A framework for evolving system of systems engineering

T here is an ongoing debate that questions whether there are differences between a system and a system of systems (SoS). For purposes of this discussion, we define a system to be a construct or collection of different elements that together produce results not obtainable by the elements alone. The elements – or parts – can include people, hardware, software, facilities , policies, and documents; that is, all things required to produce systems-level results [1]. On the other hand, an SoS is a condition where a majority of the following five characteristics are present: operational independence, managerial independence, geographic distribution, emergent behavior, and evolutionary development [2]. There are numerous definitions for both systems and SoS, but in essence there are three schools of thought that exist on the issue of similarities and differences. The first school believes that there are fundamental differences between the systems and SoS and, as a result, they warrant different names, methodologies, and tools to bring them to realization. The pro-difference camp appears to represent a majority on the debate, as evidenced by the amount of its advocacy across government, industry, and academia. Examples within the last five years include the following: The second school believes that SoS is simply an unnecessary term for a system, and that society should not be influenced by this subtlety. To follow Rechtin's heuristic: One man's architecture is another man's detail [4]. Similarly, an SoS for one organization may be seen as a system to another. The advocates for this approach believe that traditional systems engineering practices along with effective program management will be sufficient for both systems and SoS. The third school is agnostic on the issue, playing the wait-and-see game. Since there may not be enough evidence to prove that there is or is not a difference, then there is no sense in claiming loyalty to either of the first two schools of thought to avoid the risk of being wrong. This opinion represents the minority, since from a practical perspective there are few engineers who are unwilling to take a side on technical issues. At the center of the debate about the semantic difference between systems and SoS are the engineering activities that are involved in each. For the definition of systems engineering, for example, we can turn to widely accepted standards [5, 6, 7] used to define the what and how of the activities needed to …