Abstract:Reliability sensitivity analysis is used to find the rate of change in the probability of failure (or reliability) due to the changes in distribution parameters such as the means and standard deviations. Most of the existing reliability sensitivity analysis methods assume that all the probabilities and distribution parameters are precisely known. That is, every statistical parameter involved is perfectly determined. However, there are two types of uncertainties, epistemic and aleatory uncertainties that may not be perfectly determined in engineering practices. In this paper, both epistemic and aleatory uncertainties are considered in reliability sensitivity analysis and modeled using P-boxes. The proposed method is based on Monte Carlo simulation (MCS), weighted regression, interval algorithm and first order reliability method (FORM). We linearize original non-linear limit-state function by MCS rather than by expansion as a first order Taylor series at most probable point (MPP) because the MPP search is an iterative optimization process. Finally, we introduce an optimization model for sensitivity analysis under both aleatory and epistemic uncertainties. Four numerical examples are presented to demonstrate the proposed method.

Study on confidence interval estimation of asymptotic method for sensitivity test

Interval estimation of the two-parameter exponential constant stress accelerated life test model under Type-II censoring

On the Asymptotic Analysis of Quantile Sensitivity Estimation by Monte Carlo Simulation.

Effect Study Of D-Optimality-Based Sensitivity Test'S Distrubuting For Reliability Evaluation

Effect of Distribution Choice on Reliability Evaluation of D-Optimality-Based Sensitivity Test

Study on the Influence of Instrument Resolution on Estimates Precision of Sensitivity Test

An efficient method based on Bayes’ theorem to estimate the failure-probability-based sensitivity measure

Estimation and Its Applicability of D-Optimality-Based Sensitivity Test under Different Prior Information

An Estimate of a Change Point in Variance of Measurement Errors and Its Convergence Rate

Studies on Estimation Accuracy of Calculation Methods for Standard Deviation in Up-and-down Method

A new efficient simulation method based on Bayes' theorem and importance sampling Markov chain simulation to estimate the failure-probability-based global sensitivity measure

Calculation of confidence intervals for simulation output

Studies on Estimation Accuracy of Point Estimation in Up-And-Down Method

Applicability Analysis of Sensitivity Test Method

Reliability sensitivity analysis for structural systems in interval probability form

Reliability-oriented global sensitivity analysis using subset simulation and space partition

Asymptotic normality of LAD estimates in two-tailed Tobit regression models

Estimation Method of Software Reliability for Safety-critical System

Pathwise Estimation of Probability Sensitivities Through Terminating or Steady-State Simulations

Sample Size Uncertainty Analysis in the System Vulnerability Assessment Results for IEMI

Improved Confidence Regions in Meta-analysis of Diagnostic Test Accuracy