System-of-Systems Complexity

Abstract: The global availability of communication services makes it possible to interconnect independently developed systems, called constituent systems, to provide new synergistic services and more efficient economic processes. The characteristics of these new Systems-of-Systems are qualitatively different from the classic monolithic systems. In the first part of this presentation we elaborate on these differences, particularly with respect to the autonomy of the constituent systems, to dependability, continuous evolution, and emergence. In the second part we look at a SoS from the point of view of cognitive complexity. Cognitive complexity is seen as a relation between a model of an SoS and the observer. In order to understand the behavior of a large SoS we have to generate models of adequate simplicity, i.e, of a cognitive complexity that can be handled by the limited capabilities of the human mind. We will discuss the importance of properly specifying and placing the relied-upon message interfaces between the constituent systems that form an open SoS and discuss simplification strategies that help to reduce the cognitive complexity.