The book on “Modeling and Simulation Support for System of Systems Engineering Applications” (Rainey and Tolk 2015) addresses several of the original research agenda items identified by the community, such as described by Valerdi et al. (2008). Chapter 22 of this book explicitly enumerates topic of a research agenda, which are used for this special issue.
Our understanding of a system of systems (SoS) is captured in the second chapter of our book, in which Mark Maier summarises and extends his work in the context of support opportunities for modelling and simulation:
- Operational independence of the individual systems: An SoS is composed of systems that are independent and useful in their own right. If an SoS is disassembled into the component systems, these component systems are capable of independently performing useful operations independently of one another
- Managerial independence of the systems: The component systems not only can operate independently, they generally do operate independently to achieve an intended purpose. The component systems are generally individually acquired and integrated and they maintain a continuing operational existence that is independent of the SoS
- Geographic distribution: Geographic dispersion of component systems is often large. Often, these systems can readily exchange only information and knowledge with one another, and not substantial quantities of physical mass or energy
- Emergent behaviour: The SoS performs functions and carries out purposes that do not reside in any component system. These behaviours are emergent properties of the entire SoS and not the behaviour of any component system. The principal purposes supporting engineering of these systems are fulfilled by these emergent behaviours
- Evolutionary development: An SoS is never fully formed or complete. Development of these systems is evolutionary over time and with structure, function and purpose added, removed, and modified as experience with the system grows and evolves over time
Modeling and simulation in general and agent-directed simulation (ADS) in particular already support systems engineering successfully. Yilmaz and Ören (2009) dedicated a whole book to the synergisms of agent-directed simulation and systems engineering. Simulated systems can be used to obtain, display and evaluate operationally relevant data in agile contexts by executing models using operational data exploiting the full potential of M&S and producing numerical insight into the behaviour of complex systems. ADS have been shown to have the ability to support the development of robust, fault tolerant, adaptive, self-optimising, learning, social capable, autonomous, and agile solutions. ADS also expose emergent behaviour similar to SoS, so that they can be used to better understand and utilise this criterion and enforce positive emergence while avoiding negative emergence.
While some problems have been solved by recent research results, other challenges are still obviously unresolved, but the application of M&S methods promises to offer an improvement. The list of recommendations below regarding topics for the research agenda is neither considered to be complete nor exclusive and is meant to be modified and extended by the community of scholars and practitioners to fit their needs. It hopefully can serve as a core to grow from. The topics below for a research agenda are compiled based on the chapter contributions to Rainey and Tolk (2015) serving as a starting point towards developing a common body of knowledge and educating the workforce on the SoSE approach. This special issue is the next step.
References
DeLaurentis, D.A. A taxonomy-based perspective for systems of systems design methods. Proceedings of the IEEE Conference on Systems, Man and Cybernetics, Vol. 1, pp. 86-91, IEEE Press (2005)
Kuhl, M. E., Kistner, J., Costantini, K., Sudit, M. Cyber attack modeling and simulation for network security analysis. Proceedings of the 39th Conference on Winter Simulation, pp. 1180-1188, IEEE Press (2007)
Rainey, L.B., Tolk, A. (eds.): Modeling and Simulation support for System of Systems Engineering Applications. Wiley, Hoboken, NY (2015)
Tolk, A., Adams, K.M., Keating, C.B. Towards Intelligence-based Systems Engineering and System of Systems Engineering. Tolk, A., Jain, L. (Eds.). Intelligence-based Systems Engineering, Intelligent Systems Reference Library, Vol. 10, pp. 1-22, Springer, Berlin, (2011)
Valerdi, R., Axelband, E., Baehren, T., Boehm, B., Dorenbos, D., Jackson, S., Madni, A., Nadler, G., Robitaille, P., Settles, S.: A research agenda for systems of systems architecting. International Journal System of Systems Engineering, Vol. 1, Nos. 1/2: 171–188 (2008)
Yilmaz, L., Ören, T. (eds.): Agent-Directed Simulation and Systems Engineering. Wiley, Berlin (2009)
Suitable topics include, but are not limited to, the following:
- Taxonomy for SoSE. A literature research on SoSE methods immediately shows that the community is not speaking a common language. Publications are full of synonyms and homonyms, which make it hard to communicate and reuse results from other relevant domains. Mapping results to a common taxonomy will support a better understanding of the common concepts. DeLaurentis' (2005) work is already a good starting point. This common understanding of concepts, relationships, and processes is also critical for the efficient use of M&S methods to improve SoSE. This is because successful M&S methods can be associated with taxonomical concepts, ensuring their reuse; otherwise, experts from other SoSE domains may not recognise application specific elements as being relevant for them
