Bus Infrastructure New York

Combining Local and Express Bus Services in One Lane

» New York and Chicago debate putting BRT lines in street medians.

Last week, the New York City Department of Transportation announced that in the Bronx’s planned Webster Avenue bus rapid transit corridor, buses will run in lanes along the side of the street — not in the median lanes previously being evaluated. For this 5.3-mile route through the center of the borough, the decision will reduce bus travel speeds, increasing rider commute times and ultimately limiting the benefit of the BRT investment. The move evoked concern that the city was settling for less-than-best when it comes to bus transport in New York.

Yet the issue is more complicated than that, since many BRT lines share their routes with local buses.  This has implications for cities across the country that are investing in BRT.

Here’s the problem: In addition to BRT along Webster Avenue, New York plans to continue offering local bus service that would hail at all the existing, very frequent stops (whereas BRT would only stop every half-mile or so). If bus lanes were placed in the middle of the street, local buses would either be stuck in regular traffic lanes or have to weave in and out of the middle lanes to get to and from local stations. Or the city would have to put up the funds for local stops every block along the median bus corridor, which would be financially — and probably physically — prohibitive, in addition to likely holding up express buses while local buses are stopped.

Moreover, median lanes would quite inconveniently require automobile drivers to either enter the bus lane to turn left, or go around the block to make a turn.

So the city has settled for lanes along the side of the street that can be used both by BRT and local services. In this design, Webster Avenue joins the other BRT corridors already up and running in New York and almost every other city in the country that’s building BRT. Those side lanes will almost definitely get obstructed by car traffic, which will have to move through the lanes to make right-hand turns. In addition, the insistence of community groups and local businesses that parking be preserved along the sidewalk means that parallel parking, positioned to the right of the bus lane, will be a constant source of interference into the BRT traffic.

The Webster Avenue corridor, currently served by the Bx41 from Gun Hill to the Hub, is more than just a local route; it is currently being considered for express service to and from La Guardia Airport, providing the Bronx its first (potentially) quick, direct connection to a major jobs center.

Chicago, which has its own major BRT plans in development, is currently studying whether to place its BRT routes for its Western and Ashland Corridors in the median or along the side of the street. Four alternatives are up for consideration, two of which would include a median routing (and buses with left-side doors), and the other two which would run on the side of the street, sharing the lane with local bus services.

That city’s initial analysis demonstrates the advantages of the median alignment: Faster service for customers (up to 80% quicker than current operations), and a resulting larger increase in ridership. A comparison of options by Grid Chicago provides an overview of those benefits, in the context of their relatively higher cost. If the goal of the transit system were to simply to increase the ridership of the express transit services — ignoring all other issues — the median would appear to be the right place to go.

But it does seem rather incongruous to promote a median-running BRT lane when local services remain relegated to the side of the street, especially when some passengers are likely to choose on the spot between the two options when taking their trips. Does it really make sense to use the bus-only lane half as much as there are buses available to fill it? Shouldn’t dedicated bus lanes be shared by both local and express buses, with the only difference in service being the number of stops at which each calls?

Moreover, the side lanes BRT alternatives being proposed in both New York and Chicago will have considerable advantages for customers, as shown in the rendering below. On Webster Avenue, bus-only lanes will run along 4 miles of the route, and car travel lanes are expected to be eliminated, reducing automobile traffic to one lane per direction. Pedestrian crossings and walking areas are expected to be significantly upgraded. Transit vehicles will be provided signal priority at intersections. Bus stations will be positioned adjacent to sidewalk extensions, making it unnecessary for buses to turn when making a stop.

The New York design does suffer several failings, however, that could be relatively easily resolved. As shown in the plan at the top of this article, the current proposal for Webster Avenue is to stop local buses as is currently done, along the sidewalk, replacing current parking. This makes sense for local-only stops, since a local bus pulled up to the sidewalk would allow an express bus to pass it without moving out of the bus-only lane.

