New York City Autonomous Transit

NYC Autonomous Transit Web2-01
Link to PDF

New York City Autonomous Transit (NYCAT) is a transportation strategy that would unify the city’s disparate Subway, commuter rail, and highway networks into one system focused on shared autonomous vehicles. New York’s vehicular transportation systems are disconnected; Subways and commuter rail are run separately and require separate fares, and highways are used primarily by private vehicles as opposed to transit. Each of these systems can also be very unreliable. Subways often get delayed, commuter rail service is infrequent, and highways and roads are subject to wild fluctuations in traffic speeds, slowing down cars, trucks, and buses alike.

Under NYCAT, the rails of Subway and commuter lines would replaced with paved roads, and new ramps would connect them to nearby highways and streets. All roads and highways would be priced to promote the use of shared autonomous vehicles when they’re busy; they would become an integrated network of ‘minibusways’ offering consistent travel times, frequent service, new connections, and faster journeys.

A Brief Introduction to the Minibusway

A minibusway is a road designed to ensure a smooth flow of traffic of predominantly shared, high-capacity autonomous vehicles. Like trains, these autonomous minibuses (AMs) would be space-efficient, high capacity, affordable, and protected from traffic. Like taxis, they would be versatile, offer journeys with few or no intermediate stops, and could run on regular streets and highways. Unlike taxis, they’d pick up and drop passengers within a short walk of their destinations, not always right at the door.

A Protoypical Autonomous Minibus

Unlike trains, minibuses wouldn’t be confined to one line. To emphasize this distinction, I refer to the corridors minibusways travel on as ‘routes’ rather than ‘lines.’ A minibus would change routes like a car would change between highways. For example, the D Train always runs from the Bronx to the 6th Avenue Line, but a minibus might start on the D Route in the Bronx (not shown) but then switch to the A Route in Manhattan to deposit passengers along 8th Avenue in Midtown.

NYCAT Minibus Switching from A to D Route
West side of Manhattan with the Minibus Journey Described above in White: The white circles with black borders indicate interchanges. The A Route continues down 8th Ave, while the D Route turns east.

Many minibusways would have passing lanes, allowing AMs to bypass others that are stopped at stations. People going to the same station or area would cluster into the same minibus, which would then proceed there with few or no stops in between. This would make journeys a lot faster (see table below); every Subway stop skipped would save about 40 seconds, and that can add up on longer journeys. A non-stop minibus journey would typically average 60 mph, compared to 16 mph and 20 mph respectively for local and express Subway trains.

mbw 2 passing lanes
Minibusway with Two Passing Lanes (i.e. four lanes total at stations)

In New York City, any Subway line with 3 or more tracks (or room for that many) would allow for passing, representing about 70% of the network. This means that NYC could benefit even more from minibusways than cities whose subways typically have only two tracks.

This design allows for up to 268,000 people (65 people per 20-foot minibus) to travel in the through (passing) lanes because produces a continuous flow of vehicles, plus over 100,000 more in the local lanes if those extend between the stations (as they do in Subway tunnels). By comparison, the 4/5/6 Trains together can carry up to 93,000. Plus, with up to 4,100 vehicles per hour running in the through lanes alone, no one would have to wait long for a minibus, so not only would journey times be shorter; they’d also be more consistent.

Crowded NYC Subway Train.jpg

On routes that are too narrow for passing lanes, such as the L Train Bullet, minibuses would travel in caravans that resemble trains. However, even in these cases, minibuses could move more than three times as many passengers as trains can—138,000 versus 40,800 in the case of the L because the caravans could travel closer together. Thus, a well-functioning minibusway system would go above and beyond solving the Subway’s capacity issues.

Minibuses would have stations in some places and ramps to street level in others. In dense neighborhoods, it would likely be disruptive, expensive, and impractical to have highway-style ramps onto and off of the minibusway. So there wouldn’t be any ramps in the heart of Midtown, but there would be some in places with more room.

