Preventing financial, environmental and safety related business losses caused by critical assets is undoubtedly the central role of reliability engineers, maintenance practitioners and the discipline of physical asset management.
In fact, it is more than a role - it is our duty to our co-workers, the communities we live in, and lastly to our employers. We do have a very unique role in industry as the guardians of physical asset performance. And this unique responsibility increases every year as we become more and more beholden to equipment and machinery to take care of every activity, need and requirements that we or our businesses have.
THat much is pretty straight forward and ever since Moubray wrote his paper on "The Case Against Streamloned RCM" it has been at the forefront of most peoples thinking on this issue.
Where I differ from others is in the way that we define which are our critical assets.
The standard approach either involves some for of FMECA approach, or some form of criticality matrix (reminiscent of the RBI criticality matrix) and they then dutifully apply this to the asset base to determine which is the most critical assets.
The thinking, which I agree with a little, is that they want to try to understand which assets they should be directing their afforts to.
All fair enough so far....
I have a lot of problems with present criticality approaches. From my years bouncing around the world I have found them to be defficient in many areas. They waste resources, make projects larger than they need to be, they distract attantion and often they produce results that are potentially dangerous.
But that isn't the focus of this post...
Which are our most critical assets then?
Understanding which is the most critical assets on almost any plant is not a process but a workshop or even just a question.
I have yet to work with any manager or senior line supervisor, in any of the industries or companies I work with who didn't know which were his most important assets.
It is obvious. The loader, not the haultrucks, the ore concentrator, that single point of failure electrical distribution line, the drainage sump, the rail line. And so on.
Defining the most critical assets is not a real problem.
But then where to? The standard argument is that we need to define our critical assets to sequence our (say) RCM or RCA initiatives.
This is where it all falls apart in my experience.
Understanding critical assets is all about loss prevention. Normally (what's normal anyway) they are working fine, but if they did fall over they would cost millions of dollars, potentially kill people and so on. All dramatically undesireable events.
But if you are looking at sequencing your physical asset improvement program then you need to be able to deliver tangible and cashable value. So just going to the next most important asset may not do that. It works on loss prevention - not so much on loss elimination.
Why? Because you can lose a heck of a lot of money before you start to have problems with critical assets.
So when is a good time?
Determining which are the plants most critical assets is easy and does not need a criticality approach. Period. And if there is some doubt it is a matter of small differences between two or three highly critical assets. For all intents and purposes they are all bearers of highly un-desireable results.
But there are times when it is useful.
Any approach to safety instrumented systems for example can be fraught with problems. There may be 5,000 - 10,000 instruments with safety functions on the plant, often more. Where do you start? There is no doubt, at least in my mind, that all of them need to be done, and there is also no doubt that they cannot all be done at once.
This is a real example of where criticality is useful.
The instrumentation loops have generally one, or no more than three functions. And the primary function is the dramatically most important. So doing a primary function criticality analysis (as most of them are) works here exceptionally well.
Also there are relatively few failure modes of most complex instrumentation systems these days. Design is vastly superior to what it was a few short years ago, and many of the failure modes and common causes of degradation have been managed or designed out entirely.
Another area where it may be wise is where there is a lot of spending to be made, and a limited budget to do so with.
A good example is piping for water companies. An absolutely huge cost area for most infrastructure organizations, and growing every year as the asset base continues to age.
One of the principle drivers of this spending is degradation and anticipated failure in the near future, or within the next pricing round.
But normally they haven't failed yet.
So criticality analysis here helps to balance the consequences of failure with the costs required, and to direct funds to where they are most likely to prevent significant loss to the company, people, or the community.
And again, the pipes have generally one function. And if they happen to havbe more than that then the primary function is dramatically more important than all the others.
Don't fall for the cult of criticality. It prevents more success stories than it creates, and it is by far one of the greatest misunderstandings of the modern physical asset management discipline.