Here is an excerpt from the famous Nowlen & Heap RCM report of 1978. It describes some basic difficulties in the analysis of reliability related data.
The problem of interpreting failure data is further complicated by differences in reporting policy from one organization to another. For example, one airline might classify an engine removed because of loose turbine blades as a failure (this classification would be consistent with our definition of a potential failure). This removal and all others like it would then be counted as failures in all failure data. Another airline might classify such removals as ‘precautionary,’ or even as ‘scheduled’ (having discovered a potential failure, they would then schedule the unit for removal at the earliest opportunity). In both these cases the removals would not be reported as failures.
Similar differences arise as a result of varying performance requirements. The inability of an item to meet some specified performance requirement is considered a functional failure. Thus functional failures (and also potential failures) are created or eliminated by differences in the specified limits; even in the same piece of equipment, what is a failure to one organization will not necessarily be a failure to another. These differences exist not only from one organization to another, but within a single organization over a long calendar period. Procedures change, or failure definitions are revised, and any of these changes will result in a change in the reported failure rate.
Another factor that must be taken into account is the difference in orientation between manufacturers and users. On one hand, the operating organization tends to view a failure for any reason as undesirable and expects the manufacturer to improve the product to eliminate all such occurrences. On the other hand, the manufacturer considers it his responsibility to deliver a product capable of performing at the warranted reliability level (if there is one) under the specific stress conditions for which it was designed. If it later develops that the equipment must be frequently operated beyond these conditions, he will not want to assume responsibility for any failures that may have been caused or accelerated by such operation. Thus manufacturers tend to “censor’ the failure histories of operating organizations in light of their individual operating practices. The result is that equipment users, with some confusion among them, talk about what they actually saw, while the manufacturer talks about what they should have seen.
The above impresses upon us the importance of a clear reliability information strategy. Such a strategy goes beyond the development and application of failure codes. Not only must our strategy link significant work orders to the the appropriate RCM records, but it must also include an understanding, related to current context, of what constitutes a potential failure, a functional failure, and a suspension. No software system, on its own, regardless of how sophisticated, will obviate the requirement by your maintenance organization to develop and promulgate standards that encompass a (LRCM) reliability information strategy.
© 2011, Murray Wiseman. All rights reserved.