QPRC 2016

Continuous Reliability Growth Modeling Using a Grey Systems Model

Tom Talafuse, Ed Pohl

University of Arkansas

Abstract 

When performing system-level developmental testing, time and expenses generally warrant a small sample size for failure data.  Upon failure discovery, corrective actions can be implemented to improve system reliability.  Current methods for estimating reliability growth, namely the AMSAA model, stipulate that parameter estimates have a great level of uncertainty when dealing with small sample sizes.  For purposes of handling limited failure data, we propose use of a modified GM(1,1) model to predict system reliability growth parameters and investigate how parameter estimates are affected by systems following a poly-Weibull failure intensity function.  Monte-Carlo simulation results are used to compare the accuracy of the GM(1,1) model to the AMSAA growth model.  With small sample sizes, the GM(1,1) model shows greater accuracy than the AMSAA model for prediction of growth model parameters.