But for stations shared between BRT and local buses, having two separate stops — one along the sidewalk (for the local) and the other next to a bulb-out (for the BRT) — makes little sense. In off-peak periods, when bus headways are 10 minutes or more, smart customers may simply want to take the first bus that shows up, express or local. They shouldn’t have to move to a different place on the sidewalk beforehand, so at shared stops, BRT and local vehicles (neither of which is supposed to be passing the other there) should both open their doors at the sidewalk bulb-out.

Of course, it bears repeating that as the Chicago study demonstrates, placing buses along the edge of the roadway significantly increases travel times compared to median-running buses. Even if buses in the side lanes make the most sense overall, they pose a challenge to actually speeding up riders, which is what BRT is supposed to do.

The advantages of running buses along the outside lanes of the roadway address the problems of places where there are both BRT and local services offered. But along certain routes, one might argue that differentiating express and local services is inappropriate and that the most important goal is to provide the highest quality operations on all routes, not just those designated “BRT.” In Paris, many of the Mobilien “BRT” corridors have relatively frequent stations, but buses are often placed in the median lanes, with stations in the center of the roadway. There are no local or express services, just good-quality buses that hail at many stops but at least don’t get caught up by traffic in between them.

Footnote: My thoughts go out to the people of New York City, who are surely more interested in recovering from the storm than debating how to build BRT. But perhaps this will be a welcome distraction. Apologies for the extended period away from new writing at The Transport Politic.

Images above: At top, corridor plans for Webster Avenue, from New York City DOT; at bottom, a rendering of the corridor, from NYC DOT.

34 replies on “Combining Local and Express Bus Services in One Lane”

In San Francisco, they plan to solve this issue on Geary by placing local stops in the median, with both locals and BRT running in center lanes and enough room for BRT buses to pass stopped locals. On Van Ness (also center BRT) the solution is to use one stop pattern a stop spacing that’s wider than the existing local stops but not as wide as full BRT.

Both solutions are better than BRT curb lanes, and both are better than having BRT in the center and locals on the curb. Neither require left turning vehicles to enter the BRT lane due to use of transit priority signals.

We could swap the parking with the bus lanes to reduce interference by motorists, making them “protected” like the 8th & 9th Ave bike lanes, but that would eliminate the ability of an express to pass a local… I agree that side-running is better for overall urbanism and customer experience though, and reducing the travel time is a problem that can be solved through good design (if only this would actually happen).

As I commented on my input sheet during the public meetings last week in Chicago, there are opportunities to switch between Median and Curb lanes at points along the 16 mile corridor.

There are indeed areas where eliminating left hand turns will have major negative impacts, and using curb side lanes would be ideal. But on much of the corridor queuing drivers for right hand turns, and parallel parking will make curbside lanes slower even than the average speeds found in the study. Most important is having controlled prepaid boarding, and along much of the route this can only be accommodated with a median platform without dominating sidewalks at corners.

One size does not fit all, and while it doesn’t make sense to switch alignment often, it could be done in conjunction with existing dedicated turn signals and other eccentric right of way layouts which are especially common on Western Avenue.

I find this naively optimistic that any speed gain will be had by signalization changes when you’re still stuck behind right-turning cars. The cars will still be waiting to turn right on green – maybe because of a parallel crossing pedestrian; maybe because the side street lacks room to queue; maybe because traffic is backed up from the next intersection(s).

Signalization changes without reserved lanes are, in my judgement, unlikely to help much if at all. Pretty much anywhere where traffic is bad enough to want these changes in the first place, traffic will also be bad enough that this will hold – you won’t get any benefit unless the cars are completely out of the way.

This configuration is nearly identical to the Silver Line Washington St branch in Boston. The biggest issue is double parked cars in the bus lanes, many whom are stopping for “just a minute” to run into an adjacent store. The second biggest issue is that the City only allows the buses to have signal priority if they are running “behind schedule”. This is very short-sighted, IMO, but goes to show that just because the Mayor has declared “the car is no longer king” doesn’t mean that those running the transportation department have fully embraced this concept. Unlike the NYC example, there are no local buses that parallel the route, though, so the only stops are the major ones used by the Silver Line buses.

