Designing Engineered Resilient Systems Using Set-Based Design

Abstract

Designing resilience into a conceptual system has proven challenging to quantify and difficult to evaluate. Designing for resilience requires increasing complexity, requirements, and advancements in modeling capabilities. This paper details the use of set-based design to explore design concepts and identify designs with the potential to be resilient systems. Set-based design is a technique that explicitly considers sets of design alternatives with some common design choices. Contrary to point-based design, which selects a single concept and architecture, set-based design selects a set composed of a concept with multiple potential architectures, allowing multiple architectures to advance in the design process. With multiple architectures progressing, set-based design retains the flexibility to reevaluate affordability as requirements are refined and test data is generated. For the analysis, value and cost models are constructed using multiple objective decision analysis to define the value verses cost tradespace. The practice of set-based design expands the tradespace by generating thousands of alternatives across the design space, each alternative relatable in the tradespace by the alternative’s associated and unique cost and value. This paper demonstrates the feasibility of set-based design to evaluate system resilience. Specifically, this paper quantifies and evaluates mission resilience. Mission resilience is the value of performance maintained and recovered by the system in the face of potential performance losses suffered under a threat to the system. These concepts are applied to a military squad enhancement portfolio of systems illustration. Mission resilience value models are used in conjunction with cost models to determine affordability over the life cycle of the enhancements. Set-based design is used to explore the value versus cost tradespace, and a technique called iterative set-based design is employed to focus the tradespace on efficient and affordable solutions for further exploration. The results of this illustration identify solutions on the efficient frontier using an integrated trade-off analytics framework adaptable to requirements changes and able assess the impact of a change in near real time.