CSAIL’s Daniela Rus and her team created an algorithm that crunched data from three million New York City taxi rides, calculating routes and schedules for two-person, four-person, and ten-person vehicles. The results showed that 3,000 four-person cabs could help handle 98 percent of the City’s demand (with a waiting time of 2.3 minutes), while 3,000 two-person cabs could handle 94 percent and just 2,000 ten-person vehicles could handle 95 percent.
“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 in a press release. “What’s more, the system is particularly suited to autonomous cars, since it can continuously reroute vehicles based on real-time requests.”
Many of today’s ridesharing systems, like those used by Uber and Lyft, are relatively inflexible when it comes to planning and assigning routes. In contrast, Rus’ system can rematch rides to different vehicles when appropriate and could prepare for high demand by sending idle cars to busy areas. The researchers said this lets the service run 20 percent faster than standard services.
“Ridesharing services have enormous potential for positive societal impact with respect to congestion, pollution and energy consumption,” Rus said. “I think 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.”
Rus and her team published an article titled “Ride-Vehicle Assignment and Analysis of the Benefits of High Capacity Vehicle Pooling” in this week’s Proceedings of the National Academy of Sciences.
