Modeling Resilience in System of Systems Architecture

The objective of this study is to investigate the value of an ecologically inspired architectural metric called the degree of system order in the system of systems (SoS) architecting process. Two highly desirable SoS attributes are the ability to withstand and recover from disruptions (resilience) and affordability. In practice, more resilient SoS architectures are less affordable, and it is essential to balance the trade-offs between the two attributes. Ecological research analyzing long-surviving ecosystems (nature’s resilient SoS) using the degree of system order metric has found a unique balance of efficient and redundant interactions in their architecture. This balance implies that highly efficient ecosystems tend to be inflexible and vulnerable to perturbations, while highly redundant ecosystems fail to utilize resources effectively for survival. Motivated by this unique architectural property of ecosystems, this study investigates the response to disruptions versus affordability trade space of a large number of feasible SoS architectures. Results indicate that the most favorable SoS architectures in this trade space share a specific range of values of degree of system order. This suggests that degree of system order can be a key metric in engineered SoS development. Evaluating the degree of system order does not require detailed simulations and can, therefore, guide the early-stage SoS design process toward more optimal SoS architectures.

Many software systems today have system-of systems (SoS) architectures comprising interrelated and heterogeneous systems, which are developed by multiple teams and companies. Such systems emerge gradually and it is hard to analyze or predict their behavior due to their scale and complexity. In particular, certain behavior only emerges at runtime due to complex interactions between the involved systems and their environment. Monitoring the behavior of SoS at runtime is thus essential during development and evolution. However, existing monitoring approaches are often limited to particular architectural styles or technologies and are thus hard to apply in SoS architectures. In this paper we first analyze the challenges for monitoring SoS based on an industrial SoS for the automation of metallurgical plants. We then propose a flexible framework for monitoring heterogeneous systems within a SoS. We demonstrate its feasibility by applying it to two systems of an industrial SoS. We also report results of an evaluation assessing the framework's performance and scalability.

Fuzzy Optimization of Acknowledged System of Systems Meta-architectures for Agent based Modeling of Development

: A major challenge to the successful planning and evolution of an acknowledged System of Systems (SoS) is the current lack of understanding of the impact that the presence or absence of a set of constituent systems has on the overall SoS capability. Since the candidate elements of an SoS are fully functioning, stand-alone Systems in their own right, they have goals and objectives of their own to satisfy, some of which may compete with those of the overarching SoS. These system-level concerns drive decisions to participate (or not) in the SoS. Individual systems typically must be requested to join the SoS construct, and persuaded to interface and cooperate with other Systems to create the new capability of the proposed SoS. Current SoS evolution strategies lack a means for modeling the impact of decisions concerning participation. The goal of this research is to model the evolution of the architecture of an acknowledged SoS that accounts for the ability and willingness of constituent systems to support the SoS capability development. Since DoD Systems of Systems (SoS) development efforts do not typically follow the normal program acquisition process described in DoDI 5000.02, the Wave Model proposed by Dahmann and Rebovich is used as the basis for this research on SoS capability evolution. The Wave Process Model provides a framework for an agent-based modeling methodology, which is used to abstract the non-utopian behavioral aspects of the constituent systems and their interactions with the SoS. In particular, the research focuses on the impact of individual system behavior on the SoS capability and architecture evolution processes. A proof of concept agent-based model (ABM) of the system interactions is developed and integrated with a genetic algorithm (GA) to explore the potential architectural design space, using a fuzzy associative memory (FAM) to evaluate candidate architectures for simulating SoS creation and evolution.

Behavior models are key models in SoS (System of Systems) modeling and simulation in aircraft top-level demonstration. After architecture modeling to comprehensively describe the SoS, it is essential to transform the models for SoS mission simulation to achieve the virtual verification and validation in MBSE (Model-Based System Engineering). With the analysis of behavior modeling, two methods are introduced. Driving the entities in mission simulation system directly by the state machine models in SoS architecture after the interface definition is a proven method. Through the format conversion of the SoS architecture outputs, the operational/system state diagrams can be transformed to the corresponding organizations/systems behavior models in the mission simulation system. A sample case is provided to illustrate the continuous behavior modeling by format conversion in an air-to-surface attack mission. Compared to the traditional manual behavior modeling, the format conversion method leaves out the repeated behavior modeling and saves the corresponding working load.KeywordsBehavior modelingSystem of systemsModel transformation

