I am often asked by a prospective customer this simple question – do you do Prognosis?
If the customer is in DoD/Aerospace world, or has an established R&D and Health Management program, the short answer is yes!
However, customers asking this question often are field service organizations that maintain expensive assets and are looking for an alternative to the traditional break-fix service model. The term Prognosis has become quite popular over the years, thanks to millions of dollars invested by the US Department of Defense in programs such as the Joint Strike Fighter. The promise of Prognostics is simple — wouldn’t it be nice if you could predict how much life each component has left, so that you could replace them just before they failed? But is Prognosis the right tool for you?
First, how good is your Diagnosis? If you are struggling with faults that have already happened, chances are you won’t do any better with faults that have not happened yet! Think of Prognosis as something that involves Diagnosing impending failures and predicting when they will develop into full blown faults. So, Diagnosis is the foundation for Prognosis, and the need for predictive capability makes Prognosis a powerful but expensive technique that should be used wisely only where it is necessary.
This brings us the second question: do you really need Prognosis? For example, your car has two headlights. While it is no fun driving in the rain with one headlight (have you noticed how headlights always seem to fail on rainy days?), having two headlights means that you can still get back home when one has failed. So, redundancy and fault-management are effective ways of reducing unscheduled downtime. For your critical components, evaluate what is the most cost-effective method to avoid disruption in service, and choose wisely.
And now, the all important third question: why do you want Prognosis?
Some people will answer I want to use less maintenance. For example, you may be used to changing the oil in your car every 3 months (schedule-based maintenance) or 3000 miles (usage-based maintenance), but by monitoring the condition of the oil and the filter, you could change the oil only when needed (condition-based maintenance or CBM). This is a valid application of condition monitoring, although strictly speaking, this is not Prognosis. Also, keep in mind that you may not be able to extend maintenance interval for safety critical components without exposing yourself to more liability.
However, most of our customers answer they need prognosis to reduce unscheduled downtime by doing preemptive repairs. Here too, Prognosis is not the only answer.
Let’s take an example — supposing you want to avoid being stranded on the highway due to tire failures. You could add sophisticated techniques that monitor the tread of the tires and how it is wearing out, how the underlying structure of the tire is holding up, the stress on the tire, etc, and predict when failure is imminent. You could develop such Prognosis at significant R&D expense, or, you could simply replace the tires when they look worn (CBM) (e.g., cracks on sidewall and/or tread-depth of 1/32nd inch or less). The second method may cause you to use at most one extra set of tires over the life of the car since you will throwaway tires with still some useful life left on it, but newer tires also improves your safety and performances, which has its own reward. Best of all, you can use the second technique on your current installed base without having to develop new technology.
Let’s also not forget, Prognosis or CBM does not completely prevent unscheduled downtime. You could still hit a pothole and get a flat tire – no matter how new your tire is!
To sum up, there is more than one way to reduce unscheduled downtime: Prognosis, Condition Based Maintenance, redundancy and fault tolerance and fault management. QSI can help you implement all of the techniques discussed here in a balanced health management solution. Health Management is not just Diagnosis or Prognosis, but an effectively engineered delivery of uptime at a reasonable cost.
Let us help you find the right balance of techniques to achieve your objectives.
