Title :
Modeling strategies for reparable systems having multi-aging parameters
Author :
Singh, Jagmeet ; Vittal, Sameer ; Zou, Tong
Author_Institution :
Energy Div., Gen. Electr. Co., Greenville, SC
Abstract :
In this paper we study different modeling strategies for reparable systems, that have more than one aging parameters. We propose four approaches that use Monte-Carlo simulations, proportional hazards models, multi-variate Weibull along with numerical optimization methods and maximum likelihood estimation. The objective of this study is to investigate modeling strategy that predicts system´s reliability and availability more accurately and to see if different strategies fit well on different systems based on the interactions among aging parameters. This study will provide a comparison of modeling strategies to estimate reliability and availability of reparable systems.
Keywords :
Monte Carlo methods; Weibull distribution; hazards; maintenance engineering; maximum likelihood estimation; reliability theory; Monte-Carlo simulation; maximum likelihood estimation; multiaging parameter; multivariate Weibull model; numerical optimization method; proportional hazards model; reparable system; system availability; system reliability; Aging; Availability; Hazards; History; Mathematical model; Maximum likelihood estimation; Monte Carlo methods; Optimization methods; Predictive models; Runtime; Bi-Variate Weibull; Monte Carlo Simulations; Proportional Hazards Model; Weibull;
Conference_Titel :
Reliability and Maintainability Symposium, 2009. RAMS 2009. Annual
Conference_Location :
Fort Worth, TX
Print_ISBN :
978-1-4244-2508-2
Electronic_ISBN :
0149-144X
DOI :
10.1109/RAMS.2009.4914694