Littoral combat is a critical component of naval warfare that has gained importance since the late 20th century due to a shift from matched blue-water combat to agile, asymmetric warfare near the coast. With the increased relevance of this type of naval combat, the ability of a networked naval warfare System-of-Systems (SoS) to operate effectively given the challenges of operations in the littoral zone becomes imperative. Resilience in an SoS in the face of a variety of threats is vital. These threats may take various forms: traditional warfare, cyber attacks, or communications breakdowns. The challenge for any SoS practitioner is identifying such architectures from a vast design space. We define an architecture as the constituent systems in the SoS and the connections among these systems. We seek to apply an agent-based modeling approach to create an architecture evaluation tool that leverages the modular nature of an SoS to quickly conduct trade studies and analyses to identify architectures that are resilient to a variety of threats. The simulation tool is built on the principles of the Open Systems Architecture put forth by the Department of Defense (DoD). We present an analysis that seeks to find a suitable SoS architecture configuration that maintains operability and performance during the loss of command authorities. This loss can be attributed to battle casualties or other forms of threats including communications failures. We demonstrate that distributing managerial control in the architecture improves, to some extent, SoS performance in terms of time to threat elimination. This single analysis demonstrates how SoS practitioners can use agent-based models to quickly construct and evaluate potential architectures against a variety of threats and disruptions, ultimately building resilient SoS architectures that perform to prescribed levels even in the face of network failures and attacks.

At present, the configuration of combat system-of-systems (SoS) equipment is highly complex, and the dynamic changes in the states of individual equipment during operations are difficult to predict. Traditional SoS effectiveness evaluation methods are not capable to keep up with these challenges. This study focuses on the OODA (Observation, Orientation, Decision, and Action) operational process, covering different stages such as reconnaissance, intelligence, planning and attacking. First, based on the DoDAF (Department of Defense Architecture Framework), a SoS architecture model is constructed to qualitatively describe the SoS's operational process from multiple perspectives. Second, the ADC (Availability-Dependability-Capability) model is used to calculate the effectiveness of individual equipment. Third, in combination with the Graphical Evaluation and Review Technique (GERT), quantitative calculations are performed to analyze the SoS 's effectiveness. Finally, the contribution of equipment to SoS effectiveness is calculated, as well as the impact of different network architecture in the kill chain on the SoS 's effectiveness, providing guidance for SoS optimization.

A System-of-Systems (SoS) is a collection of distributed independent individual systems that interact with one another to achieve an SoS capability requirement that cannot be achieved by individual systems alone (in contrast, in a monolithic system, hardware or software components are integrated to form a single entity). Interdependency between SoS systems, while enabling new capability, also means that failures can cascade throughout the SoS, creating development delays or additional system failures. Here we develop a method based on Bayesian networks to evaluate the resilience of SoS design alternatives to failures during operations. We propose a conditional resilience metric that measures each constituent system's contribution to overall SoS resilience, and a resilience pattern that shows how SoS performance degrades as systems fail. SoS resilience is determined by both the SoS architecture and the constituent system reliability. In a simple example based on the Littoral Combat Ship SoS, we determine the two most critical systems using the conditional resilience metric. Adding a communications link between these two systems increases the resilience, resulting in higher expected performance and slower expected performance degradation as a result of system failure. The conditional resilience and resilience pattern can be used by designers and other decision makers to identify SoS architectures and systems that increase SoS resilience to failure.

Architecture frameworks continue to evolve. The Unified Profile for the Department of Defense (DoD) Architecture Framework (DoDAF) and the UK's Ministry of Defence Architecture Framework (MODAF) (UPDM) provides a standard means of representing DoDAF, MODAF, and NATO Architecture Framework (NAF) conformant architectures using the Unified Modeling Language (UML), and Systems Modeling Language (SysML). Since the UPDM V2.0 publication, further information has emerged such as the June 2011 NATO study entitled: “Development of The AMN (Afghanistan Mission Network) Architecture In 2010 – Lessons Learned,” by Torsten Graeber of the NATO C3 Agency. This report identified the following in section 4.1‐ARCHITECTURE FRAMEWORKS, sub‐section 4.1.2 Observations (Need for a Unified Architecture Framework) and stated that:• differences in DoDAF, MODAF, and NAF make it difficult to match the metamodel one to one.• some of the concepts in the frameworks have the same name but different definitions, i.e. different semantics.• difficult to cross‐walk the concepts between the different frameworks leads to miscommunication between architects using different frameworks.Based on the above, the NATO Architecture Capability Team (Architecture CaT) meeting on Sept. 10‐11, 2012 committed to move to a single world‐wide Architecture Framework. Consequently, a new architecture framework profile supporting a unified framework is needed. It is intended that this framework bring the different architecture frameworks together. The UPDM V3.0 domain metamodel shall be derived from MODEM (the MODAF metamodel) and the DoDAF 2.0 metamodel (DM2), both of which are based upon the International Defence Enterprise Architecture Specification Foundation [IDEAS]. This paper will document the rationale behind the UPDM 3.0 as well as its new name of the Unified Architecture Framework (UAF).

