Integrated DEsign Automation Laboratory

Integrated DEsign Automation Laboratory

Design Under Uncertainty

In this research area, the principles of robust design and reliability design are integrated to develop generic probabilistic approaches to engineering design under uncertainty. Our research has been focused on all aspects of uncertainty quantification, uncertainty propagation, and probabilistic optimization.  We have been investigating efficient algorithmic approaches that can reduce fundamental computational complexity for uncertainty propagation of practical problems with expensive simulations and/or large dimensions. Various sources of uncertainties in simulation-based design have been investigated. An integrated methodology for the propagation and management (mitigation) of uncertainties associated with model inputs (design variables and noise factors) as well as the uncertainty of model itself has been developed.  The statistical sensitivity analysis (SSA) techniques have been investigated to manage the complexity in design under uncertainty.

Commercialized software code:  The Sequential Optimization and Reliability Assessment (SORA) method (Du and Chen 2004) provides an effective strategy for decoupling probabilistic analyses from optimization loops to improve the efficiency of probabilistic optimization.  The method has been integrated into Hyperworks®, an optimization software developed by Altair Engineering.

Representative Papers

vide.gif

Integrated DEsign Automation Laboratory

McCormick School of Engineering Home | McCormick Chinese (中文网站) | Contact Us | Maps
Northwestern Home | Northwestern Calendar | Emergency Plan | Legal and Policy Statements
© 2012 Robert R. McCormick School of Engineering and Applied Science, Northwestern University