predictive maintenance

The case for moving to predictive maintenance

Preventive programs have become standard, but may miss variations in real-time conditions
Thursday, August 25, 2016
By Saar Yoskovitz

Chances are, many facility managers have been running on a preventive maintenance (PM) program for years now and they feel generally comfortable with the results. Routine inspections have caught a decent amount of problems before they were critical, such as cracked belts or worn gears, but operations still suffer from downtime due to other, harder-to-see issues.

Facility managers have also likely heard of predictive maintenance (PdM), but the associated costs may have kept them from digging too deep. With recent developments in PdM technology, however, it may be time to reconsider.

What preventive programs miss

A simple analogy for preventive maintenance would be how one might keep a car running — follow a maintenance schedule, take it in for an oil change, replace the brake pads and the belts, all according to mileage and not necessarily the real-time conditions. While the schedule has been created based on normal wear and tear, if conditions vary, the schedule may be off — sometimes drastically. Preventive maintenance may also involve a visual inspection of systems, which can miss some problems.

Adopting a PdM program can fill in the blind spots where PM routines fail. By default, much of a PM program involves a run-to-fail approach, where unplanned downtime is a norm of operations. PdM, however, helps to predict issues and allows downtime for repairs to be scheduled at a time most convenient to operations. It relies less on guesswork and more on measured data. PdM uses sensors and other methods to track real-time machine conditions to alert maintenance crews to the first signs of potential breakdowns.

When considering adopting a PdM program, ask some of the following questions:

  • Could PdM methods, such as oil analysis, thermal imaging, or vibration analysis, have helped to spot and predict breakdowns where PM would not
  • Would the cost of implementing PdM be less than the cost of those unexpected repairs and downtime?
  • Did that downtime last longer than expected because purchase parts could not be purchased in advance?

If the answers to these questions are yes, then PdM could be a cost-saving step to consider taking for an operation. But how?

Predictive technology becomes accessible

Once upon a time, PdM was a technology reserved for only those with the deepest of pockets — to install sensors and monitoring equipment was a timely and expensive endeavor that involved a deep level of commitment. Today, however, PdM can be performed much more easily. It is possible to use mobile equipment to predict problems, and the equipment itself runs on cloud technology.

Predictive maintenance technologies are becoming more mobile and affordable. Advanced techniques that used to be out of reach for most, such as vibration analysis, are becoming more accessible. Some new technologies help technicians diagnose problems in just minutes. Cost-effective equipment and real-time data analysis make these viable options. The data analysis software learns as it goes, too, so patterns that showed up in past problems help technicians prevent the next problem.

Vibration and acoustic analysis once involved startup costs beginning at USD$40,000 and went up from there. Using consumer-grade mobile technology and some additional hardware, the same results can be achieved at a fraction of the cost. Additionally, these new methods offer a flexibility not found with legacy systems — beyond equipment costs, older methods often required highly trained operators, whereas new methods use cloud-based computing to analyze information and deliver accessible and actionable results. These kinds of increases in efficiency help both a company’s margins and its reputation for quality and speed.

Saar Yoskovitz is the co-founder and CEO of Augury.

Leave a Reply

Your email address will not be published. Required fields are marked *