題 目：Product Design Optimization under Uncertainties
主講人：Rong Pan副教授 （Arizona State University）
時(shí) 間：2012年6月12日 下午5點(diǎn)
地 點(diǎn)：主樓216會(huì)議室
主講人簡(jiǎn)介：
Dr. Rong Pan is an Associate Professor of Industrial Engineering at the School of Computing, Informatics, and Decision Systems Engineering of Arizona State University. He received his doctoral degree from Penn State University in 2002. His research interests include reliability data analysis, design of experiments, multivariate statistical quality control, time series analysis and control. He is the recipient of 2008 and 2011 Stan Ofsthun Awards from the Society of Reliability Engineers. He is a senior member of ASQ and IIE, and a member of SRE and IEEE.
內(nèi)容簡(jiǎn)介：
In this talk I will present some recent research in the area of product design optimization under uncertainties. I will discuss how different types of uncertainties can be modeled and integrated into the design optimization process. Specifically, in the formulation of reliability-based design optimization (RBDO) design feasibility is given as a probabilistic constraint, which is used to guarantee the product’s reliability or safety. The research presented here includes two approaches to address the RBDO problem. First, a sequential sampling strategy is developed to improve the solution of RBDO with implicit constraint functions. The Kriging model and the sequential sampling strategy are evaluated and compared with other methods that deal with the model uncertainty associated with the implicit constraint function in RBDO. Second, a Monte Carlo simulation-based reliability assessment method, particle splitting method, is developed for assessing the probabilistic constraint in RBDO. I will give a brief introduction of this Monte Carlo approach to the rare event probability evaluation and discuss how it can be applied on design optimization.
