A New Blueprint for Streets That Put Transit Front and Center

This template shows how transit could be prioritized on a wide suburban-style arterial. Image: NACTO
A template for transit-only lanes and floating bus stops on a wide street with parking-protected bike lanes. Image: NACTO

The National Association of City Transportation Officials has released a new design guide to help cities prioritize transit on their streets.

How can cities integrate bus rapid transit with protected bike lanes? How can bus stops be improved and the boarding process sped up? How should traffic signals be optimized to prioritize buses? The Transit Street Design Guide goes into greater detail on these questions than NACTO’s Urban Street Design Guide, released in 2013.

Before the publication of this guide, city transportation officials looking to make streets work better for transit still had to hunt through a few different manuals, said NACTO’s Matthew Roe.

“The kinds of problems that the guide seeks to solve are exactly the kinds of design problems and questions that cities are trying to solve,” said Roe. “How do you get transit to get where it’s going quicker, without degrading the pedestrian environment? Some of that has to do with the details of design.”

For example, the guide lays out how to design boarding areas where buses can pick up passengers without pulling over to the curb and then waiting to reenter traffic — which can slow service considerably.

This diagram shows how a "boarding island" can improve transit service, as well as prioritize comfort for riders. Image: NACTO
A boarding island can improve transit service, as well as prioritize comfort for riders. Image: NACTO

Many of the finer points in the guide involve making transit-priority streets that also work well for walking and biking. The image below, for instance, shows the intersection of a sidewalk-grade two-way bike lane with the pedestrian approach to a bus stop.

Image: NACTO
Photo: NACTO

“Transit has to run in the place where everyone wants to be,” said Roe. “It has to be right through where everybody wants to be. To create transit streets that are living streets and active streets is something that people really struggle with.”

He said Loop Link in Chicago “truly exemplifies” that kind of street design. “Instead of pushing bikes and transit off to the side you make them the centerpiece.”

Chicago's Link Loop. Photo: Nate Roseberry
Chicago’s Link Loop. Photo: Nate Roseberry

Following the release of the guide, NACTO will be working directly with three selected cities — Denver, Indianapolis, and Oakland — on a “Transit Accelerator Program,” customized assistance to bring better transit streets to fruition.

The full Transit Street Design Guide can be purchased from Island Press.

37 thoughts on A New Blueprint for Streets That Put Transit Front and Center

  1. With all of the manufacturing defects, wrong components (one assumes usually cheaper, shoddier), issues from the vendors, where has APTA been? One need only look back to the development of the PCC streetcar (some of which are soldiering on 60-70 years after original delivery) which was a project of a trade group in the 1930s top see what APTA should be about.

  2. I still see a 6 lane plus swath of concrete. Still handing over a massive amount of real estate to moving vehicles.

  3. The problem with this configuration will be standing and double parking in the bus lane. What if you swap the bus and parking lanes?

  4. What about at bus stops? Have the parking lane end and have the bus lane swerve into it, with a curb to keep cars out, and a bus stop aligned with the bus lane’s original alignment adjacent to the bike lane? Narrow the sidewalk to swerve the bike lane right and make room for a transit stop without swerving the bus away from the curb? Either way it could only work if you don’t ever expect buses to pass each other.

  5. “Moving vehicles”…I assume you mean “motor vehicles”, vehicles which are parked are hardly moving. That being said, the parking lane is much more than just parking, its multi-function, bus stops, probably bike parking areas, bike share stations, and…with the proper pricing, some vehicle parking. Nothing wrong with that. As for the transit lanes, they move people far more efficiently. Given that current status quo, this is actually a huge improvement.

  6. As a cyclist, I ride this config in Chicago daily | It’s a disaster for cars, cyclists AND pedestrians. Pedestrians cross over the bike lanes without warning to catch buses or jump walk or don’t signs at crosswalks, while cars lose badly needed lanes in heavily congested downtown locations. This is NOT a solution to having all 3 modes on the same grid, it simply reconfigures the same problem.

  7. Oddly cabs and buses in Chicago play by traffic rules in Chicago. Pedestrian and Uberists… NOT

  8. Another Chicago cyclist told me he accidentally got trapped inside the mass trans lane and felt he was in no mans land. Like a target in GTA, the bus driver gunned it!

