When you think about it, most of an airline’s operations — from maintenance to crew management, booking to check-in, boarding and so much more — rely on data.
Like other segments within the travel industry grappling with generative AI hype, airlines and airports are working to uncover the uses cases that will drive operational efficiency and/or improve the customer experience.
So what are the concerns? What are the barriers to progress when it comes to data usage and generative AI developments?
Hype around generative AI is a big consideration according to airline, airport and travel technology executives, and there is a real sense of the importance of getting the basics right first.
At the recent World Aviation Festival in Lisbon, Xavier Lagardere, chief data officer of Lufthansa Group, stressed the need for good data.
“We’re in the hype of generative AI and one thing not to forget is that good AI comes with good data," he said during a session on using data. "In the current advances of bringing generative AI, generating efficiencies, driving personalized conversations with customers, we need to start with driving good data.”
Tristan Thomas, director of digital technology and innovation at Etihad, agreed and said generative AI had become an “everyday tool.”
“It’s an unusual position for a technology professional because you then have to work out how it is going to impact your world. It’s reaching an incredible hype moment, and as with all hyped technologies you need to go back to the point of what problems are we solving, that has to be the core of this," Thomas said. "We can remember being on this stage five years ago and talking about blockchain, and the dangers are there again where you start hunting down uses for a technology as opposed to what do I want to solve with this technology. I think it is life changing. We just need to work out how we use it.”
Lagardere added that one difference with generative AI versus blockchain is that “it’s in the hands of everybody at our companies.”
“So the wave that’s coming at us is not even top-down mandated or coming out of some fancy innovation department," he said. "It’s a wave that is coming from the users themselves, so in a way it’s not only different from an adoption curve standpoint, but it mandates us as corporations to bring a response and make use of it.”
Alongside the hype is also excitement about the possibilities, said Tara Mulrooney, chief technology officer and vice president of innovation at Edmonton International Airport.
“I’m really excited about generative AI, not just because of the tool, but it’s helping us learn how to be data citizens," Mulrooney said. "It’s pervasive, so if we can get everyone to learn those skills, especially our children and others, it’s what they’re going to be able to do with that capability and not have the struggles of collecting the data and managing the data.”
Though the relevance of historical data around consumer behavior was questioned as the pandemic subsided, airlines say it's not that the data is obsolete, but rather how and when they need it that has changed.
“It has become more necessary," Lagardere said. "What we’ve lost with the pandemic is the certainties we had. Last year there was a fantastic post-summer boom and everyone would say it’s pent-up demand. This year again, so maybe it’s the new normal, who knows? So we’ve lost this predictability, which means we need to be more synchronous with the signals that are coming live through the system, so it actually makes the data more relevant.”
Influencing outcomes
The good news is that airline teams now seem to be better at working together, with operations having greater influence where they might have had to cede to commercial teams in the past.
Mitrankur Majumdar, senior vice president and global head of services at Infosys, acknowledged the change.
“As new pressures start building up — whether cost related, sustainability or broader supply chain concerns, all of those have put pressure on the routing and scheduling abilities, and aircraft availability has also become a challenge. Put together, this has put huge pressure on the network optimization part," Majumdar said. "What this does is bring operations to have a higher influence in some shape, so a shift has happened. Earlier there was a lot of opaqueness in terms of the data sharing, but this whole realization has come in that for operational excellence and enhanced customer experience they have to transcend the silos.”
This doesn’t mean all the problems have disappeared with technical silos still hampering progress. Mulrooney blamed the products put forth by vending partners. "We are held hostage by proprietary closed architectures, so my plea to all of you is adopt open architectures and open standards, or we will look elsewhere for our software and solutions.”
A further cautionary note was sounded by Andrew Eliseev, global head of consumer and retail at Luxoft, who urged carriers to address their technical debt.
“The future of airlines is not just about all these AI use cases and fancy innovations in the customer experience," Eliseev said. "It’s also about the invisible but very crucial aspect of technical debt. It’s not just an IT problem, it’s an organizational challenge that has to be addressed.”
A recent Luxoft study found that 46% of executives believe technical debt “dramatically prevents them from innovating,” he said.
“Almost half are held back by outdated systems and technologies," Eliseev said. "I believe that is the main problem, and until we resolve this issue with legacy systems, you can’t build on top of all of these new technologies like AI and digital twins. So we’ll always be linked back to the lowest, weakest part of the chain.”
One way of getting round this is to have “kind of a two-speed IT,” according to Majumdar.
“The foundation has to be changed if you really want to capitalize on this kind of technology, but at the same time you cannot spend all your time on building foundations when the world has already moved on," Majumdar said. "So you have to have some kind of a balance. There are certain areas you can make some sort of progress because the world has already made some progress, you don’t want to miss the bus, but I agree the foundation has to be changed to maximize these things.”
Tags: Tristan Thomas, ETIHAD, Xavier Lagardere, LUFTHANSA GROUP, AIRLINE, AIRPORT, TRAVEL PLANNING, ARTIFICIAL INTELLIGENCE, Tara Mulrooney, TECHNOLOGY, GENERATIVE AI
