Robust Design Optimization
Boris Kramer
University of California San Diego
Mechanical and Aerospace Engineering
Reliable, robust and risk-averse design of complex engineering systems with optimized performance requires dealing with uncertainties. A conventional approach is to add ad-hoc safety margins to a design that was obtained from deterministic optimization. Safer engineering designs require appropriate cost and constraint function definitions that capture statistics and distributional information about the unwanted system behavior, which becomes stochastic in the presence of uncertainties. In this tutorial, we will start by discussing the modeling of uncertainties in complex systems, and then present various approaches to design systems in the presence of uncertainties, such as risk-based design optimization (RiBDO), reliability-based design optimization (RBDO) as well as robust design optimization (RDO). For each approach, we will offer several areas of active research and established methods to implement such.