A model-based systems engineering framework for concept development

Abstract

The development of increasingly complex, innovative systems under greater constraints has been the trend over the past several decades. In order to be successful, organizations must develop products that meet customer needs more effectively than the competitors' alternatives. The development of these concepts is based on a broad set of stakeholder objectives, from which alternative designs are developed and compared. When properly performed, this process helps those involved understand the benefits and drawbacks of each option. This is crucial as firms need to effectively and quickly explore many concepts, and easily determine those most likely to succeed. It is generally accepted that a methodical design approach leads to the reduction in design flaws and cost over a product's life cycle. Several techniques have been developed to facilitate these efforts. However, the traditional tools and work products are isolated, and require diligent manual inspection. It is expected that the effectiveness of the high-level product design and development will improve dramatically through the adoption of computer based modeling and simulation. This emerging capability can mitigate the challenges and risks imposed by complex systems by enforcing rigor and precision. Model-based systems engineering (MBSE) is a methodology for designing systems using interconnected computer models. The recent proliferation of MBSE is evidence of its ability to improve the design fidelity and enhance communication among development teams. Existing descriptions of leveraging MBSE for deriving requirements and system design are prevalent. However, very few descriptions of model-based concept development have been presented. This may be due to the lack of MBSE methodologies for performing concept development. Teams that attempt a model-based approach without well defined, structured strategy are often unsuccessful. However, when MBSE is combined with a clear methodology, designs can be more efficiently generated and evaluated. While it may not be feasible to provide a "standard" methodology for concept development, a framework is envisioned that incorporates a variety of methods and techniques. This thesis proposes such a framework and presents an example based on a simulated concept development effort.