Minibusways would be designed to prevent congestion using a dynamic tolling system (a form of decongestion pricing) where to use the minibusway, users have to pay a toll that changes based on demand. In this system, the per-vehicle price rises until it reaches a point where the road is operating exactly at full capacity. This system would be applied to every highway, former commuter rail line, Subway route. On any busy route, people would naturally gravitate toward sharing rides; a $20 toll is much more affordable if you can split it 40 ways. AM fleet owners would respond by putting space-efficient vehicles on these routes. 

I go into much more detail on minibusway design in this piece, I discuss some additional benefits of the system here, and I discuss potential reasons why minibusways don’t exist yet here.

What NYCAT Would Accomplish

Speeding up Journeys

NYCAT Travel Times 6


By allowing vehicles to run on the most direct routes without facing congestion, NYCAT could drastically improve mobility in and around New York City. On every one of the routes shown, NYCAT would be at least twice as fast as today’s transit, while still offering affordable, shared rides. It would improve commutes to the traditional job centers in Manhattan, as well as journeys throughout the region.  Methodology for the table

Making Transportation More Reliable

NYCAT would be less prone to delays and it would be more resilient when things do go wrong. The Subway suffers from years and decades of deferred maintenance. Much of its infrastructure is being used long past its intended life, and failures with switches and tracks are common; from June 2018 to May 2019, they caused 45% of all major delays. NYCAT would allow New York City to start fresh by using paved roads instead of rails, and far less infrastructure would be needed to power and direct vehicles; AVs would be self-powered and self-guiding.

F Train Derailment
Scene at the F Train Derailment in Woodside, Queens in 2014.

NYCAT would also ensure that other incidents, such as medical emergencies and vehicle breakdowns, would have minimal impacts on people’s journeys. AVs are much more flexible than trains; if a stopped vehicle or other disruption is blocking a lane, they can easily switch lanes to go around it. Moreover, AVs can instantly reroute toward other corridors that aren’t disrupted. Simply raising the price to travel on a route with a blockage could redirect enough AMs to keep traffic running smoothly and allow people to get where they’re going with little delay.

NYCAT would also provide more frequent service than current transit. When you don’t need a driver or conductor in every vehicle, you can afford to make the vehicles smaller and run them more frequently.  Supposing 30 minibuses = 1 train,  you could provide the exact same capacity with up to 30 times the frequency; one train every half hour translates to a minibus every minute. Even overnight, minibuses could easily run every 4 minutes or fewer at stations on any major transit corridor; travelers would no longer risk waiting 20+ minutes at a platform.

Lowering Pressure on Housing Costs

NYCAT would make it cheaper to live and work in the New York City metro area.  Faster journeys would allow people to move further away from their jobs while retaining a short commute, just like newly built Subway lines did a hundred years ago. For example, someone living in Harlem and commuting <20 minutes to Columbus Circle via Subway might find it preferable to move to somewhere like Parkchester in the Bronx, where median rent is currently 41% lower and where, thanks to NYCAT, their commute would remain under 20 minutes (today, it’s 56 minutes by train.)

Median Rent by Census Tract, NYC
NYCAT would allow people to spread out more into lower cost areas and areas where it would be easier to build more housing.

People who would prefer living somewhere farther out would be able to do so more easily, and those who like their dense, lively neighborhood near the center would enjoy lower rents because fewer people would be jockeying to live there. NYCAT would not solve New York’s housing crisis—problems with rent control and zoning would still constrain the supply of housing both in central areas and the periphery—but it would help a great deal.

Lightening the Financial Burden of Transit

Metrocard being swiped

NYCAT would allow the MTA, or whatever agency might succeed it, to focus on maintaining infrastructure, while allowing private firms to compete in providing service. Fares cover only 53.7% of the MTA’s operating costs (i.e. the cost of running trains and buses, not maintaining infrastructure), with the rest of the money coming from taxpayers. This is better than most US transit agencies, and it isn’t to say that transit service isn’t worth the cost in other respects. However, it remains true that the MTA is not the most efficient organization and that funding is costly and politically difficult. 

