3,000 Ride-Sharing Cars Could Replace Every Cab in New York City, MIT Study Says (theverge.com)
All 13,000 taxis in New York City could be replaced by a fleet of 3,000 ride-sharing cars if used exclusively for carpooling, according to research published today by MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL). From a report: Instead of hailing taxis, passengers that use ride-sharing services for carpooling may lead to reduced traffic congestion, pollution, and fuel use. The CSAIL researchers used public data from NYC taxi rides published by the University of Illinois to develop the algorithm. They calculated that 3,000 four-person vehicles travelling to similar destinations could meet 98 percent of taxi demand in the city with an average wait time of 2.7 minutes. Perhaps the most important part of the system is a dynamic repositioning of vehicles based on real-time demand, which makes the system 20 percent faster.
The major selling point of a taxi is that the backseat is all mine. Now I have to share a car with two other people, or a van with how many people?
No thanks.
This makes a ton of sense in NYC which is already saturated with high capacity rail systems. If you made these car share vehicles self driving and electric, you have the potential for an amazing last leg solution.
Ride sharing (zip cars, and eventually automated vehicles) will be the future, but people do need to be aware in such a future, people will most likely not "own" cars any longer. But for this to work, they can only be a last leg. Ride shares and self driving cars will NOT solve the transportation gridlock problem. Cars simply do not have the capacity of real public transit:
http://penguindreams.org/blog/self-driving-cars-will-not-solve-the-transportation-problem/
"All 13,000 taxis in New York City could be replaced by a fleet of 3,000 ride-sharing cars..."
Gee, I wonder how many jobs that will create in this new glorious economy.
"...if used exclusively for carpooling."
That's one hell of a caveat to put on these efficiency metrics, given the amount of times drunk people not needing a carpool to work use taxi cabs.
There's already 35,000+ uber drivers in nyc.
http://money.cnn.com/2016/05/11/news/companies/uber-new-york-city-union/
Yes, but they're not following MIT's mathematical model. They're following real life supply and demand, the bastards. If they stop listening to supply and demand and start following mathematical models of where people theoretically should want to go, we would only need 3,000 of them.
"That's the way to do it" - Punch
Sounds like they just came up with a plan to use a bunch of smaller buses to replace taxis. What a novel concept.
It's more to the point that rarely are people coming and going from IDENTICAL spots, isn't it?
And if I've just grabbed a cab, I'm obviously not wanting to wait 5 or 10 minutes to pick up the next 1 or 2 people even if they're within a few blocks, and same at the drop-off.
Sure, it happens people come and go from the SAME spot and are literally outside at the SAME time - in fact, I'd be happy if the algorithm popped up a message "hey that dude next to you is in your same boat, want to split this?"
But otherwise, fuck that extra 10-20 minutes, I could've taken a bus.