  9. The modes will always be on the same grid. You can unravel routes and offer alternatives but people will always want to ride bikes on main roads as well.

  10. We need more protected intersections then. And protected bike lanes. And either protected bike lanes or quite wide (2-2.5 metre with a 50 cm buffer from the moving traffic and preferably an 80 cm buffer from parked cars) bike lanes of the standard type, and the 30 km/h zones to link them together. And practically no more stop signs affecting cyclists and traffic light optimization.

  11. A better configuration is curbside bus lane, then bike lane, then either a barrier or parking lane. The buses run along the curb all the time, which is where they pick up passengers. Cyclists don’t have to deal with bus passengers crossing the bike lane. And cyclists still have ready access to stores and residences since all they need do is cross over the (mostly unoccupied) bus lane. Every time I see the configuration shown in the graphic I can’t help but think it creates unnecessary conflicts.

  12. The problem isn’t all three modes on the same grid. Rather, it’s when you have large numbers of all three modes on the same grid. At that point pretty much any solution sucks. It’s just a matter of whether it sucks more for cyclists/pedestrians, or for cars. In my opinion where you have large numbers of pedestrians, with the potential for medium to large numbers of cyclists, it’s best to take all nonessential motor vehicles out of the picture. That lets you optimize for non-motor powered modes.

  13. Do what I mentioned above. Curbside bus lane, then bike lane (with buffer), and finally either a parking lane or barrier. Buses can pass each other by going into the bike lane/buffer area. Granted, it’s not ideal, but it’s much better than having cyclists conflict with bus passengers at every single bus stop. Since my configuration also puts the bike lane about 11 or 12 feet away the sidewalk you neatly avoid the problem of pedestrians spilling over into the bike lane.

    The design shown in the graphic may work OK if there aren’t large numbers of bus stops or bus passengers but in a place like NYC it’ll be a disaster. During rush hours you’ll have a steady stream of bus passengers leaving or entering at some bus stops, plus fairly heavy bike use. Either lots of people will miss their buses waiting for cyclists, or lots of cyclists will be heavily delayed waiting for bus passengers. I can’t imagine the conflicts will endear people to cyclists, either.

  14. The sidewalk is a much less pleasant place to be when passing buses or trucks are immediately adjacent to it. Parked cars or a bike lane are an effective buffer to reduce the impact. Sidewalks are often congested, transit stops full of people waiting or getting off can be a real pain to walk around. It’s not just cyclists that can be delayed by a stopped transit vehicle. Curb extension bus stops help reduce that impact. Bus stops are also where you would conceivably want room for buses to pass each other. So say you have a curb extension – or continuously extended sidewalk, adding say 8 feet, then 20-22 feet for the bus+bike lane+bus/bike buffer to allow buses to pass each other, then a parking lane +door zone buffer (11 feet) or bike/car buffer (3 feet), then a general traffic lane (10 feet). So that’s 41-51 feet per direction in addition to the original sidewalk width. You don’t always have that much street space. And even then, it only works if bike traffic during peak hours is low enough that buses can pass each other. Or bus traffic is low enough that they won’t have to. An alternative that might work in some places is to have bare bones or no bike or auto facility on one street if necessary to accommodate two transit lanes and wider sidewalks/transit stops, to concentrate transit vehicles onto that street. Then ‘premium’ bike facilities on nearby streets made possible by moving transit out of the way.

    There are always tradeoffs. There is no perfect design for everyone. Even if you ignore cars and trucks. The ‘best’ design will vary from street to street, and still hurt many. The degree of sacrifices that must be made increases as density increases. Decrease auto space and you can reduce the sacrifices other users have to make, but you can’t eliminate them in a place like NYC. This is even true if your bikeaducts are built! (ramps/landings still take up a non zero amount of space, $$ is limited)

