Uber wants to get better at predicting customer demand


editors note: Hear Danny Lange, Head of Machine Learning at Uber speak later this year at AI Worldwww.aiworldexpo.com

Uber wants to use machine learning to predict when a surge of people will be out looking for rides. The intention is to get more cabs on the road before surge pricing would normally kick in. That way, drivers will be ready and waiting for riders when the surge happens — and riders won’t be stuck waiting around.

Here’s how Jeff Schneider, the engineering leader of Uber’s Advanced Technology Center, put it during a recent data technology conference: “This idea is if you can predict that demand, you get that information out there — and you get that supply there ready for the demand so the surge pricing never even has to happen,” he said, according to NPR. Uber already does this to some extent, but Schneider says that Uber wants “to find those Tuesday nights when it’s not even raining and for some reason there’s demand.”

There’s been a bit of confusion today over what exactly that means. NPR, which conducted the interview with Schneider, took it to mean that — as Schneider says — surge pricing could simply go away. But following the publication of NPR’s article, Uber told TechCrunch that wasn’t quite the case.

“Uber is always looking for ways to better predict supply and demand in a city. But this story is not accurate,” Uber tellsTechCrunch. “We have no plans to end dynamic pricing. While we understand that no one likes to pay more for the same trip, it’s the only way to ensure that passengers can always get a ride when they need one.”

Read the source article at The Verge