Revisiting the Question: Are Systems of Systems just (traditional) Systems or are they a new class of Systems?

Abstract

This paper revisits a question asked and debated widely over the past decade: are Systems of Systems (SoS) just traditional systems or are they a new class of systems? Many have argued that SoS are a new class of systems, but little research has been available to provide evidence of this. In this paper we share highlights of recent research to show SoS not only have a different structure than systems and thus need to be engineered differently, but also may possess different attributes for beyond first use properties (the “illities”) such as flexibility and adaptability as compared to systems. By examining historical examples and by using a maritime security SoS as a research test bed, this paper shows that the “ility” called survivability had some design strategies that were directly mapped from systems and also allowed new strategies that only made sense for a SoS (e.g. vigilance). The paper also shows that some design strategies have a different implementation and meaning (e.g. margin) at the level of a system compared to SoS level. We conclude the answer to the question “Are SoS’s just systems?” is both yes and no. They are manifestly systems but possess properties not found in traditional systems. This is shown to true of the meta-property of survivability as applied against a directed SoS.