Claim: BRT/Buses has better capacity than rail

[TheNZThrower]

According to this blog called The Rapid Transient, this table purports that BRTs and buses on dedicated lanes can achieve higher capacity than trains:

Now according to their source on frequency from New York YIMBY, which shows one line can handle that 30 train per minute frequency as of 2014, the E/F trains:

And if we look at the data sheet for the train used on the line from nycsubway.org, the R160, we see that it has a total capacity (seating + standing capacity) of 240 passengers for individual A cars and 246 for B cars:

This equates to a total of 1200-1230 people per 5-car train set (depending on whether it's the A or B model) or 960-984 people per 4-car train set. As it's most common for the trains to be coupled together, this leads to a capacity of 2400-2460 people per 10-car set and 1920-1968 for an 8-car set. Times both by 30 and we get the following results:
59

E/F Line max capacity per hour (10-car R160)	E/F Line max capacity per hour (8-car R160)
72,000-73,800 people	57,600-59,040 people

So the truth appears to be that the highest capacity NYC Subway line according to the aforementioned sources can handle 57,000-73,000ish people per hour at peak capacity, which means the low end estimate is larger than the Lincoln Tunnel Busway capacity figures cited.

In addition, the Lincon Tunnel Busway, or the XBL, is a single uninterrupted bus lane inside a tunnel which funnels out into a massive bus terminal called the Port Authority Bus Terminal. Said terminal is one of the largest in the world to boot, and far larger than a typical subway or BRT station. A BRT with stations will naturally have lower capacity than the XBL, as buses have to travel with a far enough headway to permit stopping at stations. Thus a better comparison would be with an existing BRT line, such as the Istanbul Metrobus. The Metrobus can achieve a peak capacity of 45,000 passengers per hour according to Mercedes Benz, one of the bus suppliers for Metrobus.

What this means is that your typical metro can easily match if not exceed the capacity of the highest capacity BRT AFAIK.

 

[Mendel]

30 train per minute frequency

 

[Rory]

Current operating average is something over 460 buses per hour - so 725 represents quite a step up.

https://www.panynj.gov/bridges-tunnels/en/lincoln-tunnel/xbl.html

 

[Rory]

According to this blog called The Rapid Transient, this table purports that BRTs and buses on dedicated lanes can achieve higher capacity than trains

Where does their 1,437 passengers per subway train figure come from?

 

[Mendel]

It was the 725 buses per hour that made my eyes pop out. Imagine the fumes!

that's one bus every 5 seconds, not a lot of traffic

imagine gas (not petrol) powered buses, or hybrids

Article:

BAE Systems announced today that is has been selected by New York City Transit Authority to supply 435 electric hybrid power and propulsion systems for its new fleet of transit buses, reducing carbon emissions and fuel consumption.

Recently, BAE Systems announced that buses with the company’s Series-ER (Electric Range) electric hybrid systems can see their engine switched on or off in predefined geofenced areas thanks to an automatic tool developed by Dutch company ViriCiti.

Source: https://www.sustainable-bus.com/hybrid-bus/hybrid-buses-in-new-york-bae-systems-selected-itll-provide-435-systems/

Article:

Clean Energy to provide renewable natural gas for New York MTA buses

By switching from fossil CNG to RNG, the MTA is expected to reduce its greenhouse gas emissions annually by more than 25,000 metric tons.

Source: https://www.masstransitmag.com/bus/vehicles/gas-diesel-cng-lng/press-release/21153050/clean-energy-clean-energy-to-provide-renewable-natural-gas-for-new-york-mta-buses

It's much easier to have a fleet of buses go green (hydrogen even) than wait for everyone to switch to better cars.

 

[Mendel]

It was the 725 buses per hour that made my eyes pop out. Imagine the fumes!

Current operating average is something over 460 buses per hour - so 725 represents quite a step up.

https://www.panynj.gov/bridges-tunnels/en/lincoln-tunnel/xbl.html

Your source:

The Exclusive Bus Lane (XBL) is a 2.5 mile contra-flow bus lane traveling along New Jersey Route 495, leading from the New Jersey Turnpike to the Lincoln Tunnel. It operates during the weekday morning peak period (from approximately 6:00 a.m. until 10:00 a.m.),

Today, the XBL averages over 1,850 daily buses,

Content from External Source

Where did you get your numbers from?
1850 buses / 4 hours ≈ 460 buses/hour

 

[Mendel]

Where does their 1,437 passengers per subway train figure come from?

the peak hour ridership table has 40846 as the biggest number

40846 passengers/hour on 30 trains ≈ 1362 passengers per train

if they have slightly different traffic volumes, it fits

 

[Mendel]

I think it's uncontroversial to say that a bus system can achieve a similar capacity as a subway system. But it's hard to design a bus terminal that can handle this kind of flow as easily as a railway.

