The additional complexity in a system of systems leads to complicated models and advanced computer simulations. Although modeling and simulation are popular methods for researching a single system, there have been fewer attempts at modeling and simulating systems of systems. Agent-based modeling is an effective approach for researching systems of systems, but validation of agent-based models is difficult, especially if data are not available. Finally, communicating an agent-based model is more difficult than communicating an analytical model because analytical models use familiar mathematical notation. The purpose of this research is to increase the knowledge of system of systems engineering by developing, executing, and analyzing an agent-based model of team collaboration in a real-world, operational system of systems. This research has several goals. The first goal is to address a current need to increase the understanding of the behaviors and characteristics of systems of systems. More specifically, this research aims to model and explain how collaboration and integration in a real-world system of systems affect the achievability of the overall goal of the system of systems. There is an emphasis on the operations and integration of heterogeneous component systems of the collaborative system of systems. This includes understanding the behaviors, characteristics, and interactions among the component systems. The second goal is to develop and thoroughly document a new, repeatable agent-based model of the real-world system of systems. The final goal is to develop a useful tool for understanding and predicting the achievability of the overall goal of the system of systems. Specifically, this research explores team collaboration in a National Basketball Association offensive lineup. This lineup possesses the necessary characteristics to categorize it as a system of systems. Players are the individual, heterogeneous component systems that belong to and operate in the system of systems. This research introduces a new agent-based model and simulation to understand how the individual component systems affect the achievability of the system of systems goal. The NetLogo modeling platform provides an effective environment for executing the model. Data for initialization and validation come from the National Basketball Association. Results show that the overall goal of scoring is an emergent behavior of the collaborative system of systems. Top performing combinations of lineups and collaboration levels emerge. The heterogeneity and interactions of the component systems affect the achievability of the overall goal in different ways. Specific combinations of the collaboration levels and integration of individual component systems determine the scoring output. Observing the component systems individually offers no explanation for the achievability of the overall goal. Instead, it is necessary to view the component systems as a whole. Finally, the verified and validated agent-based model of the offensive lineup contributes to system of systems research, and it is an effective tool for understanding and exploring offensive lineups in the National Basketball Association.