Industrial Game Changer: How The IoT Is Rewriting The Rules Of Maintenance

Manufacturing, the industry which makes everything, is itself being remade. As industrial applications of the Internet of Things (IoT) add both continuous connectivity and precise monitoring to more and more machines, data is transforming the way we operate and maintain them. Maintenance is an unsung hero. From printers to power plants, all the amenities of our modern life require machines, and those machines in turn require some TLC.

From the Industrial Revolution until now, there have been two general categories of maintenance: Repair (fixing things when they break) and preventative maintenance (periodic cleaning and part replacement intended to prevent breakdowns). Usually, the best way to decrease machine downtime and keep labor and replacement part costs low is an approach which combines both categories. The IoT, however, is ushering in a novel, third category: predictive maintenance, a key enabler of smart production. To understand why predictive maintenance is such a game changer for manufacturing and the service industries, let’s summarize both in such a way that their shortcomings become clearer:

Repair: When a machine breaks down or seriously malfunctions, service technicians often are able to determine what needs to be replaced. Usually, this involves some troubleshooting, aided by diagnostic tools (like electrical meters) and manual observations. Even when this troubleshooting is successful, and the faulty component is identified, the reason why may still not be clear. Therefore, after the repair is performed, there’s sometimes no assurance that it won’t malfunction in the same way again.

Preventative maintenance: Since it’s so difficult to predict what could fail in complex systems like industrial machinery, service personnel make educated guesses about how long certain parts can last, and replace those regularly at some point before that estimated time. This approach can save the machine owner downtime if the maintenance schedule is adhered to, but machine failures still occur.

Since the schedule determines when parts are replaced, it doesn’t matter if the machine went through 1 cycle or 100,000; the part will be replaced anyway. One big assumption behind preventative maintenance is that the downtime due to these periodic replacements is less than the downtime that would have happened due to failure if they weren’t performed. Similar thinking is applied to the material costs: replacing relatively inexpensive parts now saves money in the long run by preventing the replacement of more expensive components. This assumption is largely unchallenged for specific maintenance activities on specific machines because real data is needed from the actual machine to either prove or disprove it, and that data generation and reporting ability wasn’t feasible until the advent of IoT. 

Why predictive maintenance is a real game changer 

Yet what if the repair technician’s diagnostic tools can be integrated into the machine itself, in the form of continuously operating connected sensors installed at multiple locations and in multiple sections or subsystems of the machine? What if all the data those sensors were generating were stored and analyzed? Such a collection of data would allow field service techs to pinpoint the root cause of the failure. 

What if all that sensor data was being analyzed and intelligently visualized during nominal operation, before any parts break and cause the machine to go down? Such analytics could identify the parts which really do need to be replaced, as well as spot the smaller anomalies in the machine’s data before they turn into a major malfunction. 

When all those “what ifs” become realities, the result is predictive maintenance, which uses IoT-generated data and predictive analytics to accurately forecast when certain parts will need to be replaced. Predictive maintenance avoids the downtime of repair and the waste of time and material of preventative maintenance. 

The result is less machine downtime from both preventative actions and malfunctions, fewer service personnel needed to support a given number of machines, less travel and lodging costs due to repair trips, and many other secondary cost savings. Another result is happier customers who are much more likely to buy from the OEM again, due to the reliability of their product. Using the rich behavioral data being collected from each instrument 24/7, the OEM can determine, order, and ship the correct replacement part to the field tech much sooner, further reducing downtime when repairs do need to be made. There are even health and safety benefits to the machine’s operators. Fewer breakdowns and less abnormal operation results in less risk to machine operators, supervisors, and everyone else in the vicinity. 

Predictive maintenance delivers financial dividends for OEM. Once customers perceive the added value of real-time performance data combined with predictive analytics, higher product prices become easier to justify.

The new era of data-driven smart production

This IoT-driven transformation of the service industry and the manufacturers who will adopt this smarter machinery will drive up production yields through decreased downtime, increase profit margins by decreasing preventative maintenance and repair costs, and drive the entire industry forward as the operational data of connected machinery informs the next iterations of product design.

By driving increased efficiencies, OEM revenue, and worker safety, the Internet of Things is producing a revolution in production.