 

[Rory]

that's one bus every 5 seconds, not a lot of traffic

Tell that to someone who lives on a street with 12 buses a minute!

(No need to research whether anyone actually lives on that route, just being light-hearted.)

Where did you get your numbers from?

Yep.

the peak hour ridership table has 40846 as the biggest number

40846 passengers/hour on 30 trains ≈ 1362 passengers per train

Isn't it supposed to be "maximum capacity"?

I think it's uncontroversial to say that a bus system can achieve a similar capacity as a subway system.

I agree, but that's not their claim. Their claim is:

"Buses can carry more people than the New York City Subway or the city’s highest-capacity commuter rail line, flipping the conventional wisdom on its head."

https://therapidtransient.wordpress...g-the-capacity-of-trains-buses-cars-and-more/

Content from External Source

 

[Rory]

Where does their 1,437 passengers per subway train figure come from?

Just noticed that it links to here:

https://www.transit.dot.gov/ntd/data-product/2019-annual-database-service-vehicle-inventory

But I don't see what is being referenced.

 

[Mendel]

Isn't it supposed to be "maximum capacity"?

Yes, where "maximum" is understood as "actual during peak hours". Note that they use "average occupancy" for automobiles as well, not maximum.

I agree, but that's not their claim. Their claim is:

... on their carefully selected turf, with their numbers, buses outperform rail by 3%.

For me, that comes out to "about equal capacity", which means I'd be looking at other factors that could influence a decision.

Tell that to someone who lives on a street with 12 buses a minute!

I've lived on a street that had 25000 vehicles per day, which includes trucks. If you have just the buses and no trucks or cars, that'd be an improvement in terms of fumes, especially with clean buses.

 

[Rory]

I've lived on a street that had 25000 vehicles per day, which includes trucks.

And did you think, "ah, how nice it is to live on a street with not a lot of traffic"?

There are about 25 vehicles a day where I live. A bit much but I suppose it could be worse.

Note that they use "average occupancy" for automobiles as well, not maximum.

Yes, good point.

So for cars they're using the "US average" (rather than the tunnel average or the NY/NJ average); for subway they're using the E/F train average (plus a bit); and for buses...?

I think the answer for the subway question might be it's the 40,846/30 = 1,362 passengers per train and then adding another 5% since the table says those trains are operating at 95% of "maximum peak guideline capacity". That brings it to 1,434, which is pretty close to the figure quoted.

For buses, though, if we look at the PANYNJ figures they state:

Today, the XBL averages over 1,850 daily buses, which translates to over 463,000 buses and over 18.5 million passengers a year.

Content from External Source

That's an average of around 40 passengers per bus.

Also, checking the capacity of buses, a Greyhound carries around 56 passengers and a standard 40-foot bus such as the Orion VII Original pictured on the PANYNJ webpage carries around 42.

So that figure of 69 is one I'd want to look at. It's said to come from this paper:

https://pdfs.semanticscholar.org/7e17/b88aab362f9f33236a61aacfda11a24233db.pdf

But that paper assumes 45-foot buses with 57 seats for its figures; 69 doesn't get a mention.

Looking further down the Off The Rails page I see that the 69 figure comes from an assumption of 57-seat buses with every seat filled and 12 people standing.

I suppose that's the answer: he's using average real figures for cars and subways against maxed-out (and then some) imaginary figures for buses that add 72% to the average real passenger number per vehicle.

 

[Ann K]

It's much easier to have a fleet of buses go green (hydrogen even) than wait for everyone to switch to better cars.

It's also much easier to reconfigure the bus routes to suit the current traffic demand, rather than dealing with fixed rail lines with limited stopping points. Flexibility is as important as volume, I think.

 

[Mendel]

It's also much easier to reconfigure the bus routes to suit the current traffic demand, rather than dealing with fixed rail lines with limited stopping points. Flexibility is as important as volume, I think.

"Reconfiguring" a rail line is about as easy as creating a new suburb....

Traffic patterns (where people come from and where they want to go) don't change much over time. If you build a new stadium or a new airport, you might as well plan the public transport that's going to serve it.

Rail-based transport is great for major arteries and inner cities. To reach the areas in-between and farther away, you typically configure a bus network to have stops within everyone's walking distance that is fairly low-volume per route, but that transports people to their closest rail access points. This network adapts to smaller changes well.

