January 4, 2017
Just 3,000 ride-sharing cars could replace 14,000 taxis in New York City
Ride-sharing companies like Uber and Lyft have completely disrupted the transportation industry and according to a new study from MIT, a fleet of 3,000 ride-sharing cars could completely replace the nearly 14,000 taxis currently serving New York City.
To reach their conclusion, the study team designed an algorithm that revealed 3,000 four-passenger cars could replace 98 percent of taxi demand in New York City, at an average wait-time of just 2.7 minutes.“Instead of transporting people one at a time, drivers could transport two to four people at once, results in fewer trips, in less time, to make the same amount of money,” study researcher Daniela Rus of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), said in a news release. “A system like this could allow drivers to work shorter shifts, while also creating less traffic, cleaner air and shorter, less stressful commutes.”
The team also discovered 95 percent of current taxi demand would be absorbed by only 2,000 ten-person vehicles, as opposed to the almost 14,000 taxis that currently work with New York City.
Computer Models of New York City
To develop their algorithm, the CSAIL researchers used public data from NYC taxi rides provided by the University of Illinois. In a computer model the team developed, the algorithm operated in real-time to direct cars according to incoming requests. It also sent idle cars to locations with high demand, a measure that accelerated service 20 percent, Rus said.
“To our knowledge, this is the first time that scientists have been able to experimentally quantify the trade-off between fleet size, capacity, waiting time, travel delay, and operational costs for a range of vehicles, from taxis to vans and shuttles,” Rus said. “What’s more, the system is particularly suited to autonomous cars, since it can continuously reroute vehicles based on real-time requests.”
Uber and Lyft have utilized smartphone information in a manner that has made ride-sharing an affordable, handy option. However, current strategies are still very simple. For instance, some ride-sharing programs mandate that user B be along the way for user A, and require to have all the calls entered before they can develop a route.
In comparison, the MIT system makes it possible for calls to be rematched to several vehicles. It can also evaluate an array of various kinds of vehicles to figure out which would offer the greatest benefit.
The MIT system uses something called “integer linear programming” to determine the optimal assignment of vehicles to trips. After vehicles are designated, the system can then reconfigure idle vehicles, sending them to higher-demand zones.
“A key challenge was to develop a real-time solution that considers the thousands of vehicles and requests at once,” Rus said. “We can do this in our method because that first step enables us to understand and abstract the road network at a fine level of detail.”
The MIT team said they were able to produce an “anytime optimal algorithm,” meaning it gets better the more times it is run.
“Ride-sharing services have enormous potential for positive societal impact with respect to congestion, pollution and energy consumption,” Rus said. “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.”
Image credit: Thinkstock