Architecting systems of systems: A tertiary study

Context: The term System of Systems (SoS) has increasingly been used in a wide variety of domains to describe those systems composed of independent constituent systems that collaborate towards a mission that they could not accomplish on their own. There is a significant volume of research by the software architecture community that aims to overcome the challenges involved in architecting SoS, as evidenced by the number of secondary studies in the field published so far. However, the boundaries of such research do not seem to be well defined, at least partially, due to the emergence of SoS-adjacent areas of interest like the Internet of Things.Objective: This paper aims to investigate the current state of research on SoS architecting by synthesizing the demographic data, assessing the quality and the coverage of architecting activities and software quality attributes by the research, and distilling a concept map that reflects a community-wide understanding of the concept of SoS. Method: We conduct what is, to the best of our understanding, the first tertiary study on SoS architecting. Such tertiary study was based on five research questions, and was performed by following the guidelines of Kitchenham et al. In all, 19 secondary studies were evaluated, which is comparable to other tertiary studies. Results: The study illustrates a state of disconnection in the research community, with research gaps in the coverage of particular phases and quality attributes. Furthermore, a more effective approach in classifying systems as SoS is required, as the means of resolving conceptual and terminological overlaps with the related domains. Conclusions: Despite the amount of research in the area of SoS architecting, more coordinated and systematic targeted efforts are required in order to address the identified issues with the current state of research.