Digitally enabled reliability: Beyond predictive maintenance
To capture everything digital can offer in increasing reliability and reducing costs, companies should boost their digital-maintenance ambitions.
Are we entering a world of smart machines that can warn their operators before they break down? Advanced predictive maintenance (PdM), enabled by extensive sensor integration and machine-learning techniques, is one of the most widely-heralded benefits of the fourth industrial revolution. The idea is certainly a compelling one, and it is encouraging companies in asset-intensive sectors to pursue investments in digital maintenance and reliability.
In our view, however, treating PdM as panacea for maintenance and reliability challenges may prove to be short-sighted. In part, that is because today’s advanced predictive techniques can only be practically applied to a subset of use cases. But it is also because an over-emphasis on one approach means companies won’t position themselves to capture all the potential benefits of a fully digitized maintenance and reliability function—one that’s focused on increased uptime and improved maintenance efficiency.
And those benefits are significant. Based on our observations of digital maintenance and reliability transformations in heavy industries, we see the potential for companies to increase asset availability by 5 to 15 percent, and reduce maintenance costs by 18 to 25 percent.
It’s easy to see why advanced predictive maintenance has been seen as a killer app for Industry 4.0. The approach combines many of the technologies that underpin the new wave of industrial digitization, such as networked sensors, big data, advanced analytics, and machine learning. It is a powerful technique that, by identifying complex patterns over hundreds or thousands of variables in ways that traditional analysis cannot, enables operators to develop a deeper, data-driven understanding of why failures occur. Most seductively, it promises a very tangible benefit: machines that don’t break down.
But in practice, economically viable, real-world uses for these advanced PdM techniques are less than universal. Where a machine is prone to a narrow range of well-understood failure modes, it is often possible to address a potential problem in a simpler way, for example by monitoring the temperature or vibration of a component against a set threshold, or by consistently and rigorously applying data-driven reliability analysis techniques to address the root causes of failure modes. Conversely, where a machine can suffer hundreds or thousands of different kinds of failures (some of them very rare), it can be impractical to create sufficient models of high-enough quality to adequately predict them all.
When factoring the effort and expertise required to develop accurate machine-learning models, model-based predictive maintenance becomes a breakthrough way to solve selected high-value problems, rather than the whole universe of maintenance opportunities. The approach has the most potential where there are well-documented failure modes with high associated downtime impact, for example in a critical machine on a larger production line. It also works well when it can be applied at scale to a large fleet of identical assets where there is sufficient reliability history to spread the development and management costs, as in offshore wind farms or fleets of locomotives. Thus, equipment manufacturers are strategically positioned to drive predictive-model development and deployment at scale for their end users—but these efforts have yet to materialize widely.
Does the relatively limited scope PdM has achieved mean that maintenance and reliability are somehow exempt from the digital imperative? Absolutely not. In fact, we propose that companies press well beyond one particular type of digital tool and think about how digital and advanced analytical techniques can transform their entire maintenance and reliability system.
