For more than five decades, systems engineering has been known as a well-established field. Different definitions for a system, in diverse context, have been provided in literature. Examples are ANSI/EIA-632-1999, DAU Systems Engineering Fundamentals, IEEE Std 1220-1998, ISO/IEC 15288:2008, NASA Systems Engineering Handbook, and INCOSE Systems Engineering Handbook. For instance, IEEE Std. 1220-1998 defines a system “A set or arrangement of elements and processes that are related and whose behaviour satisfies customer/operational needs and provides for life cycle sustainment of the products”. Or as per ANSI/EIA-632-1999, a system is “An aggregation of end products and enabling products to achieve a given purpose”.
To engineer a system (design, develop, and maintain), a typical process such as the one illustrated below is followed:
Systems engineering (SE) enablers and facilitators area series of toolsets that has been developed, enhanced, and optimized to assist Requirement Management, Design, Project Management, Problem Solving, Collaboration, and Decision Making in a typical SE process. Other engineering fields such as Cognitive Systems Engineering, Configuration Management, Control Engineering, Interface Design, Mechatronics Engineering, Program and Project Management, Reliability Engineering, Security and Safety Engineering and Risk Management and Software Engineering also endeavor major growth to empower ever-increasing SE demands.
The technological, social, economic, and environmental evolution in recent years requires connectivity between ever-increasing complex systems.This new environment implies and leads to a new concept and systems definition called System of Systems. System of Systems (SoS) is a collection of interconnected complex dynamic systems each of which are independent in structure and governance. Thought they are occasionally competitors in their activities, they collaborate, by force or in a volunteer basis, to achieve specific objectives and emergent properties which are not otherwise achievable. Generally, the high-level changes to transition from System to SoS are centred around coordination and interoperability. In a SoS, the focus is therefore on performance optimization, robustness and reliability.
To make it more clear, below are some well-known and already existing/operational SoS examples:
Let’s start with Complex Dynamic Systems (CDS).In CDS, the traditional SE tools are still applicable and useable whilst in SoS, SE tools face major shortcomings and fail to respond appropriately to the SoS issues because of interoperable and autonomous properties. On the other hand, Large Scale Systems(LSS) can be decomposed into sub-systems, leading to hierarchical control and its output information can be distributed which leads to decentralized control. Another difference between the two systems is that LSS sub-systems cannot operate independently and the LSS capability is not beyond the sum of individual capabilities of each system. Multi-Agent Systems(MAS) are another type of traditional systems, a special case of SoS, that have homogenous system members. In SoS, having homogenous system members is not a necessary condition.
The main challenges for the transition from SE to SoSE can be listed as follow:
The key benefit of using SoSE is to improve the competitiveness in the market, which in turn, heavily rely on applied policies and governing rules to assure that the expected emergent properties are met and the implementation and governing costs and efforts are minimized. Policy/Rules/Governance optimization is therefore the end goal in the design procedure of any SoS. The expected emergent properties encompass all the system requirements which include, but are not limited to the following dimensions:
It is important to emphasize that SoS already exists, and the SE discipline has been effectively functioning for over 5 decades, but the ability to transition into SoSE is still an opportunity that should be explored.