Every day, the machines underpinning the industrial world become a little smarter.Thanks to the rise of big data, operators are eliminating unplanned downtime, setting new standards of interoperability and extracting additional value from heavy assets. Welcome to the industrial Internet of Things, where breakthrough levels of performance are just a software platform away.
While the use of big data to leverage better performance and utility from assets is a relatively new proposition in the industrial space, it’s a shift that’s already well underway on the consumer side. Amazon knows what you want to buy before you do. Target
can use existing data to determine if you’re pregnant. Even your car can automatically let you know when it needs a brake change. All this is powered by data. But when it comes to industrial equipment, operators are only now beginning to mine data for actionable insights.
This deficiency affects a broad variety of assets in industries ranging from aviation to energy. Unlike consumer hardware, multi-million dollar jet engines and power transformer don’t generally provide real-time visibility into their health. This deficiency plays out in a number of important ways, perhaps most significantly, in the way organizations plan for short and long-term maintenance.
“If I'm an operator, I'm traditionally responding to alarms as my standard way of operating, which is reactive,” explains Fred Schults, Principal Architect at GE.
“We want to move from a reactionary operation center towards a proactive and planning reliability center. This allows you to plan and optimize your maintenance.”
The problem with relying heavily on alarms is two-fold: first, alarms don’t always correlate to a true problem. An operator might dutifully measure oil temperature on a daily basis, even if that reading doesn’t have a significant impact on anything. Meanwhile, important data slips through the cracks.
Conversely, overreacting to alarms often results in unnecessary maintenance, which can have just as negative an impact on asset life. Common wisdom suggests, “if it’s not broken, don’t fix it.” But without data, that’s easier said than done. How does a technician know if an asset is broken without peeking inside unless there’s a system in place tracking real-time analytics and comparing them against historical readings?“
Every time you're doing maintenance on an asset or opening an asset up and inspecting it, you are potentially decreasing the life because there's a risk to that action. If we can minimize unnecessary actions and maximize the outcome of those actions so that they're done at the optimal level, then we're extending the life of an asset as well as lowering the operating expense,” Schults said.
To a large extent, the status quo approach to asset maintenance is the result of utilizing a plethora of piecemeal solutions all working in parallel to monitor and diagnose equipment. Until recently, a holistic solution for tracking assets across the organization simply wasn’t available, making it difficult to correlate and cross-reference readings, or mine for insights. But solutions like APM (asset performance management) are increasingly bringing operators into the 21st century.
“In the monitoring, diagnostic, and reliability management world, it's not like we're coming in and replacing another enterprise level solution,” said Derek Porter, General Manager of asset performance products for GE. “What's happening is, they're using handfuls of point solutions and cobbling the results together across those in order to give direction and make decisions.”
For many operators, the path to a more proactive approach to maintenance begins with an APM solution
With APM, industrial assets can enjoy the same level of data accessibility as consumer electronics, leveraging both current and historical analytics to paint a clear picture of the facts on the ground.
While maintenance no doubt continues to be a big cost for heavy asset industries, solutions like these are already saving tens of millions of dollars each year and promise an even brighter future.
How can big data benefit your industrial business?