Ahead of the game

Dirk DeNutte explains to NGF why Predictive Maintenance is the most effective way of detecting equipment problems earlier and keeping costs down.

“Most failures are developed over a period of weeks, months or years”
-Dirk DeNutte, Managing Director of GPAllied, LLC

Unless you've been living on another planet for the last fifty years, you already know that the case for doing preventive maintenance is watertight. Done right, preventive maintenance will preserve, protect and extend the life of your equipment - and boost overall return on assets.

However, most maintenance and reliability professionals are unhappy with their PM program, as they believe it consumes too many resources and is simply bigger than it should be. They feel like they don't have enough manpower to manage all of their PM's along with the other important maintenance work.

Despite all of the time and money being spent on preventive maintenance, there are still far too many unexpected equipment failures.

For example during a recent chemical plant tour, the frustrated maintenance manager said, "We just PM'd that machine, and it failed a short time later anyway. So why didn't we catch the problem with the PM?" Why indeed?

In a nutshell, the problem with preventive maintenance is that it takes too much time and produces too few results.

By definition, all PM's are time-based. That means that either calendar time or operating time dictates when an asset should be inspected, cleaned, adjusted, replaced or reconditioned. But is there really a direct relationship between the time equipment spends in service and the likelihood it will fail? In short, the answer is no.

In fact, for complex systems, the majority of failures will occur at random. The reality is that 89 percent of equipment failures are not age-related. Therefore, there's no amount of time-based preventive maintenance that can manage these failures effectively.

That's why using time as the primary basis for your maintenance strategy is inherently flawed. It will have very little impact on overall reliability. From a risk standpoint, it's much safer to assume that equipment failures can happen at any time.

Despite what you may have heard, the basic maintenance process is really simple: detect problems; plan and schedule repairs; make repairs.

The secret to good maintenance is in the first step. There is a direct correlation between early detection and maintenance costs. Simply put, the earlier you can detect problems, the faster, cheaper and easier it is to make repairs. Emergency repairs can be astronomic. There is a huge difference between emergency maintenance and planned, proactive maintenance.

The fact is that problems start small and get worse with time. It's important to realise that the equipment will send off early warning signals along the way. These early warning signs might be slight changes in temperature, vibration or sound. Not all of these changes can be detected by human senses, but they can be picked up with special equipment designed for that purpose. And that's what the whole field of condition monitoring and predictive maintenance is about.

With advanced technologies in vibration analysis, infrared, ultrasound, oil analysis, motor current analysis and non-destructive testing, trained technicians can routinely monitor and inspect equipment and detect these early warning signals. The difference between the time taken for a Predictive or Condition Monitoring specialist to detect a problem and for an operator to detect one is huge. Remember, detection time equals money. Big money!

In spite of the evidence, the reality is that most asset-dependent companies are not doing nearly enough Predictive Maintenance. Even though Predictive Maintenance has been around for over 40 years, it is still "new" to some organisations. That leads to late detections, emergency maintenance, and all the painful costs that come with it.

In summary Predictive Maintenance should be an integral part of your reliability strategy - and account for at least 50 percent of your maintenance work.

For more information on Preventive Maintenance contact GPAllied on info@gpallied.com or +32.9.210.17.20.

Dirk DeNutte became the Managing Director of GPAllied, LLC in March 2009 after the Joint merger of General Physics and Allied Reliability. Previously, he was in charge of Allied Reliability's European consulting business. Dirk has over 24 years of experience in the marine and industry maintenance sector.