Armed with mountains of data, artificial intelligence is emerging as an important tool for airlines.
Armed with mountains of data, artificial intelligence is emerging as an important tool for airlines to find the ideal fare to charge passengers, helping them make as much revenue as possible as the industry emerges from its biggest crisis.
Fed by data on everything from Internet search and outbreak of Covid to weather soccer forecast and results, the computer is learning how everyday life affect the demand for flights. In its most advanced form, AI blows away complicated airline ticket codes and price bands that have had direct ticket sales for decades.
By weighing the data, technology providers can determine how much passengers are willing to pay for tickets and continuously re-price seats. According to Fetcherr, an Israeli startup that operates a live pricing engine, calculating fares using AI can increase an airline’s revenue by 10% or more.
Roy Cohen, chief executive and co-founder of Fetcherr, whose directors include Alex Cruz, former chief executive of British Airways Plc, said: “We can identify at any price. “It is very difficult to escape from a system like we. ”
Brazilian carrier Azul SA last month announced the first public test of Fetcherr’s pricing and demand forecasting technology. Azul did not respond to an email requesting more information about the trial.
According to Cohen, Fetcherr’s demand simulation is so accurate that fares determined by algorithms for flights six months away have barely changed at the time of take-off. “Almost in the right place,” he said. “Sometimes to the coin.”
Air needs all the help it can get. Travel has evaporated in 2020 as governments around the world close borders and impose Covid-19 restrictions. The recovery from the pandemic will bring global airline revenue to $782 billion this year, still below the $838 billion figure in 2019, according to the International Air Transport Association. typically have been in single digits since the financial crisis more than a decade ago.
While airlines have for years used software to manage airfares, what passengers ultimately pay is still governed to some extent by seat availability in other fare brackets. together. The AI seeks to compare fares more closely with passengers’ desire to pay, something that has become harder to determine after two years of shutdown.
Amanda Campbell, solutions marketing manager at global travel technology provider Accelya, said in an interview: “Traditional techniques are almost blunt tools for delivering certain products at a certain market price.
AI’s influence on aviation is in its infancy, but the streams of information are already too large to be sensibly captured. Cohen calculates that Fetcherr alone processes many petabytes of data from around the world every second as it increases the size of travel demand. One petabyte is estimated to be the equivalent of 500 billion pages of standard printed text. “The older we get, the better we get,” he said.
Conor O’Sullivan, product manager at Datalex Plc, a provider of real-time pricing services, says the supply of data is endless. Last year, the Dublin-based company announced a trial with Aer Lingus, the Irish airline owned by IAG SA. Aer Lingus did not respond to emails seeking details about the tests.
Datalex still relies heavily on historical information such as flight bookings and schedules to estimate current and future flight demand, O’Sullivan said. However, computers increasingly consider one-time events like concerts and sports tournaments, as well as hotel reservations and airport queues. Changes in government and policy, or even a minister being dismissed, can affect the market. The job of the algorithm is to determine the relative importance of each byte.
“All of these things work,” says O’Sullivan. “Then you come down to all sorts of behavioral psychology. If it’s raining outside, you’re more likely to book to a sunny destination than if it’s sunny. ”
While AI-powered giant retailers like Amazon. com Inc. clearly show the benefits of machine learning, the aviation industry’s inherent aversion to risk means it is likely to adopt technology at a much slower pace. Change in the industry moves at icy speeds, stymied by legacy networks and old ticket distribution partnerships.
“A lot of trust needs to be built before they start doing things like this,” O’Sullivan said. “They see this as really high potential value, but also high risk.”
Revenue on some routes can grow up to 8% with the benefit of AI-driven fare pricing, although across an airline’s entire network the benefit can be closer to 2%-3%, he said.
“If it rains outside, you are more likely to book to a sunny destination”
Frequent flyers can provide useful data – although it’s not personalized – when they log into airline websites to plan trips. Browsing sessions that don’t end with a reservation are sometimes just as useful as work sessions.
“How many searches are abandoned? You have to figure out why,” said Tim Reiz, Product and Technology Manager at Accelya. “It’s about finding the optimal price at which the airline can fill the aircraft to capacity.”
