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: Leroy Meyers
Date posted: Mon Mar 5 15:54:19 US/Eastern 2007
Subject: Preventive Maintenance - - Reliability
I have a subcontractor that is providing an MTBF for a system. He is saying that since certain items will undergo preventive maintenance (ie. TWT, Motor), these items are not part of the reliability model (not included in the MTBF calculation). The subcontractor reasons that these items will be changed before they fail.
I say that this is not valid. Preventive maintenance has no affect on a reliability figure such as failure rate or MTBF. A failure rate for these items should be included.
Am I correct or incorrect?
Subject: Maintenance versus Reliability
Reply Posted by: John M. Cloarec
Date Posted: Mon Mar 5 17:37:56 US/Eastern 2007
If certain items are replaced during preventive maintenance, this means that, in theory, they will never experience total and definitive failures. However in the statistic world, it is not because the failure rate is very low that it will never happened. It is likely that item with a very high level of reliability can fails in the first hours of utilization. As you say, preventive maintenance has no effect on the failure rate. The failure rate is an intrinsic feature of each item. If (and only if) we are 100% confident that preventive maintenance will prevent all (and I say ALL) failures, then, yes, itís like these items are 100% reliable and, in this case, there is no need to include them in a reliability model. However, these items can experience partial failures leading to degraded modes, use of redundancy, that will affect the system reliability and in this case you need to consider them in the reliability modeling and prediction. You can find several reliability software packages where you can model the reliability of the system and input not only the failure rate, but also the frequency of preventive maintenance task or the test interval, or others maintainability and testability features (I remember of an old tool, RELTREE, that offers this possibility). Another point to consider: whether your subcontractor replace systematically the items prior to their failure may lead to dramatically increase the LCC (especially the maintenance costs). Maintenance on condition could be a cheaper alternative. Good luck. John