What is Predictive Maintenance in OEM?

Reviewed by David Carmichael, Solution Engineer (January 2024)

There are different schools of thought for maintenance in original equipment manufacturing, and predictive maintenance is one of the most effective. It uses condition monitoring, Internet of Things (IoT) technology, and specialised software to monitor asset health.

With predictive maintenance in OEM, you can detect equipment problems in their initial stages and address them before they become much more costly issues. This avoids major equipment failure, thus reducing downtime and improving safety.

The regular checkups of preventive maintenance are an improvement over addressing problems after they happen (corrective maintenance), but often require more financial and time resources than predictive maintenance.

Preventive maintenance schedules equipment for service (which means downtime) using fixed periods like dates or operating hours. This is indeed a robust OEM maintenance system that can find and address all equipment issues. There are less frequent maintenance costs with predictive maintenance vs preventive maintenance, though. With the condition-based monitoring of predictive maintenance in OEM, you can ensure your assets get only the maintenance they need and when they need it.

Key predictive maintenance examples in OEM include:

• Vibration monitoring: Monitor whether high-speed rotating equipment is producing the expected frequencies. Vibration irregularities on a driveshaft can indicate imbalance, misalignment, damage, or loose equipment. Addressing these when they begin avoids the extreme costs of shaft replacement
• Infrared thermography: Keep eyes on your equipment temperature with strategically placed infrared cameras. They can detect electric circuit anomalies, deterioration like wear or rust, inadequate heat transfer in HVAC systems, and problematic hydraulic system hot spots
• Oil analysis: Hydraulic oil is the lifeblood of countless industrial processes, but it needs to remain clean and cool to do its work. By continually sampling and analysing your oil’s impurities, you can detect recent component damage or determine if the oil is causing system damage. This is more labour-intensive among the predictive maintenance examples since it involves continually sending samples to labs for analysis rather than constant condition-based monitoring, but its benefits to equipment life are well worth it
• Motor current signature analysis: Keep a constant watch on the current draw and profile of your motors. This system can spot developing shaft imbalance, bearing defects, and static and dynamic eccentricity
• Acoustic analysis: This can detect similar issues as with vibration monitoring, but it measures the sound equipment produces rather than directly monitoring vibrating parts. By recognising the telltale sounds of specific equipment damage, predictive maintenance software can alert technicians that bearings need lubrication or replacement, for example. Of course, technicians themselves can hear and respond to the sound of damage, but with the addition of acoustic analysis you’ll have data for ongoing maintenance improvement
• Quality control: By monitoring the dimensions of parts as they’re produced, predictive maintenance software can identify defects in manufacturing equipment. This allows the OEM to address the issue and reduce defective output

Ideally, any existing electronic instrumentation you have monitoring your machines will adapt to your predictive maintenance system. This can soften the costs of implementing predictive maintenance as an original equipment manufacturer.

Either way, though, predictive maintenance in OEM has high up-front costs but higher long-term benefits. It requires investing in Internet of Things devices and predictive maintenance software and developing internal procedures and training for using them.

Internet of Things devices are pieces of equipment that monitor conditions using instruments and can send that data over the Internet for analysis and action. OEMs will likely be familiar with using electronic sensors for things like temperature and pressure, and monitoring them through digital displays or human-machine interfaces (HMIs). Internet of Things devices go further by linking their data to a plant’s broader computerised maintenance management system (CMMS), allowing predictive maintenance to collect operational data on the device and extrapolate the maintenance it needs.

Not all equipment is ready-made as IoT devices, but adapting or replacing your existing instrumentation to link to your CMMS can convert them into having that capability.

Predictive maintenance software often falls within an OEM’s CMMS. A CMMS plans when company assets need to go offline for maintenance by tracking how often they operate (preventive maintenance). With predictive maintenance tools in a CMMS, you’ll gain deeper ongoing data analysis and alerts about issues as they arise.

Once you connect assets to the system, you’ll need to establish normal-state values for all the conditions you’re monitoring and what deviations are concerning. The software will monitor these properties and consider their values in the context of the asset’s age and its operating conditions and frequency.

When the software has enough data (lots and lots of data), it can identify when machines are showing abnormal behaviour and alert you of the need for maintenance.

Browse RS’ maintenance solutions to learn how to realise the many benefits of predictive maintenance in OEM at your business.