When the Large Hadron Collider discovered the Higgs Boson particle, scientists were giddy because it helped them prove the fundamental theory behind the Big Bang. Exactly how it did that is something only physicists can really understand. But, in layperson’s terms, the experiment was a triumph of data collection and analysis.
So, what does that have to do with mechanical and electrical (M&E) engineering? More than you might think.
Data is changing our understanding of the world around us in all sorts of ways, from the universe’s origins to the mundane, everyday details. Just as the scientists at CERN used a particle accelerator to see the world at an atomic scale, M&E engineers are now using rapidly evolving technologies such as internet of things (IoT) sensors, artificial intelligence (AI) and analytics platforms to understand what’s happening beneath the surface of buildings and assets.
For facilities managers, maintenance has always been crucial to ensuring compliance, high health & safety standards and business continuity. Now, however, these technologies are transforming maintenance into a more complex and strategic service. By collecting more accurate data on the condition and performance of assets, facilities managers can help their organisations make evidence-based decisions regarding real estate and workplace transformations, net zero targets and even employee experience or wellbeing initiatives.
But what do we do with it?
The most significant challenges with data are scale and ownership. There’s so much of it most organisations and business units don’t know where to begin.
Research by JLL in 2020 revealed that many companies believe smart buildings represent the future and some have even started collecting data, but a significant number don’t know what to do with it. Eighty-seven percent of facilities, security and IT managers said AI would become a necessary part of smart building management, and 77 percent of building managers already keep data generated from sensors in their facilities, yet 42 percent of those surveyed don’t analyse the data to identify variations and patterns to improve building operations. Additionally, more than half reported that they need third party support for tech implementation, tech strategy and system integration.
IT professionals know how to measure and read the data but struggle to apply it to real-life situations or business priorities. On the other hand, facilities managers may know what people want or the business expects, but don’t understand the data.
Out with the old
To exploit the data generated from buildings and assets, you need to understand how technology and data have transformed maintenance as a discipline. Historically, there were two types: reactive and planned preventative maintenance (PPM), the latter comprising statutory and scheduled works.
Both of these methods are effective ways of reducing downtime and saving on breakdown costs. But they can also be inefficient uses of either an organisation’s resources or an engineer’s time. Reactive maintenance can’t be scheduled according to the performance or condition of an asset. Meanwhile, PPM often leads to engineers carrying out works when the asset doesn’t strictly need it. The latter can also lead to the same maintenance schedule for newly installed systems and ones coming to the end of their lifecycle.
In with the new
Technology allows us to understand maintenance at a far more sophisticated and detailed level. Equipping assets with sensors that measure vibration and heat, for example, help engineering teams determine how they are performing and project when they are likely to fail, allowing them to make the decision to refurbish, upgrade or replace assets.
There are two core methods of data-led maintenance: predictive and condition-based. Both enable engineering teams to take a proactive approach to service delivery and reduce the need for both reactive maintenance and PPM, though they differ slightly.
Unlike preventative maintenance, which estimates an asset’s condition on average or expected life statistics, predictive maintenance uses a mix of real-time and historical data, enabling analysts to anticipate issues before they arise. This method also enables engineers to schedule corrective works only as and when they’re needed.
Likewise, condition-based maintenance relies on real-time data from sensors measurements, but this method alerts the system when the condition of an asset – e.g., the level of vibration – comes close to or surpasses pre-set parameters. Once again, engineers perform maintenance work only when needed, resulting in fewer unplanned downtime events and making maintenance schedules easier to prioritise.
That’s not to discount reactive or PPM; both still have their place in dealing with crises or unexpected events. Yet, as buildings become smarter, keeping them in optimal condition is increasingly a data-led undertaking.
Now that the economy is reopening, many companies have also begun to rethink their long-term real estate and people strategies. A permanent shift to flexible or hybrid working would place new demands on the operation and management of buildings. But there’s no precedent for change at this scale. That’s where predictive and condition-based maintenance come in. By adopting these data-led engineering techniques, companies could support these new work models, improve the efficiency and effectiveness of operations, and start to build a picture of this new world. Perhaps not to the detail of the Big Bang, but almost.
