All 14,000 NYC Taxis Could Be Replaced with 3,000 Rideshare Cars

A real-time algorithm illustrates all possible trip combinations and then assigns the best vehicle for each trip.

New York City's yellow cabs could go extinct - well, at least that's what a new study by MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) suggests.

MIT researchers, led by CSAIL professor Daniela Rus, developed an algorithm that crunched data from 3 million taxi rides to arrive at a surprising conclusion: The Big Apple's 14,000 taxis could be replaced by just 3,000 four-passenger carpooling vehicles.

Former CSAIL postdoctoral associate Javier Alonso-Mora, who was involved in the study, told Seeker that the MIT team set out to explore a solution that would lead to "a more efficient transportation system where taxis are available on demand with lower waiting times." He said a hope to reduce the traffic and pollution in cities also inspired the study.

The algorithm operates in real-time to direct ride-share cars to meet taxi requests people make using app-based carpool services, such as Lyft or Uber, and redirect idle cars to areas of higher demand. The algorithm creates a graph of all requests and the position of all vehicles. Then, it creates a second graph that illustrates all possible trip combinations and assigns the best vehicle for each trip.

The average wait time for each person would be 2.7 minutes - and the system could cut pollution and traffic by 300 percent, according to the study.

RELATED: Will Driverless Taxi Pods Help Solve India's Traffic Problem?

When rerouting cars to areas of higher demand, specifically, there was a 20 percent increase in the speed of service. On the flip side, people would have to be willing to share rides and accept the fact that their routes might be extended to make the system a success.

Either way, carpooling could be steering the future of public transportation, Alonso-Mora explained. "People use such services with two passengers in some cities, and I think ride-sharing will be more popular in the future with these efficient algorithms." He added that autonomous taxis would likely also play a big part in making ride-sharing more popular, and could be designed accordingly.

Of course, citywide implementation of a ride-sharing system could run the risk of putting thousands of taxi drivers out of work. But there were many perks for commuters and the environment.

For example, the team discovered that just 2,000 ten-person carpooling vehicles could meet up to 95 percent of New York City's taxi demand - paving the way for streamlined commutes, less pollution and a big drop in congestion The research article was published this week in the "Proceedings of the National Academy of the Sciences."

One of the biggest challenges of the ride-sharing solution: "That people are wiling to share their rides," Alonso-Mora said. "Technologically, it is possible."

WATCH VIDEO: Robo-Taxis and the Future of Self-Driving Cars