Do you plan to fail a core component prematurely this year? Surely not!
Australian cricketing great, Glenn McGrath was always asked to predict the outcome of an upcoming Ashes series against England. Without hesitation, he predicted a 5-0 win to Australia. When quizzed why he always predicted that outcome despite the team’s variable form, he would respond “You’re asking me to predict we’re going to lose a test match?”
His mindset would not allow him to contemplate failure before a test match or series. He always knew the possibility of failure was there but adjusted his mindset to plan for it not to occur. (He was only right on one occasion in 2007)
It’s a similar question a maintenance team might ask themselves. “How many engines are we going to blow before the end of their useful life this year?” Of course, no one will predict how many engines will fail prematurely in a year because successful maintenance teams plan for success, (just as Glenn McGrath would never plan to lose a test match).
Predictive maintenance isn’t predicting component failures, it’s predicting when to take action to prevent a failure. But what if things don’t always go to plan and the unexpected happens? Is this something that you are happy to live with? Is there a certain amount of failure is acceptable with still achieving your reliability goals?
“How many engines are we going to blow before the end of their useful life this year?”
The pursuit of a failure free equipment is not only fanciful, but it would also be outrageously cost prohibitive. Accepting that a hose or air horn is not worth spending money on to maintain or repair is fine, as you’d just replace it at the next scheduled service, however, accepting a core component ‘might’ fail is not. Choosing what is an acceptable level of failure is an important one not only from a planning perspective, but also cost management. Most CFOs wouldn’t accept an excuse that “these things happen”.
The important question then becomes, if there is a budgeted maintenance plan with an acknowledgement of an acceptable level of failure, then why do sites still lose costly core components prematurely? (Even if they have a run to failure approach)
At Relialytics we simply don’t accept that “these things happen”. Unless there is a significant manufacturing fault, all premature failures can be prevented. There is enough data extracted from components through telemetry and oil samples to give an early indication of something that’s not right which will allow for early investigation and rectification. It takes discipline to acknowledge and accept the signals that there are abnormal conditions and then act as soon as practicable.
Effective condition monitoring shouldn’t be monitoring for advanced signs of failure and then reacting, it’s about acting on changed conditions very early before any damage is done. A simple yet effective approach to core component failure prevention. Its effectiveness relies on action being taken as early as possible and not waiting to see if the issue will progress any further.
Glenn McGrath may not have had the score line he always predicted, but there is no reason at all why any mine site should lose a core component prematurely. When the early signs of failure are not identified, or worse still ignored, it creates opportunity for a component to fail. That’s predictable.
Effective condition monitoring shouldn’t be monitoring for advanced signs of failure and then reacting, it’s about acting on changed conditions very early before any damage is done.
Do you have any thoughts on this blog? We'd love to hear from you.