    In the moody ave (portland) picture some unmentioned tradeoffs….the streetcar is in a general traffic lane, not a dedicated transit lane. This is right by the new car free bridge portland opened last year. You can’t see it, but there’s a viaduct beginning to rise just to the right. It carries bus+light rail (and theoretically emergency vehicles if necessary), but no bikes or pedestrians. Streetcars can use the same rails, but they are more of a local service and the (short) viaduct would have eliminated a stop in a dense area. Behind the picture there is a huge highway bridge, the transit viaduct threads between the pillars. Supposedly there wasn’t room for a sidewalk/bike lane/shared path. They could have threaded that through a different set of pillars I’d guess, but $$$. With the existing connections cyclists continuing on Moody between downtown and the south waterfront are served fine by the cycle track where it is. But it adds a little delay for those heading from the new bridge to south waterfront, who would benefit if it was on the other side of moody. There’s a wide sidewalk there that some cyclists use to avoid the delay crossing moody twice to use the bikeway when they head south for a quarter mile or so, but a couple buses, the light rail, and one of the streetcar lines unloads by the bridge landing, so the sidewalk fills up with people walking to the south waterfront tram to get to OHSU, a hospital and huge employer. The city has signs directing cyclists across the street, but there’s always a few who will ignore it. Most of them will at least ride respectfully though. Cyclists do a kind of bad job at yielding to pedestrians leaving/going to the streetcar stop, but it’s helped that many getting off the streetcar cross moody right when they get off. Either to transfer to another service that goes over the new bridge (one streetcar line does, one heads south), head to an OHSU building right across the street, or if they’re heading to the main campus they have to cross the street anyway to get to the tram, so why not do it where there’s a crosswalk with no signal delay? So many never cross the bike lanes anyway.

    Here’s the general area. Look at streetview going back to 2007. Everything is brand new. The entire road was completely rebuilt ~10 feet higher or something as a flood control measure (I think?). Maybe you can nitpick about some of the design choices – there are plenty who do. But whatever they had done instead someone would have lost out, even when not all space is used for autos.


    Wide sidewalks with room for cafes, short pedestrian crossings, two way bike lanes on both sides of every street, transit lanes on every street with room for transit vehicles to pass each other or stopped ADA vans, freight loading zones etc…there are plenty of desirable things to do with a street if you ignore private autos. But they are in conflict. A fundamental fact of a city is that space is limited. Making more efficient use of that space is great, but it can only do so much.

  15. The problem is we have to prioritize non-auto use of space where space is scarce. Time and again we end up with suboptimal projects because we insist on keeping the parking lane. Take Jay Street for example. We could probably implement it exactly as I described if we lost the parking lanes. Going forward, I strongly feel the idea of curbside parking needs to die, at least in the denser parts of urban areas. Maybe you could make a case for it on some residential side streets but even there we should charge for it. Everywhere else it needs to go. If we want to reduce the number of private autos, no better way to start than making it hard or impossible to park in some areas. That will discourage auto use in those places. As you said, once we ignore private autos, you can do plenty of more desirable things with streets.

    Decrease auto space and you can reduce the sacrifices other users have to make, but you can’t eliminate them in a place like NYC. This is even true if your bikeaducts are built! (ramps/landings still take up a non zero amount of space, $$ is limited).

    Right, but with switchbacks we can squeeze those ramps into a space of maybe 80 feet by 10 feet, or roughly four parking spaces. Moreover, remember the concept behind these “bikeaducts” has always been as express alternatives to local streets. You could probably get by with one ramp every 1/2 mile or so, at most every 1/4 mile. That’s a really miniscule amount of street space. Cost is really the much bigger problem for something like this, although that can be offset if utilities were to run in them. You could also attach street or sidewalk lighting to them. If built above sidewalks they could shelter the space (on balance in NYC that’s probably good given the number of rainy/snowy/hot days versus the days where such shelter might be a negative). The bikeaducts themselves could even be roofed over to allow riding 365 days per year. I’ve also thought of ways to channel prevailing winds into a tailwind. If we could manage a gentle 5 mph tailwind most of the time by doing this, you average cyclist could cruise at maybe 17 mph instead of 13 mph. That’s over a 20% decrease in travel time with no extra physical effort.

    There are always tradeoffs, but when you start thinking in three dimensions they become far less. As cities get ever more dense, we’ll probably need to go that route anyway. As things stand now, trying to keep everything on one level without restricting private autos costs large cities dearly.

  16. 40+ buses per hour per direction at peak with no buffer will make the sidewalk miserable. During peak there are ~500 bikes per hour, not sure you can expect that buses will reliably be able to pass each other in your proposed bus/bike/buffer shared zone unless you want them playing chicken with cyclists. Hey, some cities have yield to bus laws, but they make more sense with cars since the cost of a crash is generally less fatal. Choosing to minimize the sacrifice to cyclists to and push more onto pedestrians, businesses (worsened sidewalk environment) and transit riders is a legitimate position to take, and the correct one in some places. I’m not sure I’d agree with it on Jay street though.

