Multidisciplinary Analysis and Design Tools For Uncertainty Modeling

Abstract

Design optimization methods have found increased application in problems of aerospace design.The interdisciplinary nature of this design problem has given rise to the field of multidisciplinary design optimization that focuses on the integration of numerically efficient analysis methods with reliable and robust design optimization algorithms.The focus of research has increasingly shifted towards the study of modeling and handling of uncertainty in different facets of the design problem. Uncertainties add to both the computational complexity and to numerical costs in multidisciplinary design.The paper discusses recent developments in computational tools for better managing the inclusion of imprecise information, including methods through which to quantify uncertainty, new formulations of the design problem that inherently contribute to lower computational costs, and numerically efficient algorithms for computing reliability measures in multidisciplinary design. The coupled nature of analysis in the design process is examined from the perspective of decomposition-based design, and appropriate problem formulations for these cases are presented.