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: Ken
Date posted: Sat Aug 31 0:59:37 US/Eastern 2002
Subject: Reliability Growth Testing Mil-STD-189
Message:
I would like to ask if there is Reclassification of failures in RDGT. In RD, we allow to reclassify the failures if proven that the corrective actions are effective through further testing. Is it stated anywhere in MIL-189 that we can or cannot reclassify failures in RDGT. Secondly, if my RDGT target is to meet MTBF of 700 hrs, what is the recommended test duration required for RDGT. Thirdly, what distribution does AMSAA model follows? (e.g exponential or weibull?) is it applicable for complex mechanical and electronic system?


Reply:

Subject: Growth Testing
Reply Posted by: B.W.Dudley (bdudley@alionscience.com )
Organization: RAC
Date Posted: Tue Sep 3 15:11:33 US/Eastern 2002
Message:
Reclassification of failures is usually done for reliability demonstration tests but not for growth testing. Paragraph 5.4.1.3.5 indicates that “purging of failures” is not recommended as it implies that design fixes reduce the probability of failure to zero. This case is very seldom, if ever, true. In fact the handbook tells the procuring agencies to specifically eliminate this action by specifying that this practice be removed from the growth test process. The length of any growth test is dependent on the start and end points that you need to navigate. The start point always depends on the maturity of the equipment and the stresses being applied. If the start value is one tenth the need, then a long test will be required. Also, the growth rate is important in determining the length of the test. Low growth rates mean much longer tests. The time factor can be computed using the following equation. MTBF(need) = MTBF(initial) [T(end time) / T(initial)]^a(growth rate) x (1-a)^-1 The AMSAA model follows a non-homogeneous Poisson process. This process is applicable to electronic, electromechanical and mechanical equipment.


Reply to this message