By Dr. Raed Jaradat
The 21st century has presented a special class of problems. This class is referred to system of systems (SoS) and has been receiving increased attention in the literature, including a journal, International Journal of System of Systems Engineering, devoted to the study of this field and associated phenomena. Traditional approaches to engineering of systems (e.g. traditional systems engineering) has been challenged as suspect (Keating, et al., 2003; Adams, 2011; Adams and Keating, 2011b) for application to this new class of problems marked by high levels of ambiguity, uncertainty, and emergence. The traditional science based approach (system engineering) to deal with problems is by reducing (reductionism) the problems into parts and deriving solutions as a function of the understanding of the parts. This approach is sufficient in systems where problems are well bounded and relationships can be understood in direct correlation to performance (outputs). However, this is not the case in SoS problem domain.
Despite being successful for many years, traditional systems engineering is not intended to address problems that are mired in; i) turbulent environmental conditions; ii) ill-defined problem conditions; iii) contextual dominance; iv) uncertain approach; v) ambiguous or changing expectations and objectives; vi) unclear integration of multiple complex systems; and finally, vii) excessive complexity (Keating 2009, 177). In sum, the ability of traditional reductionist based approaches to dealing with the emerging class of “system of systems” problem domain is in doubt. For engineers who must operate on these problem domains, it also suggests that a different level of thinking is necessary.
There is currently a lack of knowledge and development of purpose built and tested techniques supportive of the maturing of SoS problems. In conclusion, we need to develop and create purposefully designed and structured tools and methods that can efficiently deal with today’s real- world complex problems.