Paula Doyle likens engagement with Big Data to that of heading out to eat. We pull out our phones to browse the menu, check reviews and make a reservation. An app plots the swiftest route to the restaurant to avoid traffic or a different app summons a driver to take us there.

Such is existence for people in the 21st century. So why, when it comes to managing data, is industry stuck in the 20th?


Paula Doyle will participate in a panel discussion, “Evolution of Robotic Technologies in Energy Systems,” at CERAWeek by S&P Global on Tuesday, March 8 at 8:30 a.m.

“In the 1950s, industry was really a forerunner of technology, inventing the [programmable logic controller] and really driving forward with electronics primarily. Now, industry is a laggard when it comes to the use of tech, particularly software and the use of data,” Doyle, who is senior vice president of energy at Oslo, Norway-based Cognite AS, told Hart Energy. “So, while in our consumer lives, we’re fully equippedAI is just like a means for us to get aroundin industry, it’s really not.”

How bad is it? In many industrial plants, she said, data is still managed with pens and paper, which means that executives struggle to determine which of their systems needs attention, where the problem might be, even who owns the equipment in question. It’s a level of uncertainty that those executiveswho trust phone apps to make dinner reservationswould never tolerate in their personal lives.

Siloed data sources

Of course, much more is at stake involving data at an oil and gas facility than in determining whose jambalaya is truly the best in town. That might account for hesitation in committing to a significant change, but the higher stakes might also make embracing that change more necessary.

Paula Doyle (Source: Cognite AS)
Paula Doyle (Source: Cognite AS)

“We recognize that industrial data is problematic in its current state, but we also recognize that that problem needs to be solved if industry is going to achieve the thousands or the millions of optimizations that it needs to do to move from its present state, which is actually quite wasteful in terms of resources,” Doyle said. “And we can get there. There’s no doubt about it.”

It’s not that companies aren’t aware that better data management is needed to address issues like emission reduction and chemical consumption. What holds them back, she said, is access to their data. The root of that problem is their sources of data are siloed.

“Querying industrial data today is just an absolute nightmare,” she said. “You can’t look at it and say, ‘how many pumps do I have that failed in the last two years?’ That’s an easy question to ask but it’s going to take you a long time to get the answer. And the answer you get, most likely, is not accurate.”

That speaks to a prime challenge facing oil and gas companies: methane emissions.

“We have a problem with accuracy and we’ve a problem with good decision making,” Doyle said. “Obviously, in industry there’s a lot of decisions being made but they’re not the best possible decisions purely because people are driving a little blind in terms of being able to access the data to make the good decisions.”

Trust the engineers

Part of that may trace back to the enormity of the challenge. How does one open the floodgates of data, knowing that the next step is figuring out how to handle hundreds of thousands of bits of information per second?

Doyle advises executives to set priorities on what they want to accomplish and which areas really need improvement. That’s the easy part. They’re typically quick to identify what their problems are because they’ve been struggling with them, she said.

And don’t worry about who will put the data to use once a new system is installed.

“We have industries full of very well-educated technical people and engineers and others who are literally with their hands behind their back in terms of how quickly they can move and the kind of changes they can enact because they just don’t have the tools,” Doyle said.

The key is not just collecting but using the data to make a business impact. Achieving goals in the areas of net-zero, sustainability, efficiency and emission reductions almost always leads to a positive result in terms of dollars.

Doyle has spent a fair number of years in her career in R&D so she is familiar with Big Data’s enthusiasm spectrum.

“I’ve definitely been a perpetrator of technology for technology’s sake in the past,” she admitted. “But here, we’re really focused on technology for business impact.”