Do Preventative Maintenance Routines Increase Reliability?
Reliability is defined in the broad sense as the science aimed at predicting, analysing, preventing and mitigating failures over time. In the narrow sense it’s defined as the probability that a device will operate successfully for a known period of time and under specified conditions when used in the manner and for the purpose intended.
Getting high equipment reliability is mostly within the power of every organisation. You improve reliability by setting policies, using appropriate methods and adopting the standards that reduce the chance of equipment failure. Preventative maintenance (PM) routines are the backbone of maintenance policies and methods in most organisations. So how much do they add to equipment reliability?
We’ve probably all seen the classic bath tub curve showing a relatively high rate of failure at the beginning and end of equipment life. Research carried out in the USA by the airlines and Navy show that the classic bathtub profile only applied to only 3 or 4% of equipment failures. The research clearly showed that failure can only be predicted based on age or cycles around 15% of the time. The remaining 85% of failures occurred randomly when measured against time or cycles.
The purpose of any maintenance routine is to reduce the chance of failure and in so doing increase reliability. In some cases our PM’s are undoubtedly doing this but in many cases they achieve little and can actually contribute to equipment failure. Take the routine of partially dismantling a piece of equipment to periodically inspect a bearing. Presumably, we are checking the bearing because we know the mean time to failure and are inspecting at more frequent intervals than this to pick up signs of wear. For this to be worthwhile the bearing would have to follow the expected pattern of wear and failure. Knowing the figures from the US research and considering that the same bearing is probably operating in many different applications & environments, it’s unlikely that it will follow a known pattern. By dismantling a serviceable piece of equipment to inspect a bearing that may or not fail according to a pre-determined pattern cannot be the most effective or efficient way to operate.
The medical profession learnt many years ago that the less invasive a procedure is the less likely it is to cause complications. Maintenance could learn a lesson there. Surely it’s far more effective in terms of time, effort and expense to move our maintenance away from preventative routines and use predictive techniques wherever possible. In the case of the bearing this could take the form of a vibration or thermography check rather than a visual inspection of the bearing.
Many maintenance routines we carry out were originally laid down by the manufacturer. In some instances we slavishly obey their maintenance instructions way past the warranty period, knowing that the routine has not prevented failure. Usually we continue with the routine because nobody is prepared to take the risk of removing it from the schedule, not because we feel it worthwhile.
Preventative Maintenance routines attempt to avoid the penalty of failure by using time or cycle based maintenance. If we know we can only predict failure in this way 15% of the time then we have to accept that there is considerable waste inherent in the preventative maintenance strategy.
While I’m not advocating that we should immediately stop all PM’s I am suggesting that we should question them. Any maintenance we carry out must be done because it is effective in increasing reliability, not because we’ve always done it that way. Good maintenance evolves though the strategies of Reactive, Preventative, Predictive and finally to Pro-active maintenance. Employing solely preventative routines means that we’re not even half way there!