If You’re Not Running This Calculation, You’re Losing Money.
Design is a funny word. Some people think design means how it looks. But of course, if you dig deeper– it’s really how it works.
Math doesn’t lie. People can interpret equations incorrectly (or even falsify the inputs)…but math is consistent. Unless we are dealing with crazy quantum mechanics or some other science oddity, “2 + 2” will never equal anything but 4.
Application in real life, however, is anything but straightforward. That’s why there is a course in torture they inflict upon engineers called physics. When trying to apply mathematics to real life, nothing seems to work right. The same happens when attempting to predict patterns in fixed asset operation. So how do we calculate asset failure probabilities and revenue risk?
Heuristics
Big absurd word, but a simple meaning — find or discover. Heuristic algorithms are not intended to be exact answers; they are the best approximation at that moment. Another way to put it: heuristic calculations are practical answers for immediate goals. There is no sense in spending countless dollars trying to perfectly model your fixed assets as they operate today. Tomorrow, who knows how this will change. Heuristics is a pragmatic approach to a very complex problem.
Calculating Probability of Failure
Let’s look at a heuristic method for the failure probability of an asset. First and foremost, we need the age of the asset. Then, the life expectancy, taking into account the actual operation of the asset, must be estimated. For instance, if a standard packaged rooftop unit (also called unitary HVAC unit) is serving a normal office operation and is sized correctly, the economical life expectancy would be 15 to 20 years (by contrast if this was involved in an industrial process, the life expectancy could be as little as 5 years). The actual age divided by the expected age will give a starting percentage. To clarify, 0% would mean it should have no probability of failure; 100% would mean it has failed or is at risk of imminent failure.
Again, real-life application is not that simple. Assets are not like a tube of toothpaste. You can’t measure exactly how much was there to begin with, or how much is remaining. Look at it this way: just because your odds of winning the Powerball are 1 in 292,201,338 doesn’t mean you won’t play and win. Enter the maintenance big-data piece.
We never think of maintenance and repair technicians as creators of big data, but we are wrong. Every single, seemingly insignificant, interaction with an asset is a chance to capture “loads” of data — a technical term, I assure you. Even a picture from a smartphone documents much about an asset’s present state. We humans, typically, cannot process the data efficiently, so we do not bother with it. Image recognition and machine learning allow a simple visual condition cue — is the condition excellent, satisfactory, poor, or critical? Four choices that we can increase the life expectancy ratio with (or in some cases, decrement it). Also, a well-rounded maintenance technician gives feedback as to their opinion of the life remaining. Obviously, if we weight their opinion too highly, the calculation could skew out of control…but we should never discount the value of sound human feedback when it is available.
Revenue Risk
Financial Directors never like anyone other than their immediate staff seeing finance data, which is understandable. However, in order to fully manage assets correctly, the revenue must be tied to the exact generating (or supporting) asset in the magnitude that particular asset serves. This “revenue footprint” is not easy to develop. It will take time and will be updated as the business progresses. Though the exact revenue amounts will be encrypted, an index from 1 to 100 can provide enough information to calculate the impact of each asset’s condition to the revenue. If the impact is 1, then it should be prioritized lower than an impact of 80. The total revenue calculation: probability of failure x revenue index value = risk magnitude. This may be relatively simple, but simple is good.
Bottom-line: the calculation design is simple and easy to understand. As Einstein’s famously attributed quote says: “Make things as simple as possible.” The execution, however, is where the rubber meets the road.
I have been working on this execution throughout my entire career. Lately, my main focus has been on building this cloud platform — thus the reason I have written these articles: to document the theory behind the platform in order to further advance in the business realm. From here on out, the articles will focus on issues and updates with The ROI Reports’ take on solutions.
Follow us for next week’s update in the world of business, finance, and operations. We welcome your feedback and suggestions. Remember: “Lack of money is the root of all evil.”
Originally published at roireports.com.
