AI Advances Toward MRO Potential

Artificial intelligence graphic
Credit: Skorzewiak / Alamy Stock Photo

Artificial intelligence (AI) tools, including natural language processing (NLP), are improving all the time. ChatGPT can now apparently mimic a human in responding to open-ended questions and even in writing like a fairly skilled wordsmith. Does this have implications for AI and NLP in aircraft maintenance?

Yes and no, at least for now, according to Vinay Kumar, senior vice president of engineering at ATP, which makes two very popular AI tools, ChronicX and Spotlight, to assist airlines in troubleshooting defects.

“We definitely have our eyes on the new wave of AI hitting the market, such as ChatGPT and Bing chat,” Kumar says. “I personally am of the opinion that this is a revolutionary step for a new way for the general public to interact with technology and simplify their lives.”

However, even for consumer uses, Kumar thinks these kinds of tools are still in early stages. “The technology has a long way to go,” he says.

Furthermore, “aircraft maintenance is not something that can be taken lightly and easily tested,” Kumar stresses. “The stakes are too high when safety and maintenance costs are on the line.” ATP understands the importance of its algorithms providing deep contextual and accurate results, which should lead to airlines and other operators getting more uptime.

Kumar says there is a potential for the new AI technologies to improve troubleshooting aircraft maintenance, “but it’s not there yet, and we’ll be keeping a close watch on it to see if and when it would be the right time to leverage as part of our solutions and product offerings.”

Meanwhile, ATP has been following its own technology roadmap.

ChronicX uses NLP to find an airline’s repetitive defects quickly so engineers can focus their efforts on fixing these most common troublemakers. ATP has now developed the capability to gather anonymized data from many airlines to spot defects that affect many carriers. ATP has now also put ChronicX on mobile devices.

Spotlight uses a combination of OEM data, NLP and troubleshooting experience to guide workers on the most effective troubleshooting path. It, too, has been put on mobile devices.

ATP has already begun demonstrating the data gathering capabilities of ChronicX and the mobile capabilities of both tools to select customers to get their feedback. Kumar says its customer base has continued to grow, from about a quarter of the world’s fleet 18 months ago to 30% of the feet now.