As manufacturers move to more of a ‘service-based’ business model, they are becoming increasingly motivated to manage the complete lifecycle of an asset. As a result, we’re seeing how predictive maintenance can help in this journey—as this approach is applied to products, it can also be applied to high-value manufacturing assets. Take Rolls-Royce. Instead of limiting itself to remain a standard manufacturing-based enterprise, it has blurred the lines between product and service by selling ‘hours of flight’ instead of simply selling airline engines.
The linkage between selling flight hours as a service and selling engines as individual units relates directly to the increasingly pervasive role of predictive maintenance in manufacturing performance: the whole idea that above and beyond simply maintaining our assets on a preventive-maintenance basis, we can actually begin to predict when machines—based on a whole array of aggregate factors—require maintenance in order to operate optimally, avoid failure, and, in essence, mitigate costs on a long-term basis.
Moves like Rolls-Royce’s are illustrative of a broad shift we’re seeing across manufacturing industries. But while the whole impetus for preventive maintenance is so often tied directly to asset health and financial performance, in the long game it is linked directly to something much broader and—some would say—more important: sustainability performance.
In this article, we’ll discuss the linkages between asset health, preventive maintenance, and overall sustainability performance.
The Three Levels of Maintenance
For the benefit of those new to the concept of preventive maintenance, let’s step back momentarily from the manufacturing environment and look at something we all deal with on a recurring basis in a domestic environment: the light bulbs in our homes. Light bulbs are used on a daily basis in most cases, and as assets that deplete through use, they need to be replaced. We can use this lens to better understand the three levels of maturity when it comes to maintenance.
- Reactive or Break-Fix Maintenance: For light bulbs in a household environment, this is the most widely adopted model. We notice that a kitchen or living-room light has gone out, then we find another bulb, or we go out to a store to get a new one, and we replace the fixture in question with a fresh bulb. Beyond the realm of home lighting, while this approach can be spotted in low-maturity manufacturing environments, it is obviously far from optimal in any manufacturing environment. Critical machinery needs to be operating to spec at all times, at minimum.
- Preventive Maintenance: To counter the reactive model, we have the mainstay of the mature manufacturing environment: the preventive maintenance model. While in most cases we wouldn’t use this model on the domestic homefront, in an office or plant environment it makes a lot of sense. Instead of waiting for light bulbs or an array or light bulbs to go out, we refit a whole set of fixtures on a regular basis, at regular intervals, with new bulbs. Some light sources might have been rarely used, while others might have been used on a more exhaustive basis. However, by treating the set as a whole, we’re mitigating the chance one or more bulbs will die in between maintenance intervals, thereby mitigating the need to respond ad hoc to one or more minor failures, and as a result reducing associated costs.
- Predictive Maintenance: This model takes the latter approach one step further. What if we could have detailed data on energy use from certain fixtures, at certain times? What if we tracked the flickering and fading of individual bulbs? What would happen if we could break the performance of our fixtures down into finely nuanced details, and track that information accordingly? Well, if we had the right analysis software we’d probably be able to determine exactly when our lights were going to go out, as well as why and how.
Surely implementing something along the lines of the latter example would be absurd from a home-lighting perspective. But it is illustrative of why large, forward-thinking organizations are starting to look at the health of their manufacturing assets in a detailed way. But just as we think of light usage from an energy-cost perspective, often organizations think of predictive maintenance solely from a bottom-line perspective.
However, there’s another enormous factor here at play: ultimately, all of our assets—from light bulbs to complex automated machines—need to be maintained, yes, but they also have direct collateral impacts on that huge performance metric the business world continues to discuss: sustainability. Yes, sustainability tends to connote environmental performance—i.e. how many emissions is a certain unit producing, or, how much energy is another unit consuming? But more broadly it speaks to all the aspects of business we need to assess and maintain to stay alive and thrive as organizations: environmental, energy, health and safety, and, yes, social responsibility.
EHS, asset, and sustainability performance are inextricably linked
You might think social responsibility is a stretch in this equation, but consider the fact that a poorly maintained machine sometimes doesn’t just fail to maintain optimal performance. Sometimes a poorly lubricated machine, or a machine that hasn’t had the right parts replaced at the right intervals, actually bears the possibility of exploding or falling apart entirely with a worker in its midst. Think of the health and safety hazards this creates.
This happens regularly, and while the examples are unfortunate, they point to the direct collateral impacts we see in not only environment, health and safety (EHS) performance, but also asset health management, manufacturing operations, and ultimately sustainability performance management, for sustainability is as much about the safety and health of the individual worker as it is about environmental performance at large.
Now that we have set the stage for predictive maintenance from a sustainability perspective, in an upcoming post we’ll get into the key recommendations you need to know when developing a predictive maintenance program.