The algorithm is more efficient than the ones used by Lyft and Uber.
Researchers from MIT have conclusively proven that not only is carpooling less polluting, but it is also a more cost effective and efficient solution as opposed to utilizing the services of traditional taxis.
Researchers from MIT’s CSAIL computer science lab have come to this conclusion after gathering data from 3 million taxi NY city taxi rides using 3,000 vehicles from NYC’s 14,000 strong taxi fleet, Lyft Line and UberPOOL.
Turns out by using a carpooling service, New York City could reduce congestion by three times and on average you would have to wait just 2.7 minutes for a ride, thus barely impacting your travel time.
This highly academic study assumes that you could convince 15,000 gainfully employed drivers to go ahead and get employed elsewhere. Apart from that assumption however, the results of the study shows that the ideas garnered from it could lead to a series of benefits, including, a more efficient transportation infrastructure, lower pollution levels, better traffic management and reduction in lost time, said researchers from the study.
With carpooling as its central theme, the MIT researchers created an algorithm that is a lot more efficient than the ones used by Lyft and Uber, which typically requires you to create a route while booking the taxi.
The algorithm however can reroute taxis on realtime, incoming requests basis, and can dispatch idle vehicles to high demand areas. In the process, carpooling services can be sped up by more than 20%.
“A key challenge was to develop a real-time solution that considers the thousands of vehicles and requests at once,” said Daniela Rus, a CSAIL Professor.
The algorithm first creates a graph of all requests and vehicles, it then calculates every possible trip combination to determine the best assignment. Vehicles without trips are dispatched to high-demand areas thus rebalancing the load distribution of the carpooling system.
What’s amazing is that the algorithm gets better in time: the more extensively it is used the greater is its efficiency.
“A system like this could allow drivers to work shorter shifts, while also creating less traffic, cleaner air and shorter, less stressful commutes,” said Rus.
However, the algorithm does not factor in the possibility of people wanting to ride by themselves, since the idea is to see the efficiency of a carpooling system, in the first place.
Another issue with this system is that many cities are against ride-sharing services that compete against taxis, for a variety of reasons, which include safety of the passengers.
Nevertheless, the point of this study is to envisage a more efficient future. As much as Lyft and Uber have changed urban transportation, the coming age of automated autonomous electric vehicles could shake things up even more.
“It’s important that we as researchers do everything we can to explore ways to make these transportation systems as efficient and reliable as possible,” says Rus.