(It's also often possible to simply cycle to your rail station and be faster than the bus.)

 

[Mendel]

And did you think, "ah, how nice it is to live on a street with not a lot of traffic"?

no, but it goes to show that one big vehicle every 5 seconds is not much for a through road—which is the only kind of road you'd route thousands of buses along.

Living near a busy street car line wasn't as disruptive.

Worst was living under an airport approach.

 

[Landru]

The claim to be discussed is BRT/BUS has better capacity than rail. Not which is better.

​

 

[Mendel]

So, side question: do the bus capacities also equate to bus station capacity? do the Lincoln tunnel bus lane busses all stop at the Port Authority bus terminal?

it'd be misleading to claim "look at all this capacity" when you can't get that capacity off the transport

The Jackson Hts.-Roosevelt Av. metro station serves more than just the E and F lines

but I expect the Port Authority bus terminal does not serve every bus through the tunnel, or does it?


(Maps from https://new.mta.info/maps and https://commons.m.wikimedia.org/wiki/File:NYC_subway-4D.svg )

If the claim counts every bus through Lincoln Tunnel, but not every train through Jackson Heights station, then it undercounts the train capacity, I think? unless they want to argue a single "track", but then the Port Authority bus terminal must fail that majorly.

 

[Rory]

Going by the link in the OP the claim is really about:

"whether more, smaller vehicles offer more capacity than fewer, larger ones"

Content from External Source

with New Jersey-New York used as an example of how many people can travel using a dedicated bus lane vs how many can travel by train.

Spoiler: Some irony about the claim

There's a kind of irony in the claim really, for at its ideal his suggestion is:

• Use a dedicated direct bus lane with all other forms of traffic prohibited
• Automated braking and acceleration allowing them to travel in "platoons"
• Buses travelling as closely together as possible (up to five seconds apart)
• Double-deckers instead of single

One might suggest this could be improved even further by replacing the asphalt bus lane with something like metal rails and rather than having a wasteful five seconds between buses join them together and add extra doors.

His obvious conclusion should really be: double-decker trains.

 

[Mendel]

Current Sydney transport.

Subway (metro rail)

Article:

The Sydney Metro is a fully automated rapid transit system serving the city of Sydney, New South Wales, Australia. Currently consisting of one line that opened on 26 May 2019, it runs from Tallawong to Chatswood and consists of 13 stations and 36 km (22.4 mi) of twin tracks, mostly underground.

Source: https://en.m.wikipedia.org/wiki/Sydney_Metro

High frequency bus services

Article:

Metrobus (stylised as metrobus) is a high frequency, high capacity bus network in Sydney, Australia. Metrobus services run every 10 minutes during peak periods, 15 minutes during off-peak weekday periods, and 20 minutes on weekends, linking key commercial suburbs and centres throughout the city, with the intention of making timetables obsolete.

Source: https://en.m.wikipedia.org/wiki/Metrobus_(Sydney)

The lines used to have a separate numbering scheme, but have now been integrated into the regular network.

Discontinued: Monorail


@econ41 , do you have any information on passenger numbers / route capacities?

 

[econ41]

@econ41 , do you have any information on passenger numbers / route capacities?

Sorry, no. I've never studied it in any quantified approach.

 

[TheNZThrower]

I've found a Reason Public Policy Institute article which claims that a bus lane can handle up to 1,200 buses per hour:

Thomas Rubin, a California based consultant and former senior transit agency official in Los Angeles, says spacing of buses at 264 feet on 3-second headways at 60 mph is safe. That would yield 1200 buses/hour and passenger flows of up to 70,000/lane/hour. This is the huge capacity that could be achieved with busways. However, Rubin notes there are not many places where this huge capacity is needed. He also challenges the notion of heavy rail systems being capable of running at one-minute headways. The shortest headways in U.S. rail systems are achieved by the San Francisco Bay Area’s BART, at 2 minutes and 40 seconds. Boston, Philadelphia and San Francisco run trolleys at 60-second headways, but only at low speed.

The Transportation Research Board’s highway capacity manual recognizes a bus in a traffic stream as the equivalent of two cars so the capacity of an expressway lane commonly rated at 2,300 car equivalents/lane/hour will be able to handle 1,150 buses/lane/hour and at 40 to 120 passenger spaces per bus this puts busway capacity at 46,000 to 138,000 passengers/lane/hour—about six times the capacity of light rail and about twice that of heavy rail.

Of course, my main objection, assuming this is all true, is how are the buses going to stop? How large would the bus station have to be?

