Reliability-Based Design Optimization with Confidence Level for Non-Gaussian Distributions Using Bootstrap Method

Abstract

For reliability-based design optimization (RBDO), generating an input statistical model with confidence level has been recently proposed to offset the inaccurate estimation of the input statistical model with Gaussian distributions. To do this, the confidence intervals of mean and standard deviation are calculated using the Gaussian distributions of input random variables. However, if the input random variables are non-Gaussian, use of the Gaussian distributions of input variables will provide inaccurate confidence intervals and will yield an undesirable confidence level of the reliability-based optimum design meeting the target reliability. In this paper, the RBDO method using the bootstrap method, which does not use the Gaussian distributions of input variables to calculate the confidence intervals of mean and standard deviation, is proposed to obtain the desirable confidence level of output performance for non-Gaussian distributions.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 2010
Accession Number
ADA558432

Entities

People

  • David Gorsich
  • Ikjin Lee
  • Kyung K. Choi
  • Yoojeong Noh

Organizations

  • University of Iowa

Tags

DTIC Thesaurus Topics

  • Coefficients
  • Data Sets
  • Distribution Functions
  • Engineering
  • Estimators
  • Experimental Data
  • Fatigue Life
  • Gaussian Distributions
  • Identification
  • Industrial Engineering
  • Intervals
  • Monte Carlo Method
  • Probability
  • Probability Distributions
  • Random Variables
  • Sampling
  • Simulations

Fields of Study

  • Engineering

Readers

  • Approximation Theory.
  • Computational Modeling and Simulation
  • Life Cycle Cost Analysis