The P-F Curve, in Sickness and in Health.

My wife and I have been married for 25 years - a significant achievement we take pride in. Yes, there have been good and not-so-good times, but we made an oath to look after each other, in sickness and in health, and we’ve kept it. But we’re not the only ones.

Taking an oath has been institutionalised across many fields; my local doctor, for one, may have taken the Hippocratic Oath in the spirit of ‘first do no harm’. But what about maintenance and reliability staff, engineers, and asset managers? Do they swear by some secret oath? Perhaps there should be ‘zero harm’, an oath to put the health of their equipment first.

Admittedly, when it comes to maintaining the lifecycle of a machine, ‘zero harm’ would be (and is) challenging to implement. How would you know if you’re ‘doing no harm’? The approach certainly isn’t the same as managing infant mortality, but I believe the philosophy is. We should consider the possible harm that any intervention may cause.

When it comes to equipment monitoring, this comes down to the maintenance strategy you employ.

Luckily, the ‘harm’ that each approach causes has already been mapped out for us on the P-F Curve. For those unfamiliar with the theory,

"The P-F curve is a way of representing an asset's behaviour or condition before it has reached a failed state. It illustrates an asset's progression toward failure. On the chart (see below), the x-axis represents the time to failure, starting with an asset's installation, and the y-axis represents an asset or component's resistance to failure." ~ International Society of Automation, 2019

 

MA_2019_special-section-fig1
Source: https://www.isa.org/intech-home/2019/march-april/features/improving-maintenance-by-adopting-a-p-f-curve-meth

So how do we know that we have moved out of ‘doing no harm’ to damaging the machine?

Along this curve, assets shift from health to sickness, and the maintenance strategy does too. Suppose you let your machines run until they break (corrective/reactive maintenance) or wait until they’re noticeably very sick (preventative maintenance), coughing and spluttering like me with the ‘man-flu’. Not pretty. In that case, it’s not hard to fathom that these ‘sick’ maintenance strategies are damaging the machine.

In my mind, the solution lies in condition-based and precision maintenance, ‘healthy’ maintenance strategies that get ahead of any advanced failure. I do this as part of my health management, and I’m sure you do too. When we get ahead of situations that may become stressful later – scheduling that massage when your back starts to ache, rather than waiting until you’re unable to walk– we’re always better off. Sometimes only a little realignment is needed for peak performance, and implementing strategies at the start of the curve allows you to do so.

Importantly though, not all condition monitoring is the same. For example, you may already monitor the condition of your machines using a preventative strategy, and it may seem to have worked. But of the many tools and techniques that help you track the ‘health’ of a machine, which you can see in the graphic below, some are more harmful than others.

P-F IntervalAsset Failure Curve by Dustin M. Etchison 

Here are four questions that will enable you to qualify if your condition monitoring is appropriate:
  1. Do you have the human resources available to capture condition monitoring data on a regular basis?
  2. How fast will a failure occur? Do failures happen between your inspection points? For example, say you run monthly assessments. If your assets can fail within four weeks, are monthly assessments enough?
  3. Do your machines operate in a consistent and predictable manner? Perhaps you have process changes or fluctuations in machine operation. In this case, setting alarms at ‘x’ or ‘y’ vibration/temperature won’t do. These alarms get ignored, as your team will soon realise they’re time wasters, like the boy who cried wolf.
  4. Do you have the skills and the resources available to do your own analysis? (the graphic below is a good guide)

Division of condition monitoring.
Source: https://www.researchgate.net/figure/Division-of-condition-monitoring_fig4_245079171 

To summarise, the strategy you use to maintain your machines will directly impact how healthy or sick your machines are. Choose the most appropriate to your needs.

We believe adopting a condition monitoring strategy is ideal for helping you and your team to understand the progression of a healthy asset. However, before jumping into a particular tool or technique, understand your needs. For example, do you have the skills to diagnose manually, do you have machines that vary, and can you afford to have gaps or delays in your understanding of machine health? The significant variable on the P-F Curve is TIME.

Hopefully, this article has shed light on the motivation to move from sickness to health. We’d like you and your assets to live a long and healthy life.

What oath are you going to make to your hard-working machines?

 

---

References

Cover image sourced from: https://smartloving.org/making-the-wedding-vows-stick/

https://www.isa.org/intech-home/2019/march-april/features/improving-maintenance-by-adopting-a-p-f-curve-meth

https://www.researchgate.net/figure/Division-of-condition-monitoring_fig4_245079171

 

Further Reading

https://reliabilityweb.com/articles/entry/why_people_do_not_understand_the_p-f_curve

 

Leave a Comment