Problem Formulation for Multidisciplinary Optimization

TLDR

Multidisciplinary design optimization couples multiple analysis disciplines with numerical optimization, and the key distinction among the three fundamental MDO formulations is the disciplinary feasibility that must be preserved at each optimization iteration. The paper sets out three goals: to introduce MDO to optimization researchers, to present a new abstraction and decomposition formulations for multidisciplinary analysis and design, and to propose individual‑discipline‑feasible approaches that enable coarse‑grained parallelism on heterogeneous computing environments. The authors develop individual‑discipline‑feasible (IDF) methods that leverage existing specialized analysis codes and incorporate sensitivity calculations, thereby facilitating coarse‑grained computational parallelism. The discussion reveals trade‑offs among reusing existing software, computational demands, and the likelihood of successful problem resolution.

Abstract

This paper is about multidisciplinary (design) optimization, or MDO, the coupling of two or more analysis disciplines with numerical optimization. The paper has three goals. First, it is an expository introduction to MDO aimed at those who do research on optimization algorithms, since the optimization community has much to contribute to this important class of computational engineering problems. Second, this paper presents to the MDO research community a new abstraction for multidisciplinary analysis and design problems as well as new decomposition formulations for these problems. Third, the “individual discipline feasible” (IDF) approaches introduced here make use of existing specialized analysis codes, and they introduce significant opportunities for coarse-grained computational parallelism particularly well suited to heterogeneous computing environments. The key distinguishing characteristic of the three fundamental approaches to MDO formulation discussed here is the kind of disciplinary feasibility that must be maintained at each optimization iteration. Other formulation issues, such as the sensitivities required, are also considered. This discussion highlights the trade-offs between reuse of existing software, computational requirements, and probability of success.