The idea the Menino is in any way pro-transit is pretty laughable—he might make occasionally mention it to keep parts of the Democratic base mollified, his record reveals that his actual position on transit ranges from hostile to not giving a care (you’re right about MDOT and the MBTA too, of course).

Double parked cars doesn’t have to be a problem. Cameras on buses and fines of a hundred dollars will soon put a stop to that. Does the trick in London.

Soooo, if you’re putting the lanes in the median, why not speed things up further by putting in electric power and rails — restoring the streetcars which were ripped out back in the 50s?

All of this futzing about with BRT is just reminding me that rail is better for most purposes.

1) Because rails and electric power are much more expensive than curbs and paint.
2) Because buses are not limited to the BRT route. Once the route ends, or in the middle of the route, they can branch off and go to other destinations.

Well, rails are certainly more expensive than paint, but have you actually priced curbs lately? Didn’t think so. How about asphalt?
Electric power is of course cheaper than gasoline or diesel or natural gas.

Oh, and trains aren’t limited to the LRT route. Once the route ends, or in the middle of the route, they can branch off and go to other destinations on any rails which may be available. That is merely a “we happen to have some asphalt here already” argument, and I find it less and less compelling as street maintenance budgets drop.

I’m all for bus lanes made with paint. Paint is cheap. It’s when the capital investment starts going beyond paint that it starts seeming like a poor return on investment.

There’s also the fact that buses are less visibly intrusive than LRT – most LRT implementations require overhead wiring, and in a lot of places it’s viewed as unsightly (I think it was banned in Manhattan while trams were still running, and the prohibition may never have been repealed.)

A better future: European cities have been working on getting tram routes thru their historic downtowns without ruining the classic street scenes. The manufacturers are making remarkable progress toward meeting this demand.

Various technologies. Batteries, of course, but they are heavy and bulky. Or putting the juice just below the pavement level.

Most recently, new design trams will use supercapacitors (I looked up that term on Wikipedia, found “electric double-layer capacitor” and I still can’t explain it ha ha ha) that can store energy and can charge very rapidly. These units can be charged from below and/or overhead while a tram is stopped.

One part of the Wiki article that I did understand, “A 3-kF capacitor that cost $5,000 in 2000 cost $50 in 2011.” This power delivery system could end up being MUCH cheaper than catenaries hung from wires strung from poles.

Production model trams with ‘overhead wireless’ capability are still limited in range, a few blocks or so. Maybe in a few years we can put them on 42nd and 34th running river to river, and down Fifth Avenue, too.

Until then, I agree with Henry — and with the NIMBYs — that the wires, poles, etc. are just too ugly for most city streets. Wireless trams could change that, the way wireless changed phones.

Trams / steel-wheel-base LRTs offer higher capacity than any BRT system is able to. The longest one-unit-trams worldwide operate in Budapest (54m) just fitting between two Manhattan streets.
Even longer multiple units can be found in the German “Stadtbahn” systems (max. train length 80m due to road traffic rules).

Street lanes:
Street lanes reserved for public transport (only public transport bus lines and tram/LRT lines) increase the average speed of all (allowed) vehicles using these; this can be improved by small curbs separating public and individual modes of transport.

Trams / LRTs are in most cases when sharing streets with other vehicles electrically powered using catenary; especially French tram operators have designed good-looking and very slim catenary.
Alternatively, ground-level power supply could be installed like in once used to be e.g. in the Washington D.C. tram network.

Trams’ electromobility has actually already been developped and optimised for more than 100 years to minimise energy consumption and noise emission through electrification. Additionally, electrically powered systems are independent from gas and petrol prices; electricity rates are a matter of engergy politics.
Trolley-buses would only be of advantage along really hilly bus lines (i.e. not in NYC) or where the ridership is too low for trams.

