NYU Tandon professor Joseph Chow thinks we’re only at the very beginning of an age of smart transportation, where big data will allow us to create systems that much better match up supply and demand, and deliver better services.
“I’m trying to harness the information we have to optimize the services we have,” Chow explained by phone last week.
The National Science Foundation awarded Chow and his colleague at Tandon, Constantine Kontokosta, its Faculty Early Career Development Award, or CAREER award, earlier this month. (Editor’s note: The second part of this piece, on Kontokosta, will be published tomorrow. Update: Here’s that piece.) The National Science Foundation describes the CAREER Award as the “most prestigious award in support of the early career-development activities of those teacher-scholars who most effectively integrate research and education.”
With each of the modes of transportation now digitized and recorded, researchers are now able to light up the tunnel of otherwise unknowable transportation service. Take cars for instance. In a mildly controversial rule change earlier this year, the city demanded Uber, Lyft and other ride-hailing apps to make certain trip data publicly available. This includes bits of information like where each ride started and ended, and how long that took.
This visualization is only from yellow and green cabs, but here’s what that could look like to a researcher:
— Will Geary (@wgeary) March 27, 2017
“When people talk about congestion, people think you can get rid of it,” Chow explained. “No, there will always be congestion. [We have] yellow taxis that focus on manhattan but can go into boroughs and green taxis in the boroughs and they have different fleet sizes. There are decisions to be made about how much supply we need to use to meet demand. Transportation is a very dynamic thing, and changes at different times of day.”
One of the reasons why the new rule about Uber and Lyft’s data sharing is that if it’s in the city’s systems, it’s subject to be made public under the Freedom of Information Act. That means someone could, if they want to, go through the data and find out from which addresses people were picked up and dropped off to, and when. Likewise, the ridesharing companies don’t want to release this data for fear that their competitors could figure out how their algorithms work. Chow is working on a mathematical model that will take the data and randomize it while still allowing its usefulness to persist. He plans on presenting the model at IEEE in Japan in the fall.
Knowledge is power!
Subscribe for free today and stay up to date with news and tips you need to grow your career and connect with our vibrant tech community.