If there is one use case for harvesting device data that excited both equipment owners and equipment manufacturers, it’s predictive maintenance. The concept is that – by aggregating and analyzing data from network-connected devices and sensors – the equipment owner or equipment manufacturer could identify when a device was showing signs of failure and get it repaired or replaced prior to a catastrophic event.
This is a potential gamechanger – making it so that equipment failures that lead to lost productivity and revenue are effectively things of the past. In fact, analysts like Gartner have predicted tremendous growth in this area – anticipating that equipment manufacturers would ramp up investment to network-connect their devices and equipment owners would embrace gateways, sensors and analytics software solutions to make it a reality.
But, as Martin Keenan of Avnet Abacus laid out in a recent article in Design World, that reality hasn’t yet come to pass. Predictive maintenance hasn’t taken off like some anticipated.
To get more insight into why he thinks that predictive maintenance has yet to see massive adoption, and why he still thinks it’s the future of the industry – regardless of that fact – we sat down with Martin to ask him a few questions. During our discussion, we talked about why equipment owners have been reticent to pull the trigger on predictive maintenance initiatives, and how he anticipates that changing in five years.
Here is what he had to say:
Modern Equipment Manufacturer (MEM): In your recent article in Design World, you say that predictive maintenance was expected to be the “killer app” of IIoT. Why was that the case? What were industrial and commercial devices owners expecting to get out of predictive maintenance? What were the benefits they were hoping to realize?
Martin Keenan: As with any maintenance strategy the number one key benefit is to minimize downtime, as the cost of having production lines down while maintenance engineers repair equipment can be crippling. A better approach to predictive maintenance will maximize the lifetime of assets and decrease spending on maintenance products. So, the ideal strategy is to use technology to monitor the condition of equipment in real-time, scheduling in maintenance when it is needed before failures arise.
MEM: In 2018 Gartner predicted a massive increase in predictive maintenance by 2022. We’re obviously a few years away from 2022, but – in your opinion – are we on track to hit that $3.4B increase in spending they predicted? Why not?
Martin Keenan: We are definitely seeing an increase in sensing and communications technology being integrated into industrial equipment across all sectors. However, adding technology is just one step in transforming your production facility to a smart factory. To truly leverage the benefits of a smart factory, it will require more than simply investing in technology, organizations will need to change their entire approach to optimizing maintenance.
This is well understood by industrial production companies, so this may be holding back investment in some instances where they acknowledge that adding new technology needs to be complemented by new skills, new processes and, in some cases, new staff.
MEM: What is the difference between preventive maintenance and predictive maintenance? Why is preventive the forerunner of predictive?
Martin Keenan: There are similarities in that both are a planned approach, striving to avoid catastrophic downtime that can be mitigated by maintaining machinery before it fails. Preventative typically leverages maintenance software, which schedules jobs based on input on what the schedules should be to keep the equipment running effectively. It learns and adapts as the software is used over time.
Predictive maintenance takes this a step further, using sensing and condition monitoring to further increase accuracy. This is a more flexible approach, leveraging real-time data, further optimizing maintenance schedules.
MEM: What are the big hold-ups keeping equipment owners from embracing predictive maintenance?
Martin Keenan: At this stage it’s fair to say that the technology is available and accessible, both in terms of hardware and software. Existing devices can be retrofitted with sensing and communication capability.
The bigger challenge is in how companies must adapt to use the technology. Implementing a world-class maintenance system affects people, processes and budgets in many parts of an organization, making it seem a major step for some companies to take. But it is a step that most will take in due course.
MEM: What are some of the things that equipment owners can do to make predictive maintenance possible?
Martin Keenan: On the equipment side there are many ways to make equipment more intelligent without having to overhaul your factory. For example, installing sensing and condition monitoring devices that have communications capability will enable a better understanding of how your equipment is performing. IIoT gateways can help get that data efficiently to the cloud where it can be analyzed and acted upon.
MEM: How do you see adoption of predictive maintenance changing in the next five years? If we chatted in May of 2025, where would adoption be? How would equipment owners be utilizing predictive maintenance? How would equipment manufacturers be helping them?
Martin Keenan: I believe there will be steady progress in the coming years. Most equipment will have sensing, monitoring, diagnostic and communication ability and there will be clear benefits to leveraging this functionality. The software and data analytics will be even more mature by 2025.
Yet there will be some organizations, particularly those that are smaller in size, which remain reluctant to adopt an entirely new way to manage the health of their factory equipment.
MEM: We’re seeing an interesting movement in the HVAC industry towards what they’re calling HVAC as a Service (HVACaaS) – where the equipment manufacturer effectively manages and maintains their installed devices for the equipment owner for a recurring fee. Is this something that you could see taking off in other areas of the commercial/industrial equipment marketplace?
Martin Keenan: Absolutely. This will become a growing trend. A famous example in the maintenance world is Rolls Royce, who sells “uptime” – or operating hours – to the aviation industry and is seen as a pioneer, having introduced this approach in 1962. This is a trend we are already seeing in other areas, such as elevator companies selling a service to municipal enterprises. Some major industrial equipment providers are also moving in this direction, one example is Siemens.
We see the term servitization being used to describe companies going from selling products to selling a service or outcome. As the term HVACaaS indicates, this is a trend that has worked well for software. It will affect manufacturing and maintenance, lowering Cap-ex and maintenance costs.
This trend will likely extend into other industries, as well. For example, in the automotive and transportation industry, urban environments will have fleets of self-driving vehicles. In that environment, we may see car companies selling trips, instead of just cars.
To learn more about how access to device data can benefit equipment manufacturers and owners, click HERE to download a complimentary copy of the, “HVAC as a Service Revolution.”