Massachusetts Institute of Technology
The proposed research will model the relationship between design and requirements in SysML and propose a methodology by which this relationship can be studied and adverse effects can be avoided. The biggest challenge in early-lifecycle systems engineering is deriving a relationship between the benefit and cost of possible spacecraft designs. These metrics are difficult to quantify early in a project because the intricacies of each possible design have not been explored. Underestimation of the complexity of the final system or optimistic estimations of technology infusion effort can lead to descoping capability or overrunning on cost. Gaining a greater understanding of detailed design characteristics early in a project will contribute to reducing cost overruns and making technology infusion less costly.
One source of complexity is the relationship between requirements and design. In the systems engineering V-model, requirements development precedes and drives the design process. However, on real projects, design choices frequently force changes of requirements causing redesign and cost increases. This feedback relationship is difficult to track with traditional document-based methods due to its conditional nature and the cascading nature of changes in an integrated system. However, model-based systems engineering (MBSE) techniques and specifically SysML are able to model this relationship. Through successful development of the proposed methodology, systems engineers will be more aware of the full ramifications of their design choices early in a project and projects will be better able to make reliable cost and schedule estimations. With a better understanding of the true cost/benefit relationship, NASA can be more confident in taking risk and focus on missions which provide the biggest benefit to the nation.