There’s no getting around it—flight delays are a pain. You’ve made plans wherever you’re going, or you’ve got a meeting two hours after you land, but no matter how you slice it, you won’t get there on time. It’s frustrating for everyone involved, just the sort of thing you didn’t need on your already-stressful trip.
Google has taken a big step in ending this frustration. While it can’t control the planes themselves, it can use its legendary troves of data to predict when they’ll be behind schedule. That’s exactly what it’s doing with the new delays feature in Google Flights.
Google Flights says it won’t just be pulling information from the airlines directly, but will use its machine learning algorithms to predict delays before the airlines themselves do. This sounds complex, but it’s not. Think about it. If, over the past six months, a particular flight route has been delayed 90% of the time, there’s a good chance that today’s flight will be delayed, too. If a high number of the other flights departing your airport have been delayed, that indicates that perhaps weather or an airport-wide system failure will affect your flight; if the incoming flight that yours is connecting with was delayed on its previous trip, that will affect you as well. To avoid a frustrating user experience by way of false positives, Google Flights doesn’t make its prediction until it is 80 percent confident that a delay will actually take place.
I type “BA28” into Google to take a look at British Airways 28, which is nearly finished its trip from Hong Kong to London. Google suggests that the flight will arrive thirteen minutes ahead of schedule, at 4:42am instead of the scheduled 4:55am.
Google Flights tracks flights in real-time, giving you an up-to-the-minute look at whether a flight is on time or delayed.
You can also see predictions of delays well before your flight takes place. For example, one of the results on a flight search from New York to Los Angeles is this Spirit Airlines flight with a stopover in Detroit. As you can see, Google Flights suggests that both the first and second flight are often delayed by 30+ minutes.
Google Flights will now predict whether a flight will be delayed, using a combination of big data and past on-time performance.
If you don’t already know how to use Google Flights, it’s worth taking a look at. It’s by far the fastest flights search engine on the web, and offers helpful suggestions whenever there’s an opportunity to get a better price. For example, I recently did a search for a flight from New York JFK to London Heathrow on September 21. As you can see, Google Flights immediately highlights in green the best price available on the route: $212 in mid-October.
Article source: https://www.forbes.com/sites/bishopjordan/2018/03/20/how-to-predict-flight-delays-using-google-flights/