SRC Forum - Message Replies
Forum: Reliability & Maintainability Questions and Answers
Topic: Reliability & Maintainability Questions and Answers
Topic Posted by: Reliability & Maintainability Forum
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998
Posted by: Jeff Jackson
Date posted: Thu Aug 30 0:19:13 US/Eastern 2001
Subject: Accuracy of reliability prediction methods
Dooes anyone have any data to bound the accuracy of reliability predictions performed to various methods such as MIL-HDBK-217, Telcordia, etc?
Subject: Reliability predictions and observations
Reply Posted by: Larry George
Organization: Problem Solving Tools
Date Posted: Wed Sep 5 18:43:30 US/Eastern 2001
"MTBF vs. Age-Specific Reliability Prediction," Reliability Review, Reliability Division of ASQ, Vol. 21, No. 2, June 2001, pp13-15. This article compared predicted and observed age-specific reliability of a communications switch. The prediction underestimated infant mortality.
A sales representative for http://www.itemsoft.com said they'd post the article. Meanwhile, contact firstname.lastname@example.org for an electronic copy.
Subject: Reliability Prediction Bounds
Reply Posted by: B.W. Dudley
Organization: Reliability Analysis Center
Date Posted: Thu Sep 6 15:43:36 US/Eastern 2001
Comparing actual test or field performance data to predicted information is a difficult operation as environmental characteristics have to be determine along with electrical stress parameters for each component. Many predictions have unrealistic assumptions applied to the components resulting in either optimistic estimations or incorrect analysis.
The last real effort to analyze this situation was performed in 1989 and was documented in the technical report “RADC-TR-89-299” Reliability and Maintainability Operational Parameter Translation. This report collected equipment reliability and maintainability data from a number of military contracts and compared the predicted results to the demonstrated ones using test and operational field results. Translation equations were developed based on a regression of the data for each equipment set and environment. Overall, the results showed good correlation for predicted to field for controlled environmental conditions, but optimistic predictions compared to field results for harsh environmental conditions. The complete report is available from the Defense Technical Information Center at Ft. Belvoir VA for qualified defense contractors. The basic equations are listed in the “Reliability Toolkit; Commercial Practices Edition” which is available from the RAC product catalog.
The RAC has a new prediction product called “PRISM”. In one test case dictated by a customer, we compared the PRISM prediction to real operating data for a specific product. The results were very encouraging as we achieved a near one to one relationship for an adverse environment.