18.3 C
New York
Tuesday, April 16, 2024

Airfare Prediction Algorithms Are Going Haywire

Flying has always been a pain: the endless security lines, the hip-checking other passengers to make sure there’s overhead room for your rolly bag, the squeezing into a seat, the ear-popping, the spotty internet, the boredom. But these days, the irritation begins much earlier, when passengers start to look for tickets.

The average round-trip ticket price in the US was $408 this week, up $100 from the same time in 2019, according to the airfare sales app Hopper. Part of that is pent-up demand from people sick of their homes after a still-not-over pandemic. Another is high fuel prices, spurred upward by the war in Ukraine. Another is a shortage of air-travel-industry workers. Factor in a waterfall of flight cancellations and schedule rearrangements, due to weather plus all of the above, and you’ve got a weird moment in air travel.

In normal times, people often turn to airfare prediction products to tame the madness. These tools—built by companies like Hopper, Kayak, Google Flights, Skyscanner, and others—are machine learning algorithms. These are one of the original Big Data projects. The platforms are trained on the arcane rules of airfare, plus reams of historical data, and use that to hazard when customers should buy to get the best ticket price. Generally, these tools will tell a prospective air traveler whether prices for their targeted route are high or low, or somewhere in between. And the more sophisticated ones will make a recommendation: Buy now, or wait.

But some unprecedented strangeness in air travel has led to unprecedented strangeness in price prediction, some executives say. That means that even the most tech-savvy buyers could be paying a little more than is optimal to take to the skies. For passengers, buying a plane ticket can feel like a mix of magic and luck, and the current unpredictability could add a touch more confusion—and frustration—when planning a trip.

Airlines establish airfares through art and science. An entire class of airline-employed data analysts, who work in a field called “revenue management,” work to anticipate who will want to go where when, and they set schedules, routes, and prices accordingly. Even after an airline prices out its schedule, the passenger sitting in seat 18A may have paid hundreds more for their trip than the passenger in 18B. This is often due to a system called “fare buckets,” where a group of seats will sell at one price. Once those go, another bucket opens up at a different price. Automated systems play their part here too: If one airline reduces prices on one route, another airline might pick up on the shift and immediately reduce its prices.

This makes airline price prediction a little like Spy Vs. Spy, with one gigantic system trying to predict what another system is going to do. For those building price-prediction algorithms, though, the results are usually consistent. “When data scientists work on trying to predict prices, they're looking at a black box,” says Oleksandr Kolisnykov, a content strategist at the software firm Altexsoft, which has built price-prediction tools. But in the end, that doesn’t always matter. “We actually don't know all the airline reasoning and factors that impact the current price, but we can observe the history and we can make some predictions.”

Most PopularBusinessThe End of Airbnb in New York

Amanda Hoover

BusinessThis Is the True Scale of New York’s Airbnb Apocalypse

Amanda Hoover

CultureStarfield Will Be the Meme Game for Decades to Come

Will Bedingfield

GearThe 15 Best Electric Bikes for Every Kind of Ride

Adrienne So

But the uneasy pandemic years have made all of this more complicated. Oren Etzioni is now the CEO of the Allen Institute for AI, but in the early 2000s he built—and sold to Microsoft—one of the first airfare prediction tools. Prediction algorithms are pretty good at reweighting the importance of different factors as the world changes, and, he says, “they have a shot at adjusting automatically by having the freshest available data.” But that can take some time, according to Etzioni: days, if not weeks.

Google Flights helps customers track down the least expensive tickets for their preferred routes and dates. But since spring 2020, the search engine has significantly cut down on the number of “predictive insights”—forecasts of when prices are likely to go up or drop—it offers searchers. In general, Flights aims for 90 percent prediction accuracy, says Eric Zimmerman, the director of travel products at Google. “With the increased volatility in airfares, it has become more difficult to reach that high level of confidence,” he says. The pandemic and its effects on air travel also pushed the company to halt an experiment launched in summer 2019, in which it guaranteed fares for some specific itineraries and would send flyers refunds if the price dipped before takeoff. It could bring the project back soon, Zimmerman says, as the industry starts to stabilize.

Giorgos Zacharia, president of online travel agency and search engine Kayak, says he has a team of MIT PhDs who spend their working lives tending to the website’s price-prediction tool. While the prediction algorithm, first launched in 2013, usually needs adjusting every few years, he says, the past two have seen “serious retraining” every few months, and sometimes every few weeks. He says that the accuracy of the prediction tools, which is generally around 85 percent, may have periodically dipped in the last few years—maybe closer to 83 percent. That means that, at some low points, waiting or buying when the website told you to was less likely to have led to the lowest possible price—and could have led, instead, to some light fist-shaking toward the sky.

“Machine learning likes to learn from old and past repeatable patterns, and make predictions based on the likelihood of those patterns working again,” Zacharia says. “So the pandemic, which brings a lot of unexpected outlier events, also affects the input data of models like this and makes it a more challenging environment.”

Hayley Berg, the lead economist at Hopper, says the company’s predictive tool is trained on 75 trillion itineraries and eight years of historical price data. But today the algorithm more heavily weights what it’s seen in the past three years, which has helped the tool maintain 95 percent accuracy throughout the pandemic, according to the company. Even in the first few days of Covid-related shutdowns, she says, Hopper got its airfare price predictions right 90 percent of the time. Still, customers shouldn’t be shocked by price volatility—Hopper has found that the average domestic flight changes price 17 times in two days, and 12 times if it’s international.

All those changes lead to plenty of conspiracy theories among ticket buyers, even those who don’t bother with price-prediction platforms. No, executives say, airlines aren’t tracking cookies and hiking prices if they see you’re interested in a certain route. (Zacharia, the Kayak president, does say that fares are occasionally higher or lower depending on your location when you’re searching, because the systems do take “point of sale” into account.) No, there’s no reason why flights would be cheaper on a Tuesday than any other day, a persistent rumor among bargain hunters. “The best time to book will depend on your trip, specifically the origin, destination, departure, and return,” says Berg. “And it can be wildly different depending on where you're going.”

Today, though, it doesn’t always take a sophisticated machine learning algorithm to pick the best time to buy—there is no good time. Prices are so high, says Victoria Hart, a spokesperson for Kayak, that there aren’t “many ‘wait’ indicators these days.”

Related Articles

Latest Articles