A headway of only several second is too small for passengers to disembark and board a bus at a stop. In addition, if you run 1,200 buses per hour, that would be equivalent to 150 8 car train sets per hour assuming one bus = one train carriage. You would need every station to be gargantuan to service that many buses, certainly a larger station than is typical for any conventional rail line.

The counter argument I have is that if there is a dedicated loading bay at the bus station next to the lane, the normal buses can stop there less frequently, leaving the lane to be operated by express services at a low enough headway so that the normal buses can merge back into the lane. But this still does not account for the stations where both express and normal buses stop.

The source for Rubin's claim is another Reason article with this graph:

The citation for this graph (cite note 37), is another Reason article. I'm a bit too tired to summarise it, so debunk at will.

 

[Mendel]

a hypothetical bus line has more capacity than actual rail, right

why not compare it to hypothetical rail? automate the train control to remove signaling and you can achieve much shorter headways.

you can also design rail to occupy unused space, which is why urban rail is often underground or elevated: then the preoccupation with how many "lanes" it takes can go away.

 

[Rory]

Hypothetical rail could run underground, ten tracks wide with kilometer long double decker trains traveling over a hundred miles an hour.

Your move, hypothetical buses with your city-sized terminals.

 

[Ann K]

Hypothetical rail could run underground, ten tracks wide with kilometer long double decker trains traveling over a hundred miles an hour.

Your move, hypothetical buses with your city-sized terminals.

Hypothetical rail lacks the flexibility of hypothetical buses, however. There's an unwarranted assumption that traffic will want to go from A to B for years to come, but when that changes, as it does with some frequency, the investment in rail is lost, while buses just have to print another schedule.

 

[Mendel]

Hypothetical rail lacks the flexibility of hypothetical buses, however.

The topic is capacity, not flexibility.

Nobody suggests replacing a whole bus network with rail.

I'd also like an example for a high-capacity rail line ("from A to B") that has been abandoned. The only rail lines I know to be abandoned were quite low-capacity to begin with, and many (e.g. streetcars in Edinburgh) are being brought back.

 

[TheNZThrower]

a hypothetical bus line has more capacity than actual rail, right

why not compare it to hypothetical rail? automate the train control to remove signaling and you can achieve much shorter headways.

you can also design rail to occupy unused space, which is why urban rail is often underground or elevated: then the preoccupation with how many "lanes" it takes can go away.

Now thinking about it, there does exist BRT systems with similar capacity. The Guangzhou BRT has headways of 1 bus every 10 seconds at peak:

It has the world’s highest BRT bus flows, with one bus every 10 seconds into the city in the morning rush hour.

Content from External Source

 

[TheNZThrower]

Though to be fair, the system itself occupies two lanes at each station, though it doesn't appear to be too different size wise from a typical train station:

Image source.

 

[Mendel]

Now thinking about it, there does exist BRT systems with similar capacity. The Guangzhou BRT has headways of 1 bus every 10 seconds at peak:

It has the world’s highest BRT bus flows, with one bus every 10 seconds into the city in the morning rush hour.

Content from External Source

that's 360 buses per hour
360 buses × 60 passengers per bus = 21,800 passengers per hour
that's less than the actual peak volume of many NYC metro lines listed in the table in the OP, and certainly less than what passes through any NYC multi-line station at rush hour

In addition, if you run 1,200 buses per hour, that would be equivalent to 150 8 car train sets per hour assuming one bus = one train carriage.

a single bus, as shown in the picture in the previous post, typically has about 35 seats, and standing room for as many people.

Compare Bombardier Twindexx Vario, a two-level rail car that seats 91 people:

 

[Mendel]

Though to be fair, the system itself occupies two lanes at each station, though it doesn't appear to be too different size wise from a typical train station:

I disagree. Compare:
Photo: Marc A. Hermann / MTA New York City Transit

 

[TheNZThrower]

I disagree. Compare:
Photo: Marc A. Hermann / MTA New York City Transit

But wouldn’t a surface level station be a more appropriate comparison?

Also, what are the issues with running an elevated or underground BRT network with either battery electric buses or trolleybuses?

 

[Mendel]

But wouldn’t a surface level station be a more appropriate comparison?

No, why?

Also, what are the issues with running an elevated or underground BRT network with either battery electric buses or trolleybuses?

they're not a very common solution

you can investigate the Downtown Seattle Transit Tunnel, though its bus passenger volume seems to have been around 28 to 37 thousand per day—what a busy NYC metro line can transport in an hour.

 

[TheNZThrower]

No, why?

Why not?