- Theoretic foundations for SoSE. Several authors have already requested a SoSE methodology, i.e., a rigorous engineering analysis that invests heavily in the understanding and framing of the problem under study (Tolk et al. 2011). In particular, modelling has been identified as a successful M&S method to support a better understanding (Zhou et al. 2011). Although these efforts are a good starting point, capturing the ideas in unambiguous and rigorous formal methods is needed as well. The authors are convinced that the field of cybernetics has the potential to better support these efforts. An example is Ashby's (1958) law of requisite variety for the control of complex systems. The application of this law will provide insight into what a common operating picture would look like for a SoS where coherence, as addressed above, is the objective. Beer's (1979) viable system model has the potential to facilitate a better theoretic understanding of resilience. Finally, system thinking (Boardman and Sauser 2008) and its formalisms needs to be re-evaluated in order to provide solid theoretic foundations. The theoretic foundations for SoSE seem to be present, but they may have to be compiled from the variety of related contributing domains into a more coherent body of knowledge
- Organisational and human factors engineering. The human limit to handling complexity and the organisational constraints for systems with operational and managerial independence continue to be unsolved challenges. Currently proposed solutions often focus on technical proposals, but technical efforts cannot solve conceptual problems. A consolidated effort that brings together management expertise, an educated workforce, and supporting technical solutions is needed. Engineering management for SoSE needs to play a pivotal role to ensure that (a) the existing and new technical solutions are recognised by academicians, management professionals and (b) professional education of the workforce is provided
- Engineering emergence. The emergent behaviour of SoS is recognised. The topic of guided emergence was recognised already in the 2006 workshop and evolved into a broader task, namely to actively pursue positive emergence and avoid negative emergence. Positive emergence should not be a welcomed coincident but rather the product of engineering efforts. Now researchers are starting to work on methods to gain a deeper understanding if and how this is possible (Chen et al. 2009). Agent-based simulation is well known for its ability to produce emergence as well. Using sophisticated SoS models to drive agent-based simulations to gain a better understanding is a logical resulting recommendation. One early application may be the use of such models within serious games to create problem awareness and better training for managers (Tolk 2014)
- Cybersecurity. Another topic that evolved significantly over the last years is security. The operational and managerial independence creates a significant challenge for secure solutions. Every interface or access point provided by contributing systems within the SoS federation extends the attack surface exposed to potential threats. The security solutions proposed in the chapters of this book are necessary but not sufficient. Loosely coupling systems offers the rapid accessibility of new functionality, but it also opens the threat of unauthorised access or manipulation of sensitive information. As mentioned before, the development of new security protocols will make the SoS more secure, but managerial and educational processes to raise the awareness of these problems are also needed. Again, M&S methods can support procurement, testing, and training on multiple levels. Kuhl et al. (2007) give examples of efforts that are under development, and some of them are now in operational use to train cybersecurity personnel
- Model-based SoSE. The advantages of model-based systems engineering are well recognised by the traditional systems engineering community by now. The use of a model-based common knowledge repository with a multitude of different facets to support customers, stakeholders, and team members of all life cycles and phases of a system in the form of a consistent system architecture with multiple views or viewpoints is becoming a common approach. These ideas, methods, and supporting tools need to be adapted and evolved to support SoSE as well. As already discussed by Tolk et al. (2011), it is highly recommended to ensure that all artifacts are machine-readable so that intelligent agents and other tools can use them to support users and managers, eventually evolving the state of the art towards M&S-based SoSE
- Academic and professional SoSE education. Although SoSE gained significant academic attention over the last year, the professional education still needs improvement, as the new knowledge has not been transferred well from academia to the workforce. Specifically, managers and commanders at all levels within the command structure of complex organisations are not aware that the professional environment that they work in day-in and day-out is a SoSE application. This can be observed in government, industry and supporting Federally Funded Research and Development Centers. The requisite education will significantly enhance their daily situational awareness. One example in the focus of this book is the use of M&S methods. Several complex organisations still have adopted the discrete-event simulation paradigm as a "one size fits all" solution that worked well in the traditional environment, without awareness of the utility of agent-based simulation or hybrid approaches. As a result, decisions are not based on the latest scientific results and can be improved by providing better academic and professional SoSE education on all levels, including decision makers and managers. What exactly needs to get into curricula and continuous education lessons is open for discussion and needs to be captured as the research agenda progresses.
Important Dates
Submission of Manuscripts: 31 January, 2016
Notification to Authors: 31 March, 2016
Final Versions Due: 31 May, 2016
