System of Systems Analyses Using Model Based Systems Engineering

Steven A Zusack, Purdue University


This study aims to address two major challenges faced at early stages of development, commonly called Pre-Phase A, for complex space-based mission architectures. The first challenge is a lack of detailed system information when an assessment of competing architecture concepts is needed. To counter this challenge, the Systems Operational Dependency Analysis and architecture-level parametric cost methods are used. These methods produce meaningful evaluation metrics for direct comparison of architecture concepts by utilizing both limited, high-level information as well as any detailed, low-level information that may be available. The second challenge is the ability to quickly and efficiently alter, add, or remove systems or system-attributes from a concept architecture and reevaluate using the afore mentioned methods. To counter this challenge, the Model Based Systems Engineering approach is utilized. With this approach, a single Mars concept architecture is created and quickly altered to include a second Mars variant. After that, the model is expanded to include two Moon-based variants followed by a connecting architecture with both lunar and martian missions. This task is demonstrated much more quickly than could have been possible using traditional, document-based systems engineering. The output of the tools is valuable to guide future research and concept evaluation. However, it is important to note the results are based on high-level information and expert opinion and do not claim a level of accuracy expected from Probabilistic Risk Assessment or industry-level cost methods. The focus of this study is the demonstration of a combination of tools to counter these two common challenges of early architecture development.




DeLaurentis, Purdue University.

Subject Area

Engineering|Aerospace engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server