In a recent blog post, we discussed the direct relationship between predictive maintenance and sustainability performance. It is not a commonly drawn linkage, but the correlation between how we manage our assets and overall sustainability metrics could not be more clear.

Follow @paulleavoy

However, it’s one thing to want a predictive maintenance model, and quite another thing to actually implement it in a manufacturing environment. At times it can seem overwhelming. We have so much data to draw on, and between the rise of Big Data and the Internet of Things (IoT), it can be overwhelming to consider how to achieve true predictive maintenance, above and beyond the preventive maintenance we widely practice.

With that, let’s look at four key things we can do as we strive to implement a predictive maintenance approach across our organization, and at the same time—in the spirit of our previous message—consider how these ideas correlate to overall sustainability performance.

1. There is no turnkey solution you can buy

When we look at a lot of other enterprise software solutions, be it Enterprise Quality Management Software (EQMS) or Manufacturing Operations Management (MOM), we have a lot of existing best practices and very defined and widely adopted solutions we can implement and apply to our existing processes. Sure, we often need to make a tweak here, a customization there, but all told, a lot of these solutions speak to long-standing and widely adopted management standards and best practices.

It’s not so much the case for predictive maintenance. At this point we can’t seek out a vendor, implement a solution, and expect that—voila—somehow we’ll have an effective predictive maintenance model across our enterprise. Every manufacturing organization is so unique—from the specifics of individual fixed assets to the complexities of machine-to-machine interactions—that there’s no template we can simply apply and say, ‘here, this is predictive maintenance.’

That said, it is a very achievable journey to get there. At the core of many such systems we see today is Enterprise Asset Management (EAM) software that serves as the main enterprise application to manage maintenance activities. Then add-on applications—such as those provided by companies like Meridium and Ivara—are used to develop the analytics. These are not turnkey solutions, but they are examples of predictive maintenance models and a knowledge base that can be adapted to specific assets over time.

2. Pick the right assets

In the previous post I mentioned, we discussed how there are essentially three categories of maintenance in manufacturing: break-fix (reactive), preventive, and predictive. We analogized the manufacturing environment to the homefront by discussing how we deal with light bulbs on an ongoing basis. The metaphor can seem trite, but it can also help us figure out which are the right assets to build a preventive maintenance program from, and this is a great starting point for developing a predictive maintenance program.

Using the three classes of maintenance I mentioned can be a great way to define predictive maintenance priorities. Look around your manufacturing environments and consider which equipment you would treat on a reactive basis, as opposed to a preventive or predictive basis.

We have seen companies come in and try to provide the connective tissue between asset performance and preventive maintenance, but as we say, we are still a long way from any sort of turnkey solution to preventive maintenance. Such solutions may provide a checklist that you can map against your own specific requirements, but you really have to start from ground zero, especially if you have no predictive maintenance program in place.

So, start by categorizing your assets according to the three categories I mentioned. Begin with the assets that would benefit best from a predictive maintenance approach. These are the ones you will want to start with, especially if you are trying to demonstrate quick wins with senior management, from whom you may require endorsement for a more comprehensive predictive maintenance program.

3. The overwhelming goal is to maintain ‘optimally’

Remember the light bulb analogy we used in the first post on predictive maintenance? While it would certainly be considered responsible, it would ultimately be foolish to replace all the light bulbs in your home at a regular interval, in spite of the fact some were barely used while others had been used regularly. Likewise, we could also use predictive maintenance to analyze the fuel efficiency and oil quality in our cars, only to replace our oil every, say, month, as opposed to the manufacturer-recommended 7,500 miles. But that would be costly and inefficient.

It’s the same in manufacturing. ‘Overmaintaining’ can really be as bad as too little maintenance. With complex machinery, things go wrong. We have to tear equipment down from time to time, re-inspect, and figure out what went wrong. Sometimes things don’t always go back together as planned. The easy answer would be to overmaintain.

However, as with the car analogy, consider the cost of putting in oil every day, to go to an extreme example. The payback in preventive maintenance is not just trying to avoid failure; it is also trying to avoid overmaintenance, especially from a cost perspective.

4. Recognize the relationship between predictive maintenance and sustainability performance

As we discussed in the previous post, predictive maintenance and sustainability have inextricable ties. So, if your predictive maintenance program is in its infancy and your organization has (or plans to implement) a sustainability program, begin by tying these two operational and strategic objectives together.

Asset performance will ultimately have a bearing on overall sustainability performance. For example, if a specific piece of equipment is undermaintained and using more energy, well, that will impact overall energy usage directly. If a few vehicles in your fleet haven’t been maintained properly and have poor emissions as a result, guess what? That factors directly into overall emissions and asset safety KPIs when it comes to sustainability reporting and overall sustainability performance.

But for a moment forget about the role of a sustainability program for sustainability’s sake. The whole point of sustainability from a business/stakeholder perspective is that as we and our assets consume excess energy and water and generate excess emissions and waste, well, we pay for that. We don’t want to simply manage these impacts because they are bad for the world—while that is a noble intent. We want to reduce these impacts simply because it is bad for business to be wasteful. And in the ongoing goal of achieving operational excellence by mitigating consumption and waste, what better a way to start than by implementing predictive maintenance programs to operate profitably?