Applying social network analysis to systems of systems: The case of the 2016 Puerto Rico power outage

In September 2016, a fire in one of Puerto Rico's ten major power stations left over 1.5 million people without power and blacked out a significant portion of the island. This massive failure of the power system of systems is an example of how systems engineers require better methods for understanding complex, networked systems of systems. This paper proposes adapting social network analysis tools and methods to system of systems analysis in order to provide insights into interconnected networks of systems. First, it explores the literature for systems engineering, system of systems engineering, and social network analysis to identify opportunities to improve system of systems methods. Then it presents a methodology that applies specific metrics from social network analysis to the Puerto Rico power grid. Specifically, it shows how various centrality metrics may indicate when a system of systems is vulnerable to network wide failures. By better understanding the interoperability of a system of systems, engineers may be able to improve the overall robustness of the system of system and prevent catastrophic failures such as the 2016 Puerto Rico power failure.