Private AM fleet owners competing to serve New Yorkers could likely provide better, more flexible service at a lower cost. Plus, minibusways would lead to fewer vehicles running with empty seats, more affordable vehicles (#6), and lower maintainance costs. The city and state could use the extra cash for anything from subsidizing rides for people with low-income, to cutting taxes, to funding schools.

Providing service to ‘Transit Deserts’

NYCAT would especially improve mobility for city residents who do not currently live near Subway stations. All major streets in New York City would be dynamically tolled and therefore largely free from congestion. AMs could travel on these streets much quicker than city buses do today, and they could also make use of highways and rail corridors. In some areas, such as the eastern part of the Bronx, people would gain access to many more high speed transit routes.


Facilitating Outer-Borough Trips 

A major limitation of New York’s Subway and commuter rail networks is that they are primarily radial systems, meaning that most lines point toward Midtown and/or the Financial District. This makes other transit journeys, such as an east-west Bronx trek, longer and less consistent. The bus network addresses this to an extent, but city buses are subject to often poor traffic conditions and tend to be much slower than trains (6.6 mph vs. 16-20 mph). As a result, many destinations are poorly connected by transit, even if they each lie in the same borough.

The Verrazzano-Narrows Bridge seen from Brooklyn

The city’s highway network is a lot more useful for outer-borough trips. Three highway bridges run between Queens and the Bronx, three highways connect Queens and Brooklyn, and one highway bridge connects Brooklyn and Staten Island (the latter borough’s only non-ferry link to the rest of the city). Once decongested by dynamic tolling, these routes and arterial roads would serve as high-speed transit links. This would expand job access, support a vibrant economy, and better tie the city together.

Dealing Death to Dead Ends 

NYCAT effectively extends Subway lines for minimal cost by connecting them to highways and rail lines. Subway extensions have proven slow and costly in NYC, to say the least. NYCAT focuses instead on connections; before considering where to expand, it’s best to make the most out of the immense assets the city already has.

6 Brooklyn Bridge Connection Highlight
The 6 Subway Train currently terminates near the base of the Brooklyn Bridge in Manhattan  (marked by the red box).The 6 Minibusway Route would extend across the Brooklyn Bridge, to Boerum Place, and to a connection with the A and G Routes at Hoyt-Schermerhorn. NYCAT features over 20 such connections at or near the end of Subway lines.

In Conclusion,

NYCAT would expand access to jobs. It would free up city and state resources to be used for other goals. It would help New York City run well 24 hours a day. It would allow the government to focus on maintaining excellent infrastructure while allowing private firms to compete for riders on their vehicles. It would decrease housing costs. It would expand access to jobs. It would address in some way nearly all of the major transportation issues facing the city today. It would make New York City a better place to do business. It would make New York City a better place to live. It would strengthen the the entire region.

A system like this would take a while to come to fruition, but we could do well to start thinking about it now because building this kind of system would take a concerted, coordinated effort and because thinking about it helps us understand the challenges in the current system. Companies are plugging away at making driverless technology ready for widespread adoption; it’s anybody’s guess as to when exactly that might happen, but the day will come sooner or later. When it does, New York City should be ready to transform its transportation system.

I’ll be posting weekly pieces that dive into different aspects of NYCAT and minibusways in general. The first will focus on how NYCAT would avert te need for Gateway or other new trans-Hudson tunnels by improving the capacity of existing Hudson River crossings, and the second will discuss the work that would be needed to get NYCAT up and running. Thanks for reading!

Related Posts:

The Minibus Way: How Autonomous Vehicles can Replace Trains

Eight Reasons Autonomous Minibuses Could Replace Subways and Urban Rail

Where Autonomous Vehicles Could Start Transporting a Lot of People

Chicago Autonomous Transit

Dynamic Tolls Promote Carpooling

One Last Thing: NYCAT, not to be confused with Nyan Cat

4 thoughts on “New York City Autonomous Transit

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s