Fleet investments:
One bus life is approx. 10 years for the requirement to offer comfortable trips; one tram life is approx. 30 to 40 years. Further, one tram can replace two or three buses. This is why much less tramcars compared to buses need to be ordered (see above).

Infrastructure investments:
Trams / LRTs need additional rail infrastructure, of course; these are much higher than for common BRT systems. This pays off in a longer period of time, which can also be an advantage not to reduce traffic as the fix costs make a much higher share of all costs for trams than for buses.

Car drivers’ view:
Car drivers usually prefer buses rather than trams, and by experience they respect rail-guided trams much more than buses.

My Conclusion for Manhattan(NYC):
1) Given the high riderships in NYC and to relieve the overcrowded subway lines, trams/LRTs should be preferred instead of BRT systems; in the long-run, they will get cheaper than buses especially in Manhattan.
2) As far as possible, public transport street lanes should allow both tram/LRT and bus systems to use them, usually with additional priority at intersections.
3) In case of choosing BRT, all BRT projects should at least prepare for a later use by trams/LRTs, too.

Note: The term “tram” refers to “streetcars”, as internationally standardised, even although North Americans enjoy to develop and mixed terms continuously; the so-called “Roosevelt tram” is in fact a cable car.

Didn’t New York City also have underground power supply for streetcars? Let’s just do cable cars!

The problem with LRT cars going elsewhere on the “network” unlike a bus, is there would no other network. Better to establish BRT and then gradually convert them to LRT, rather than do the German bus subways.

Just what it sounds like. I’m too lazy to find pictures today (really buses running through a streetcar subway).

Bus tunnels have always been a niche solution for a very limited market, usually as an intermediate step (Seattle/US, Guadalajara/Mexico) or rather a test application (Essen/Germany). The fix costs of Boston(US) Silver Line tunnel sections reach dimensions of an underground LRT. So all such systems deal/dealt with too high costs for only limited benefits; also the operational costs are much higher than for trams/LRTs which can replace two to three buses each.

Cable cars within streets have only run with a very limited speed (SF/US); faster cable-driven systems like in Italy (e.g. Perugia, Venezia) need separate right-of-ways and are comparable to monorails.

Even the most efficient BRTs worldwide (e.g. Guangzhou/China, Curitiba/Brazil) can be beaten by high-capacity trams/LRTs.

That’s why trams/LRTs are being reopened in many cities around the world *because* of their long-term efficiency for demands between bus and metro systems, their managable infrastructure and operational costs and because they are easy to use/access.

Nathanael is correct’ BRT is mostly a scheme to gild the bus lily at very little advantage to riders. Separating the stops for locals and expresses just makes travel less convenient.

Chicago’s options are again constrained by Mayor Daley’s dreaded parking meter lease. Taking away street parking for any significant segment of Ashland or Western is going to add many millions to the project as LAZ parking must be paid for all lost meter revenue for the next 90 years or so. I suspect this is why they are so seriously considering taking away a lane of traffic in each direction – which is a huge change in Chicago. Western and Ashland are major thoroughfares, and Western even has a segment with an expressway-like overpass.

I wonder if the meter count clause goes away with “street reconstruction”? The city should have marked each space like they do in some suburbs, rather than claiming “oh, the meter boxes will allow more cars now, rather than just the number of meters”.

Not really. Much of the parking along both corridors is unmetered. Ashland through the North Side is more heavily metered, which is why I’m guessing they will go for a median design and narrow to one lane of traffic to preserve parking. Western will probably have a median design with parking eliminated (preserving it as a major thoroughfare).

A few commenters here and on other sites have noted that the rail may be a preferable solution on these corridors. I have an interesting argument against doing this initially in Chicago.

The CTA is building the Ashland and Western Ave. corridors as a showcase of improved bus service in the city. They intend to make it feasable to install these improvements on other routes where they are needed. Installing rail would maintain a distinction between the classes of service that would not translate to other routes and locations in the public’s mind.

