The implementation of artificial intelligence (AI) in the oilfield has transformed operations—optimizing efficiency and reducing environmental impacts. For many in the industry, this is to be celebrated.

But for some, AI’s benefits are more so seen as a last-ditch effort to save the industry.

North American shale, heavily influenced by the adoption of unconventional recovery methods, faces a dilemma, Omer Gurpinar, industry veteran and former technical director for EOR at SLB, told a group of panelists during a Q&A at SPE’s Annual Technical Conference and Exhibition.

“Nearly every field is in decline, nearly every perforation,” Gurpinar said. Remaining reserves are being squeezed out of brownfields, he added.

Efforts to improve production in subsurface reservoirs in North America have been insufficient, Gurpinar said. And, a significant portion of remaining reserves lies outside North America.

But David Johnson, digital solutions principal for Petrolink, pushed back, arguing that advancing unconventional methods is too costly.

“In my lifetime, we have gone from 38% to 40% to 42% recovery,” Johnson said. “Now we’re even higher than that, but we’re still leaving a lot of hydrocarbons in the reservoir because it’s not cost-effective.”

AI is the way forward, Gurpinar insisted, arguing that AI can accelerate laboratory decisions and enhance recovery factors.

“I think for our job, even with our white hair and old age, we need to say there must be a different way of doing things,” he said.

Producers are still stuck playing the hits, Gurpinar said. The energy industry is still using similar recovery methods to what was done in the 1960s, he said, adding that simply building a reservoir model takes two years.

Timely action is important, particularly for countries where the industry’s survival hinges on rapid technological adoption. AI-driven networks are already making significant strides in this area, Gurpinar said.

Faisal AlNughaimish, vice president and chief petroleum engineer at Aramco, stressed the importance of realistic expectations.

“When it comes [to] asking AI to increase that sort of recovery… that’s something more like sci-fi, I would say. It’ll not change nature,” AlNughaimish said. Some reservoirs in the Middle East can achieve recovery rates as high as 85%, while others struggle to reach 40% due to natural limitations, he said.

“We have to [manage] our expectation when it comes to technology, whether hardware or software, AI or physics-based, the physics will not change nature. It will not alter the capillary pressure of the rock or make unconventional reservoirs conventional,” he said.

Gurpinar still called for new and innovative approaches to improve production, suggesting that upgrading facilities and embracing new technologies could yield significant benefits.

“All I know is, including myself, many things we have done in the past are not applicable for the future. Accordingly, we should be very open minded because I know the pressure in the industry is so high,” Gurpinar said.

AlNughaimish agreed with the need for innovation but cautioned against over-reliance on AI. Other technologies, such as smart completions and improved fracking methods, will continue to play a crucial role in improved production methods.

“AI is part of a toolkit, but we should not forget about real nature,” he said.

Comments