[Review] Systems of Systems Engineering: Basic Concepts, Model-Based Techniques, and Research Directions
This is a general description of Systems of Systems (SoS). It is a huge amount of 41 pages, but the most important thing is to understand the SoS definition and the 8 dimensions that it has. So I wrote only the definitions of SoS and 8 dimensions.
SoS is part of system engineering and deals with boundaries such as interface, integration, testing, and communication between independent and distributed CSs.
The arrangement of the theoretical systems and constructs in a hierarchy of complexity.
A set of connections or associated systems to produce results that could not have been achieved with individual systems
Managerial independence: self-governance and management of cooperation.
Geographic distribution: Each system is remote.
Evolutionary development: can be continually changed by additions, modifications, and eliminations. It is never fixed.
Emergent behavior: Collaborative work that an individual system can not do.
Belonging: systems functionally cooperate for a greater purpose.
Connectivity: Synergy through highly dynamic distributed networks
Diversity: Each part is heterogeneous, autonomous, and free to develop.
Emerging: Each part cooperates to do something new.
Autonomy is the degree to which CS's behavior is governed by its own rules, not by the outside. Individual ownership. Modeling and analytical techniques should allow autonomous behavioral representation of CSs even if accurate prediction at the SoS level is difficult.
2. Independence
Independence is the ability to operate when the CS is off the SoS. This means that CS can play an unrelated role as well as a role related to SoS. So the information may be hidden by SoS engineers, so Model-based techniques should be able to support information hiding.
3. Distribution
Distribution indicates that CS is distributed and communication and information sharing should be possible. Modeling frameworks that support decentralization should be capable of assigning CS to the computational infrastructure to which the communications medium is connected.
4. Evolution
SoS is long-lasting and changing. Evolutionary development, lack of a permanent state. A model-based approach should ensure the preservation of specific functions in the evolutionary phase. As one CS evolves, it may be necessary to re-verify the interface with which it should communicate with the other CS.
5. Dynamic reconfiguration
Dynamic reconfiguration is the ability for a SoS to be able to replace structures and components without planned intervention. Real-time change to ensure resilience to breakdowns and other threats. This is the ability of the SoS to change the configuration while it is running. On-the-fly swap-in and a pluggable architecture.
6. Emergence of behavior
Emergence is the action resulting from the mutual cooperation of CSs. Modeling and analysis tools should allow statement and verification of emergence and identification of emergent behavior.
7. Interdependence
They are interdependent to each other to achieve the common goal of SoS. Sometimes they sacrifice their actions for SoS. CS are independent but interdependent. Modeling and analytical techniques should have a clear identification of interactions, tracking of interdependencies, and the ability to evaluate the impact of CS changes using these links.
8. Interoperability
Interoperability is the ability of SoS to include heterogeneous CS. This includes integration, interface acceptance, protocols, standards, and so on. Modeling and analysis techniques should support conformance testing of the CS interface.
Reference
Nielsen, Claus Ballegaard, et al. "Systems of systems engineering: basic concepts, model-based techniques, and research directions." ACM Computing Surveys (CSUR) 48.2 (2015): 18.
SoS Definition
System of Systems (SoS) is a system made up of independent constituent systems (CS).SoS is part of system engineering and deals with boundaries such as interface, integration, testing, and communication between independent and distributed CSs.
The arrangement of the theoretical systems and constructs in a hierarchy of complexity.
A set of connections or associated systems to produce results that could not have been achieved with individual systems
5 principal features by Maier (OMGEE)
Operational independence: All systems that are part of SoS operate independently.Managerial independence: self-governance and management of cooperation.
Geographic distribution: Each system is remote.
Evolutionary development: can be continually changed by additions, modifications, and eliminations. It is never fixed.
Emergent behavior: Collaborative work that an individual system can not do.
5 characteristics by Boardman and Sauser (ABCDE)
Autonomy: Each system is autonomous, independent.Belonging: systems functionally cooperate for a greater purpose.
Connectivity: Synergy through highly dynamic distributed networks
Diversity: Each part is heterogeneous, autonomous, and free to develop.
Emerging: Each part cooperates to do something new.
8 dimensions of SoS
1. Autonomy of constituentsAutonomy is the degree to which CS's behavior is governed by its own rules, not by the outside. Individual ownership. Modeling and analytical techniques should allow autonomous behavioral representation of CSs even if accurate prediction at the SoS level is difficult.
2. Independence
Independence is the ability to operate when the CS is off the SoS. This means that CS can play an unrelated role as well as a role related to SoS. So the information may be hidden by SoS engineers, so Model-based techniques should be able to support information hiding.
3. Distribution
Distribution indicates that CS is distributed and communication and information sharing should be possible. Modeling frameworks that support decentralization should be capable of assigning CS to the computational infrastructure to which the communications medium is connected.
4. Evolution
SoS is long-lasting and changing. Evolutionary development, lack of a permanent state. A model-based approach should ensure the preservation of specific functions in the evolutionary phase. As one CS evolves, it may be necessary to re-verify the interface with which it should communicate with the other CS.
5. Dynamic reconfiguration
Dynamic reconfiguration is the ability for a SoS to be able to replace structures and components without planned intervention. Real-time change to ensure resilience to breakdowns and other threats. This is the ability of the SoS to change the configuration while it is running. On-the-fly swap-in and a pluggable architecture.
6. Emergence of behavior
Emergence is the action resulting from the mutual cooperation of CSs. Modeling and analysis tools should allow statement and verification of emergence and identification of emergent behavior.
7. Interdependence
They are interdependent to each other to achieve the common goal of SoS. Sometimes they sacrifice their actions for SoS. CS are independent but interdependent. Modeling and analytical techniques should have a clear identification of interactions, tracking of interdependencies, and the ability to evaluate the impact of CS changes using these links.
8. Interoperability
Interoperability is the ability of SoS to include heterogeneous CS. This includes integration, interface acceptance, protocols, standards, and so on. Modeling and analysis techniques should support conformance testing of the CS interface.
Reference
Nielsen, Claus Ballegaard, et al. "Systems of systems engineering: basic concepts, model-based techniques, and research directions." ACM Computing Surveys (CSUR) 48.2 (2015): 18.
Comments
Post a Comment