Using value of information in quantitative set‐based design. Issue 6 (6th August 2021)

• Record Type: Journal Article
• Authors: Shallcross, Nicholas J.; Parnell, Gregory S.; Pohl, Ed; Goerger, Simon R.
• Is Part Of: Systems engineering. Volume 24:Issue 6(2021)
• Page Start: 439
• Page End: 455
• Publication Date: 2021-08-06
• Subjects: model‐based engineering -- multiobjective decision analysis -- multiobjective value of information -- program management -- set based design; Systems engineering -- Periodicals; 620.0011
• DOI: 10.1002/sys.21595
• Languages: English
• ISSNs: 1098-1241

Abstract

Abstract: Increasing system complexity requires that engineers, systems analysts, and program managers use comprehensive design methodologies to deliver affordable and resilient designs. One method is set‐based design (SBD), a product development and managerial process distinctly suited for developing complex systems under uncertainty. SBD simultaneously develops, analyzes, and matures numerous potential design sets, enabling the identification of high‐value, affordable, and resilient designs. Published SBD research is a rich source of both qualitative and quantitative methods. This research specifically focuses on quantitative SBD methods to apply a value of information (VOI) methodology enabling design convergence and selection. We build upon previous SBD research to enable design maturation and uncertainty reduction. Our methodology integrates design maturation and multiobjective VOI analysis into a comprehensive quantitative SBD process to guide system development from initial design concepts to the pre‐production design decision. In doing so, we also provide refinements and process improvements to existing quantitative SBD methods. We demonstrate our methodology with a model‐based UAV design case study using an integrated suite of system, value, and cost models. Our case study specifically focuses on the design maturation and model selection decisions enabling design space convergence. We compare our current results with those from a previous UAV case study, achieving a 41% reduction in required computation time for design space convergence. These results highlight the methodology's ability to reduce program risk and potential to improve SBD convergence efficiency.