Charles E. Dickerson

A Case for an International Consortium on System-of-Systems Engineering

A system-of-systems (SoS) conceptualization is essential in resolving issues involving heterogeneous independently operable systems to achieve a unique purpose. Successful operation as an SoS requires communication among appropriate individuals and groups across enterprises through an effective protocol. This paper presents a position on the creation of a consortium of concerned system engineers and scientists worldwide to examine the problems and solutions strategies associated with SoS. The consortium could lead efforts in clarifying ambiguities and in seeking remedies to numerous open questions with respect to SoS analysis, SoS engineering (SoSE), as well as differences between systems engineering (SE) and SoSE. The mission of this consortium is envisioned to: 1) act as a neutral party; 2) provide a forum to put forth Calls to Action; and 3) establish a community of interest to recommend a set of solutions.

A Brief History of Models and Model Based Systems Engineering and the Case for Relational Orientation

Models are at the heart of science and engineering. Model-based approaches to software development and systems engineering use technologies to include graphical modeling languages, such as the Systems Modeling Language, that support system design and analysis through machine readable models. This paper traces key historical contributions of software and systems engineers over the past five decades to show a coherent concept of models and how they can be used for software and systems engineering. Recent model-based systems engineering methodologies supported by commercially available modeling tools are also summarized. Relational orientation is seen to be the underlying viewpoint that expresses and binds these approaches. Relational orientation for systems engineering (ROSE) is then specified using a general systems methodology. Systems are seen to access each others models in ROSE much like classes in object orientation access each others objects. Object-oriented frames for software engineering are extended to relational frames to specify an innovative framework for system design and analysis. This generalizes the axiomatic design approach of N. P. Suh. A repeatable procedure supporting greater concurrency between design and verification is also demonstrated for searching the solution space in linear axiomatic design.

Relational Oriented Systems Engineering (ROSE): Preliminary report

A new viewpoint on Architecture and Systems Engineering is offered based on domains and mathematical relationships that is concordant with and expressive of the IEEE and INCOSE definition of ‘system’ and the definition of ‘system architecture’ proposed by ISO working group JTC1/SC7/WG42. In this viewpoint, referred to as Relational Oriented Systems Engineering (ROSE), relational models and transformations are the means for abstracting systems engineering concepts and information. Hierarchical paradigms such as ‘definition and decomposition’ can be expressed more precisely by the principle of ‘model specification and relational transformation’. ROSE can be applied to any systems approach to provide an expression of the approach that is model-based. Ultimately this is a model-based viewpoint that should bring precision with agility to the practice of Systems and Systems of Systems Engineering and a more holistic understanding of the role of Architecture.

Towards a logical and scientific foundation for system concepts, principles, and terminology

This paper provides preliminary results of an approach that is being used to support the efforts to standardize concepts and terminology for architecture and systems by the International Organization for Standardization (ISO) convened under working group JTC1/SC7/WG42. The approach is based on logic and the methodology of science. The results provide: (i) a demonstration of the proposed modeling approach using different definitions of system and related systems principles, (ii) a scientific interpretation of the logical model offered for system that gives further validation of the approach and reveals how system element interactions in Hamiltonian mechanics can be modeled as the transport of energy between elements, (iii) giving a sharper meaning to the engineering definition of System that (iv) extends the meaning of interactions in the INCOSE definition of system. This approach should be also useful for reasoning about definitions for systems of systems.

Using relational model transformations to reduce complexity in SoS requirements traceability: Preliminary investigation

The principles and methods of Model Driven Architecture are applied to the problem of requirements traceability for a System of Systems (SoS). Model transformations of operational threads are used to reduce the complexity of modeling mission requirements and their flow into the architecture of the SoS. The allocation of requirements to operational mission threads (OMTs) rather than to individual systems reduces the complexity of the requirements tracing. Relational transformations provide a mathematically based formalism for model transformations that permit precise computation of the transformation of operational threads into threads of systems allocated from the SoS. Connectivity requirements for the SoS are also exposed in this way and the number of permissible system threads are seen to correspond directly to the number of permissible transformations. The principles and methods are illustrated by an elementary case study for sensor fusion.

