Will Predictive Maintenance Be an Engine for Economic Transformation?

There’s a still lot of “gee whiz” around the Industrial Internet, and that’s okay. In some ways, it feels like the early days of the space program – people weren’t exactly sure how it would all turn out, but they had believed in the technology and they shared a common vision of the future.

There’s one significant difference, however. We’re already seeing positive and tangible results from our efforts. For example, applying predictive analytics to data from the Industrial Internet enables us to offer products and services that we couldn’t afford to offer in the past. This is especially true in critical areas transportation, aviation, oil and gas, and power generation.

The key capability here is predictive maintenance or Asset Performance Management (APM), which gives us the ability to lower the overall operating costs of machines and devices by fine tuning their performance. That’s a big deal, because when we significantly reduce operational costs, we open the door for a whole new generation of products and services that would have been far too expensive in the past.

APM blends the power of predictive analytics with the scale of big data to detect potential maintenance issues before they occur. Why is that important? Because if you’re operating gas turbines, diesel locomotives, aircraft engines or oil drills, you don’t want those devices to fail unexpectedly. Knowing in advance when a vital component is likely to break means you can replace it on your schedule – which is much better and much less expensive than being forced to replace or repair something in the middle of the night or during a hurricane.

Mark Grabb, the technology leader for analytics here at GE Software Research, sees a fundamental economic shift emerging from the wider use of APM. “When you can provide services at efficient price points and still generate positive returns on your investments, you can offer services that would never have been offered before, simply because they were too costly,” he said.

From Mark’s perspective, predictive maintenance allows us to apply the “long tail” concept to industry. Usually, we associate the “long tail” with the publishing business, where it refers to the economic value of a publisher’s backlist, which includes all the books the publisher has in print, not just the recent bestsellers. A publisher’s backlist is the real source of its economic strength, because it costs less to sell copies of The Time Machine than it costs to sell copies of the newest vampire romance.

Applying “long tail” economics to manufacturing, energy or transportation means that more services that are not “bestsellers” can be offered at price points that are reasonable for buyers and profitable for sellers.

I find this idea extremely exciting, because it elevates APM into a force for economic transformation. Thanks for the work of Mark and his team, algorithms for APM / predictive maintenance is already a practical reality. More than 88 algorithms have been created in the past three years and implemented into software solutions for the businesses. I think it’s also very cool that these types of software-based services offer the potential to make the world a better place by allowing companies to bring new products and services to market more quickly, efficiently and profitably.