From the Flintstones to the Jetsons?


I’ve spent most of the past year traveling around the country talking to leaders in the military and the defense industry about the potential of predictive analytics to turn their machine data into actionable insight. We’ve seen the amazing things that visionary industrial companies in asset-intensive industries (like power generation, oil and gas, mining and manufacturing) have been able to do when they focus on getting real operational value from their data. It makes us passionate about the potential of industrial data and, for people who know and love the military, passionate about delivering that value to our defense forces.

It has been gratifying to see so much interest in data-enabled maintenance operations in the military—but there have also been some alarming tendencies that pop up occasionally. One that I’ve noticed in my travels is a desire to jump from the current generation of condition-based maintenance where, in the most likely scenarios, data is gathered from assets periodically and dumped into a data repository where it is never seen or heard from again—to some completely bleeding edge new state in which data is gathered wirelessly, processed autonomously using cutting edge analytics, and all of the implications (spare parts ordering, work order scheduling, and so on) are handled by computers. I haven’t heard anyone openly discuss using autonomous, self-directed robots to perform the maintenance work—but the aspiration flows logically from the other “requirements”…

Call me a skeptic but—well, I’m skeptical. From the Flintstones to the Jetsons in one RFP? It ain’t going to happen.

Humans are vital

My view is that, at least with current generation technology, the role of human subject matter experts is vital. Algorithms and analytics can be predictive, but the goal of this whole change effort is to be proactive, and only people can do that. Advanced statistical anomaly detection, like the patented algorithms embedded in GE’s SmartSignal solution, can’t turn a wrench or borescope an engine. The analytics can’t even cajole a skeptical maintainer into doing off-schedule maintenance on a piece of equipment that looks and sounds fine. No software is going to convince the shop supervisor to extend maintenance intervals on a subsystem. 

Equipment experts with dirt under their fingernails and the right data at their fingertips—that’s what it really takes to help drive change at the ground level. We have built a whole business around that merging of big data and big iron: we call it the Industrial Performance and Reliability Center (IPRC). Located in Lisle, Illinois, GE’s IPRC is staffed by “greybeards” who know industrial equipment inside and out and can help form the bridge between predictive analytics and proactive maintenance. Lots of our customers are so taken with the IPRC that they have us set up internal monitoring centers for them using our software and training services.

The fact is that condition-based maintenance is first and foremost a culture change initiative, not a technology fix—and choosing an analytics provider is really an exercise in choosing a partner for that transformational journey. Check out this video on the IPRC to get a better idea of what I’m talking about, and drop me a line if you want to chat about getting proactive.

Todd Stiefler

Todd joined GE from the world of Washington politics, and in no time at all has moved on to his second assignment, which sees him managing business development for the services GE is increasingly looking to offer to customers, including the Proficy SmartSignal predictive analytics software.

More Posts

Add new comment