Ivisited ny old neighborhood (South Shore) to check out the 14 Jeffery Jump
So they millions are being spent on 2 “demonstration” stations 1 @ 87th 1 @ 71st which will feature shelers/waiting benches, and arrival predictors. These features are in place for instance on the Addison Crosstown @ Southport. The Jeffery will also get “special signage” , until the day the special buses are OOS and whatever is roadworthy shows up. Exactly ONE queue jump is going in @ Anthony Ave, rush hour lane priority (will believe it when I see it enforced)and the mythical ttraffic signal priority (if it is non functional as AC Transit and Muni’s installations…) The real issue, it seems to me is will the bus get any preferential lane on the Outer Drive (60+% of the route distance)? So much for BRT.

I live in South Shore, right by Jackson Park and can say the J14 has improved service. The 1/2 mile stop spacing and rush hour bus lanes (north a/m, south p/m) really have created a more consistent and reliable service. The bus lanes are heavily enforced, with seemingly several cops especially assigned. On Lake Shore there is no lane priority, but that has always been the point where buses gain lost time because there is usually only heavy traffic near Soldier Field during rush hour.

I hope it works. So far, it looks like there are no bus lanes getting installed, and probably no signal priority, but lots of money is being spent on gimcrack.

In addition to AC Transit and Muni, Toronto’s another case where it seems to be impossible to get the local streets department to actually turn the signal priority on. If you actually get it turned on in Chicago, that *would* be lovely.

To be optimistic, unlike the situation in Toronto, where permanent staff in the streets department are deliberately not doing what they’re supposed to, and getting away with it — in Chicago if the mayor and the local alderman WANT it done, I would expect that they could *force* it to get done. But will they?

Is signal priority kit expensive? Because if it isn’t, it should be rolled out across the Limited bus network.
All routes with frequencies better than 15 minutes should also get Limited routes – it doesn’t cost money to make stops farther apart.

These two improvements would make journey times shorter for a lot of people, and SBS is just those improvements + POP and bus lanes. Sure, LRT and real BRT are an order of magnitude better than SBS, but hey, an improvement’s an improvement, and the more commutes we can improve, the better it is for everyone.

I am all for signal priority. Unfortunately both SF Muni and AC Transit “claim” to have installed the hardware, BUT it isn’t working–SF Muni never activated it, and AC Transit admits that they don’t know if the units are functional.
In the case of the AC Transit route closest to me, the hardware was only deployed on the “Rapid” (BRT very lite) even though locals operate many more service hours. When fleet maintenance crashed the Rapid was being operated w/ non equipped buses.

What the traffic capacity is, in terms of persons per hour passing a given point of a traffic lane, is greatly increased by running trams/streetcars/LRVs (they are the same – just different nomenclature) instead of buses. A reasonable peak capacity for an articulated bus is about 80 people – more if you minimize seating and go for ‘crush’ loading. If you are lucky at a bus stop you can average a bus every 30 seconds, taking into account the bus braking to enter, dwell time and acceleration. This gives 120 buses per hour. Sydney tries to do more than this and has severe bus congestion on the streets where it is tried. Passenger capacity with buses can be as much as 9600 persons per hour. A tram every 2 minutes can carry that number of people per hour – they do it in Budapest. So a tram replaces not 2 or 3 but 4 buses, with substantial operating economies. Doors on buses tend to be limited because of the wheel locations and this increases dwell times – on a long tram, eg a Skoda 15T, there are two wide doors to every section, so dwell times are minimized.

With a tram every two minutes, traffic priority tends to be self enforcing – with buses every 30 seconds it is impossible. At greater headways, say a tram every 6 minutes, traffic priority for the trams is very easy. For the buses at every minute and a half, it is still dicey.

And note that a tram every 6 minutes can still carry 3200 persons per hour, cars on a traffic lane in CBD conditions are limited to carrying about 1200, 1500 persons per hour at the very best. So it is worthwhile in maximizing the traffic capacity of a street to replace one lane of cars by one lane of buses, better still one lane of trams.

Leave a Reply