Effective Preventative Maintenance - questions to ask
The purpose of any preventative maintenance routine is to reduce the chance of failure and, in doing so, increase reliability. Much of our time is consumed carrying out Preventative Maintenance (PM) routines but we often don’t take time out to question how effective the routines actually are.
Preventative maintenance is carried out to preserve the inherent reliability of equipment or to detect failure in sufficient time to plan, schedule and conduct a corrective action. If the PM’s do not achieve either of these objectives in a cost effective way then we have to examine their validity.
Firstly, let’s look at using PM’s to preserve the inherent reliability of our equipment. When commissioned, all plant and equipment has a service life and level of reliability that is determined mainly by the quality of the original design, manufacture and installation. If correctly maintained and operated our equipment will reach that level of inherent reliability. Such a result is achieved through scheduled replacement or restoration of components or through tasks such as cleaning, lubrication and making adjustments.
The second purpose of carrying out PM routines is to detect failure in sufficient time in order to plan, schedule and carry out a corrective action. This usually takes the form of condition monitoring checks such as vibration, temperature or oil analysis. The key to these checks is in understanding the data. For every condition monitoring check, data must be recorded and analysed, as simply writing data on a work order is not sufficient. If it is not recorded, plotted, analysed and used to predict failure to plan corrective action then it may prove a useless exercise.
To assess whether our PM routines are as effective as we want them to be we need to ask ourselves some questions.
1. Does the PM routine preserve the inherent reliability or predict failure in time to plan, schedule and carry out corrective action?
We can assess whether a condition monitoring routine identifies failure by simply analysing the data we have available. For example, if we are monitoring vibration on a pump set but are still getting unplanned bearing failures then we need to reassess the exercise. This could be because of the way the reading is taken or the time interval but more often is down to us not recording, plotting and analysing the data well enough.
It isn’t always a simple task to assess whether or not a cleaning or tightening task for instance is preserving the inherent reliability, as often the only method to do this is to stop the task and monitor reliability, which is often not an option. There are still other questions we can ask though to help monitor the effectiveness of these PM routines.
2. Is the time and effort that goes into the PM routine relative to the consequence of the failure of the equipment?
It is important to always remember that it isn’t the failure itself that we are trying to avoid but the consequence of that failure. Continuing with the pump set example, we may have the same pump in several locations throughout the plant. So a PM routine on one that is crucial to safety or production justifies a higher level of preventative maintenance than one that has little or no consequence if it were to fail. We must ask ourselves whether we are under- or over-maintaining each piece of equipment to ensure we are maximising the limited resources we have at our disposal.
3. Are we spending more than 20% of our time on scheduled restoration or replacement of components based on hours or cycles run?
If so we are undoubtedly carrying out too many of these tasks. Research has shown that only about 20% of failures can be predicted based on time run. Failure of these items can often be predicted, but not based on time – we must look for other indications such as vibration or temperature change.
4. Is the cost of carrying out that PM routine more than the cost of the failure of the equipment?
We sometimes continue to carry out these checks simply because we have always done them; we never take time to carry out a cost/benefit analysis. If the PM costs more than the full cost of repair, taking into account all the consequences such as production or quality loss then it is simply not worth following.
Preventative maintenance routines are vital for preserving equipment reliability but they take a large amount of our resources. We need to be sure that our maintenance routine is having a positive impact on plant reliability and we aren’t simply following such steps because that’s a traditional approach.
Perhaps it’s time that we took a step back and started analysing the effectiveness of our PM routines. After all, they say one sign of madness is doing the same thing over and over while expecting a different result.