The Curse of Criticality

Criticality is one of the most misapplied and misunderstood methodologies in a reliability practitioners arsenal today. Personally I am absolutely astounded by the continual use of criticality approaches to select assets for an improvement process. Don't misunderstand me, I like the idea, and I believe it has a lot of very powerful uses in reliability. Just not as an asset selection process - and particularly not for RCM or any other form of maintenance strategy approach. It goes like this... The newly trained RCM faciltiator takes the initiative to push forward the program by spending time developing a criticality ranking of his asset systems and then selecting the most critical (hence the most important) to begin the work of RCM. He works diligently, with some of his colleagues, and maybe he takes two weeks to get it done. (Total of 10 4 hour meetings)  Then what...? Whats wrong with this picture? In my experience critical assets (or systems) are generally known in the first place. Secondly, they are generally the first to receive resources, cash, time and whatever else they need to make sure they are running okay. Last, they are often pretty reliable anyway!!! So, the hard working facilitator goes to his manager and says "Here boss, you will be happy to know that this critical asset, which was running reliably, is going to keep on running reliably!" And his boss says what? "Thanks, now get out" I have other concerns about criticality at a system level, most of which have convinced me that it is not even possible let alone practical, but the whole issue of momentum, management support and getting projects started really tops them all. The Political Element Like everything else at work there is a political element to this. We (RCM Practitioners) realize that a post RCM analysed asset will bemore reliable than a pre-analysis asset. Period.But our bosses often do not... So what are they interested in? 1. revenue increases, 2. Cost reductions, 3. Risk mitigation and 4. Knowledge increases; generally in that order. The moral of the story? Don't waste time trying to build support by choosing the reliable important assets, start with the unreliable poor performing assets. Go for the ones that have benefits already built in to their current levels of performance. It is far easier (from my experience) to build support for something when you walk in the door with $1 million of forecast revenue increases than with a "more reliable important asset". Wait.. I hear you cry "But thats reactive, we're supposed to be proactive!" Yup, I agree. But we are where we are, and before we can start working on tomorrows problems (proactivity) we need to fix the headaches of today.