Forum: EXAKT

Forums > EXAKT > The article "Elusive P-F Interval"

The article "Elusive P-F Interval"


United_States
I take issue with your article that appears on Reliabilityweb.com. You leave too many holes for arguments in favor of the P-F Interval to slip through. This (Elusive P-F Interval article) is in the class of a fundamental paper. It needs to hit hard. Yet the story is not clearly stated.

One can say that the P-F interval is a deterministic decision approach, whereas EXAKT is a stochastic approach, that decides on maintenance based on the critical hazard. This dichotomy of approaches (deterministic versus stochastic) needed to be explored deeply in an article. It is somewhat analagous to the argument between classical and relativistic physics, where classical physics is a special case of relativity. In maintenance, the uncertainty of the time of failure is the reality. The increasing temperature of a bearing, the differential pressure increasing linearly across a filter, or the treads of an airplane tire wearing down in proportion to the number of landings, are similarly, special deterministic cases of CBM. The development of cracks in a mechanical component is more stochastic. Maintainers must adopt tools and procedures that deal with the general, probabilistic, and more frequent cases. The more general, the more accurate.

You need to give persuasive examples. In a rewrite you should include some of the arguments by Daming Lin in the plantmaint list. You should rewrite this article to emphasize the weakness of the P-F interval in the stochastic real world.

Multiple variables is another strong argument. Where a single covariate is a linear combination of weighted single covariate values, you need to show that this is a more complex function, which, generally, is not easy to force into the P-F simplification.

Summarizing, your article was not a clear logical story. It was too short. It needed to be more statistical, more fundamental. You must show how the problem can be addressed clearly only when approached in a statistical way. You must talk about managing uncertainty. A rewrite of this article needs to make a strong case for how the simplified or approximate (P-F interval) approach will not lead, generally, to realistic results. A more convincing argument is required if you are to deal realistically with changes in the resistance to failure. Remember that, “For every complex problem there is a simple, intuitive, and wrong solution.”

Naaman Gurvitz, PhD
VP Technology
Clockwork Solutions Inc.


Canada
Thanks, Naaman:

Needless to say, I think that you are absolutely right on. Daming and I will rewrite the article and send the draft to you for more of your wisdom on this important topic.

Murray


United_States
Wait a sec. I'm confused. I have seen a presentation by OMDEC where the presenter used this P-F interval graph to make their case for EXAKT. What gives?

Alan


Canada
Hi Alan:

There are many things wrong with this graph as a description of practical CBM. The most obvious being that it implies that a functional performance drop accompanies a potential failure. But we know from a decade of work in the CBM lab, that, in general, this is not the case. A significant spike in vibration, or the combination of signals found in EXAKT to have attained the optimum hazard for that working age, does not, usually, coincide with any particular drop in the asset's functional performance. Naaman is indeed speaking specifically to this graph, which has become something of a sacred cow, and explains the reactions we received on the plantmaint list.

Murray


United_States
Murray,
Let me take a crack at adding to your re-write as well. I agree with Naaman's assessment, and I have some things written that would support a re-write.

Marty


Canada
Here is the re-write. The Elusive P-F Interval - Re-write. I'm trying to strike a balance between a rigorous treatment of the subject and the need for this article to be readable by us maintenance engineers. Please let me have your comments. Registered users can view / add comments to the article by hitting the orange "Comments" button at the bottom of the article.

Murray

Show posts:
 

Features

Quick Edit a Wiki Page

Menu

Powered by Tikiwiki Powered by PHP Powered by Smarty Powered by ADOdb Made with CSS Powered by RDF
RSS feed Wiki RSS feed Blogs