SRC Forum - Message Replies


Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum (src_forum@alionscience.com )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Back to message list Show all replies Topics List About this forum
Original Message:

Posted by: Ty Pollak (pollakr@afotec.af.mil )
Date posted: Thu Aug 5 15:10:19 US/Eastern 1999
Subject: Use of Reliability predictions to supplement flight test data
Message:
I am analyzing reliability for a missile weapon system and basing my point estimate on flight test data. Due to the limited data, however, I'd like to incorporate contractor-provided reliability estimates to decrease my confidence interval about the pt estimate. The contractor solution is based on HALT and other developmental tests. My solution is to develop a distribution for Pr (system reliability) based on flight tests only, and another for the contractor's Pr. Then sample from these distributions according to a "scale factor" determined by my confidence in the contractor estimates, methodology, and correlation w/ flt test data. This scale factor is subjective. By sampling from the 2 distributions, I will generate a "total" Pr distribution, which I can pick off confidence bounds from. Any comments, suggestions, insights?


Reply:

Subject: Flight and Develoment Test Data
Reply Posted by: Bruce Dudley (bdudley@alionscience.com )
Organization: Reliability Analysis Center
Date Posted: Thu Aug 19 9:16:44 US/Eastern 1999
Message:
To maximize the test information, I would combine the development testing and the flight test data on an hour to hour basis, but not use the overstressed HALT test data. Halt testing is an accelerated hi-stressed test scheme and the failures and test hours may not represent the operating world. Scaling this halt data is very subjective based on the acceleration factors that one estimates. This information may be based on previous testing experience and not represenative of the existing equipment or environments.


Reply to this message