Systems-of-Systems (SoS) is a set of independent software-intensive systems intended to support critical domains, such as emergency and crisis response systems and health. As such, SoS should exhibit robust behaviors and guarantee that the entire system is continuously available, not failing and leaving its users helpless. However, the SoS inherent dynamic architecture can potentially affect the availability of the functionalities being offered by the SoS. Hence, it is prominently important to anticipate, at design-time, the degree of availability an SoS can offer and provide mechanisms to reinforce the availability provided by it. Usually, software architects use static models to predict quality attributes in software architectures. However, these models are not well-succeeded to deal with the dynamic nature of the SoS architectures. The main contribution of this paper is thus presenting emerging results on the establishment of a simulation-based approach for empirically predicting the availability of SoS software architectures at design time. A Flood Monitoring SoS was used as the experimental platform. Preliminary results indicate that, specifically for the context of our study, the nature of availability in SoS architectures was not only influenced by the availability of the SoS constituents, but also by the SoS topology.

Systems-of-systems (SoS) are network-enabled synergistic collaborations between systems that are operationally and managerially independent, distributed, evolve dynamically and exhibit emergence. The design of dependable SoS requires model-based approaches that permit description of contracts between constituent systems at interfaces in a SoS architecture, including functionality and interaction behaviour, and that permit verification of global behaviours. We describe an approach to formal model-based SoS engineering using complementary notations for functional, interaction and architectural aspects. A case study in modelling information flow in an emergency response SoS demonstrates the viability of the proposed approach and highlights a need for common semantic foundations.

SApHESIA: An agent-based model and a criticality-based heuristic for cooperatively coupling SoSs

Systems of Systems (SoS) are complex systems composed of managerially and operationally independent constituent systems (CS). Smart cities are examples of SoS. However, these types of systems impose challenges to traditional software architecture such as highly evolutionary architecture due to the evolution of individual CS and emergent behaviors that results from the interoperability of CS. Executable Models (ExM) are a class of models that can be executed and that can assist on architectural design of SoS. By using them, architects can predict the SoS structure and behavior by visualizing and simulating the SoS still at design-time. On the other hand, as any other emergent technologies, it suffers with absent scientific evidences of its benefits, mainly in industrial context. The main contribution of this master thesis project intends to be the presentation of evidences about the use of ExM to solve problems in the SoS software architecture design. For achieving such purpose, this project is structured in well-defined steps: (i) a systematic mapping study, (ii) elaboration of a conceptual map derived from the mapping study, (iii) a survey to obtain perceptions from software engineering professionals on the use of ExM in the context of engineering simple systems and (iv) a new survey expanding the previous study and to obtain perceptions from software engineering professionals on the use of ExM in the context of architectural design of SoS. Preliminary results reveal that ExM have been reported in diverse SoS domains, offering a dynamic and interactive view to the SoS. We also found that software engineering professionals see ExM as suitable solution for dealing with complex and critical systems as SoS.

This paper is concerned with the engineering of systems that are themselves comprised of other component systems, and where each of the component systems serves organizational and human purposes.i¾ i¾ These component purposes may be locally managed and optimized independently, or nearly so, of the objectives to be met by the composite system.i¾ i¾ There are a number of inherent characteristics of these systems, and such related terms as systems of systems (SOS) or federations of systems (FOS) or federated systems of systems (F-SOS) are often used to characterize them.i¾ i¾ It is asserted that the resultant systems generally possess the characteristics of complex adaptive systems.i¾ i¾ We provide an overview of the literature describing these engineering efforts and provide plausible strategies for systems engineering and management of SOS and FOS that are based on the principles of a 'new federalism'.i¾ i¾ Finally, the implications of these plausible SOS and FOS systems engineering and management concepts are discussed with emphasis on evolutionary acquisition in the style of DoD and Intelligence Community related programs.

This article describes the emerging roles of the systems engineering (SE) function in supporting enterprise management in information‐intensive organizations. “Enterprise Systems Engineering” (ESE) comprises three major roles or “levels” of systems engineering for successful and efficient development or procurement of large complex systems of systems (SoS). While the authors' experience focuses predominantly on government organizations acting as their own SoS integrators, the SoS ESE concept has generic applicability for any organization, public or private, seeking to attain competitive advantage through leveraging of information technology resources and systems. The processes and tools described here have been developed and successfully employed to facilitate government project management and investment decisions and control. [Carlock and Decker, 1998] This paper describes a formal three‐level SoS ESE process that, at the top level, organizes and maintains all of the details of the enterprise‐wide SoS architecture and strategic development plan in a flexible framework that accommodates the changes expected over a long SoS evolution. The information maintained in this framework allows the organization to know where it is going, how and when it is going to get there, the required capabilities and interfaces of each SoS component, and the impact of changes to system requirements, budgets, schedules, etc., on the overall SoS. The middle level processes allow the organization to perform trade studies among alternative solutions to implement required capabilities based on what is best for the enterprise‐wide SoS rather than just local considerations. The end result of the middle level processes is a selected and approved solution and its associated cost, schedule, benefits, and technical baselines. The third level processes implement the approved solutions in accordance with the approved baselines. © 2001 John Wiley & Sons, Inc. Syst Eng 4: 242–261, 2001