Category: Data and samples
-
RA requires LRCM
The idea behind EXAKT is an appealing one. What can be more desirable than a systematic way to determine the “exact” relationship between condition monitoring data and impending failure? Although I wish that I could say that EXAKT auto-magically generates predictive models, I cannot. EXAKT translates the occurrences of PFs, FFs, and Ss of failure…
-
Interpreting failure data
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…
-
Sample selection
A good sample should include both good and bad actors. For example, pumps that fail frequently as well as those that do not fail or fail infrequently should be included. This ensures an unbiased sample. Normally the analyst takes two points in calendar time a few to several years apart. Then within that time window…
-
The CMMS barrier to RCM
One company tried to connect its CMMS with its RCM program. Not just the usual upload of the RCM strategy from the RCM application to the CMMS but a more organic day-to-day integration of the two processes. However, they ran into difficulties. The CMMS being based in a relational database tries to codify as much…
-
Free text on the work order
Information technology has changed the way we work, but not the internal workings of our minds. Another 10 or 100 generations will pass before the human brain evolves the capacity to describe complex situations by selecting items from drop down lists. What exactly is the purpose of free text on the work order? A work…