Relational-Oriented Systems Engineering and Technology Tradeoff Analysis Framework

In this paper, a new framework for performing early technology tradeoff and design studies, the relational-oriented systems engineering and technology tradeoff analysis framework, is developed and applied to an initial case study to conduct a trade between two candidate technologies for potential application on a commercial jet. Relational-oriented systems engineering and technology tradeoff analysis leverages the relational-oriented systems engineering methodology coupled with the exploitation of transformations used in modeling and simulation to create a direct association between the quality function deployment methodology and standard quantitative conceptual design space exploration techniques leveraged in technology forecasting and trade studies. This association brings precision to quality function deployment that is model driven and mathematically founded. The approach highlights key deficiencies in quality function deployment when applied to early phase design and technology tradeoff studies for t...

A Relational Oriented Approach to System of Systems Assessment of Alternatives for Data Link Interoperability

ROSE is applied to a large-scale SoS tactical data link interoperability problem. A model-driven framework structure developed using the ROSE methodology is employed to prescribe a repeatable approach for determining viable candidate solutions that complete and make rigorous a previous capability-based exploratory analysis performed by the Office of the Chief Engineer of the U.S. Navy. This novel and efficient approach to a long-standing problem concentrates on the relationships between models to provide a framework and factorization of an SoS architecture for portfolio selection and evaluation. The approach is demonstrated in a simplified, but end-to-end, case study derived from the original data link interoperability analysis. The abstract approach employed can be applied to a much wider class of problems than data link interoperability.

Formal methods for a system of systems analysis framework applied to traffic management

Formal methods for systems and system of systems engineering (SoSE) can bring precision to architecting and design, and increased trustworthiness in verification; but they require the use of formal languages that are not broadly comprehensible to the various stakeholders. The evolution of Model Based Systems Engineering (MBSE) using the Systems Modeling Language (SysML) lies in a middle ground between legacy document-based SoSE and formal methods. SysML is a graphical language but not a formal language. Initiatives in the Object Management Group (OMG), such as the development of the Foundational Unified Modeling Language (fUML) seek to bring precise semantics to object-oriented modeling languages. Following the philosophy of fUML, we offer a framework for associating precise semantics with Unified Modeling Language (UML) and SysML models essential for SoSE architecting and design. Straightforward methods are prescribed to develop the essential models and to create semantic transformations between them. Matrix representations can be used to perform analyses that are concordant with the system of UML or SysML models that represent the system or SoS. The framework and methods developed in this paper are applied to a Traffic Management system of systems (TMSoS) that has been a subject of research presented at previous IEEE SoSE conferences.

Relational oriented systems engineering framework for flight training

The integration of systems of systems (SoS) associated with a flight training mission directly reflects the problem of developing a system engineering process for the design of live, virtual and constructive (LVC) experiments. Due to the complexity and disparity of the technology in a flight training SoS (FTSoS), modeling and analysis of architecture is becoming increasingly important. Relational Oriented Systems Engineering (ROSE) methodology is used to develop a framework for simulation and analysis of a navigational SoS for a typical aircraft. The framework can be used for both the prescription of navigation systems entering and exiting the SoS and for the analysis of pilot behavior as navigation quality of service (QoS) changes. ROSE offers a novel approach to developing a model-based systems engineering (MBSE) process for simulation and analysis of a complex SoS problem.

Mathematical foundations for relational oriented systems engineering (ROSE)

Relational oriented systems engineering (ROSE) is a viewpoint on systems engineering that is based on domains and mathematical relations. It is a model-based formalism that integrates and extends the work of Yi Lin and George Klir on relational structures and relational homomorphism in systems engineering. It is also appropriate from the viewpoint of universal algebra. The mathematical formalism also admits a graphical interpretation. The mathematical foundation for ROSE supports the rigorous development of structures for the design of systems and the assemblage of systems of systems that extend the methods of Nam Suh on axiomatic design theory. The ROSE formalism is rooted in the Tarski theory of models in mathematical logic but has been adapted to the practice of engineering. The various mathematical viewpoints that comprise ROSE are seen to comprise a coherent foundation for a relational orientation on systems and system of systems engineering.