Recent catastrophic failures of the rocket motor on a US missile caused concern in the US Navy that the missile might not be serviceable past its estimated service life of 20 years. This paper analyzes nearly 2000 firings of the motor under field conditions using classical maximum likelihood estimation (MLE) methods and Bayesian methods. MLE methods indicate that there could be less than a 1% chance that the motor will survive past 20 years of life and are not considered credible. Bayesian methods indicate that there is better than a 99% chance that the motors will survive past 20 years of life. The authors present reasons for preferring the Bayesian analysis and discuss testing schemes for more precise estimates. They comment on the lack of data to perform degradation analysis as a function of temperature cycling for the motor and make recommendations for future missile system data collection.
Warranty calculations for missiles with only current-status data, using Bayesian methods
2001
6 Seiten, 8 Quellen
Conference paper
English
Lifetime estimation using only failure information from warranty database
Tema Archive | 2009
|EDITORIAL - Missiles aren't the only threat
Online Contents | 2003
Reliability modelling using warranty data
Automotive engineering | 2002
|Practical guidance for homing missiles with bearings-only measurements
Tema Archive | 1996
|