Model-based system-of-systems engineering for space-based command, control, communication, and information architecture design

This dissertation presents a model-based system-of-systems engineering (SoSE) approach as a design philosophy for architecting in system-of-systems (SoS) problems. SoS refers to a special class of systems in which numerous systems with operational and managerial independence interact to generate new capabilities that satisfy societal needs. Design decisions are more complicated in a SoS setting. A revised Process Model for SoSE is presented to support three phases in SoS architecting: defining the scope of the design problem, abstracting key descriptors and their interrelations in a conceptual model, and implementing computer-based simulations for architectural analyses. The Process Model enables improved decision support considering multiple SoS features and develops computational models capable of highlighting configurations of organizational, policy, financial, operational, and/or technical features. Further, processes for verification and validation of SoS models and simulations are also important due to potential impact on critical decision-making and, thus, are addressed. Two research questions frame the research efforts described in this dissertation. The first concerns how the four key sources of SoS complexity—heterogeneity of systems, connectivity structure, multi-layer interactions, and the evolutionary nature—influence the formulation of SoS models and simulations, trade space, and solution performance and structure evaluation metrics. The second question pertains to the implementation of SoSE architecting processes to inform decision-making for a subset of SoS problems concerning the design of information exchange services in space-based operations domain. These questions motivate and guide the dissertation’s contributions. A formal methodology for drawing relationships within a multi-dimensional trade space, forming simulation case studies from applications of candidate architecture solutions to a campaign of notional mission use cases, and executing multi-purpose analysis studies is presented. These efforts are coupled to the generation of aggregate and time-dependent solution performance metrics via the hierarchical decomposition of objectives and the analytical recomposition of multi-attribute qualitative program drivers from quantifiable measures. This methodology was also applied to generate problem-specific solution structure evaluation metrics that facilitate the comparison of alternate solutions at a high level of aggregation, at lower levels of abstraction, and to relate options for design variables with associated performance values. For proof-of-capability demonstration, the selected application problem concerns the design of command, control, communication, and information (C3I) architecture services for a notional campaign of crewed and robotic lunar surface missions. The impetus for the work was the demonstration of using model-based SoSE for design of sustainable interoperability capabilities between all data and communication assets in extended lunar campaigns. A comprehensive Lunar C3I simulation tool was developed by a team of researchers at Purdue University in support of NASA’s Constellation Program; the author of this dissertation was a key contributor to the creation of this tool and made modifications and extensions to key components relevant to the methodological concepts presented in this dissertation. The dissertation concludes with a presentation of example results based on the interrogation of the constructed Lunar C3I computational model. The results are based on a family of studies, structured around a trade-tree of architecture options, which were conducted to test the hypothesis that the SoSE approach is efficacious in the information-exchange architecture design in space exploration domain. Included in the family of proof-of-capability studies is a simulation of the Apollo 17 mission, which allows not only for partial verification and validation of the model, but also provides insights for prioritizing future model design iterations to make it more realistic representation of the “real world.” A caveat within the results presented is that they serve within the capacity of a proof-of-capability demonstration, and as such, they are a product of models and analyses that need further development before the tool’s results can be employed for decision-making. Additional discussion is provided for how to further develop and validate the Lunar C3I tool and also to make it extensible to other SoS design problems of similar nature in space exploration and other problem application domains