    Getting rid of parking doesn’t help as much as you think. You already lose parking lanes at bus stops where you still don’t have enough space.

  17. On Jay Street I personally feel a center running bike lane makes a lot more sense than either what is proposed, or my alternative. That said, with 40 buses per hour during peak times there WILL be major conflicts between bus passengers and cyclists with the proposed configuration. With my configuration there will still be conflicts, but between buses passing other buses and bikes. At least that doesn’t raise the ire of pedestrians. The bike lane near the curb idea works passably well if bus stops are widely spaced and bus service is infrequent. When the stops are more closely spaced, my idea works better, provided bus traffic is still light enough so buses seldom need to pass each other. When you get into heavy bus traffic/frequent stops, I’d say neither proposal works great. In that case it’s better to either reroute bikes to another street, or if that’s not really feasible use a center running bike lane (again not an ideal solution but at least you avoid conflicts with buses or bus passengers). Or if space is really limited use a bike viaduct. That might actually work well on Jay Street since it’s mostly a through route for bikes.

    Sidewalks already have somewhat of a buffer even without an adjacent parking lane or bike lane. The 5 feet near the curb is usually a dead zone with trees, newspaper vending machines, trash cans, sometimes Citibike racks. It’s not like the buses will going 60 mph on Jay Street. They’ll probably be crawling along most times of day.

    Getting rid of parking doesn’t help as much as you think. You already lose parking lanes at bus stops where you still don’t have enough space.

    It helps primarily to discourage private auto use. That in turn can gain you several multiples of the space you gain by losing the parking. In some parts of Manhattan for example fully half the traffic is people looking for parking. Eliminate parking and the traffic drops by 50%. That could be enough to free up an entire traffic lane or more.

  18. That said, with 40 buses per hour during peak times there WILL be major conflicts between bus passengers and cyclists with the proposed configuration

    Why? The buses are pulling to the curb.

    In that case it’s better to either reroute bikes to another street

    Not sure which street that could be here.

    center running bike lane (again not an ideal solution but at least you avoid conflicts with buses or bus passengers)

    And introduce conflicts between cyclists trying to access local destinations by riding on the sidewalk and pedestrians.

    The 5 feet near the curb is usually a dead zone with trees, newspaper vending machines, trash cans, sometimes Citibike racks.

    Jay street doesn’t have a lot of that sort of sidewalk obstruction. It would, except the space is needed for pedestrians.

  19. In the case of Jay Street, there really aren’t that many local destinations. As was already mentioned by another poster in the Jay Street thread, only one block has any commercial establishments, and most bike traffic is through traffic to the bridge. Sure, with a center running lane( or a viaduct) you inconvenience a minority of cyclists whose destination might be somewhere on Jay Street, but as you said every design is a compromise.

  20. But those buses will go straight on, while cyclists can enter and exit at every single intersection – all potential conflict, of the deadly kind. In the NACTO scenario cars have that same problem, but it’s still less dangerous for cars to cross a bus’s path than it is for pedestrians to cross a car’s path, so I understand why NACTO chose this order instead of the Amsterdam configuration with the PT in the middle, then cars, then bikes, then sidewalk.
    Speaking of Amsterdam, pedestrians manage just fine crossing the bike paths, unless they’re tourists who don’t know what to look out for. So based on that, I think people in the USA just need to get used to these new street layouts.

  21. That first image at the top of the article is a step in the right direction, but still inadequate, as the Netherlands has learned over the years. Pedestrians “manage” better in Amsterdam because bike paths are (1) at a slightly different level than sidewalks (you notice a step down from the sidewalk into a bike path, similar to stepping off a curb onto a street here) (2) universally separated from vehicular travel lanes (auto/truck or transit) by a landscaped or paved median (which sometimes serves as a bus/tram stop waiting area or bike parking area), (3) paved in a distinct way (texture matters, and what they use in Amsterdam happens to be optimal for biking vs. driving or walking), and (4) often colored differently than sidewalks and vehicular traffic lanes in mixing zones (near crosswalks and bus/tram stops). Bike path treatments in Dutch cities are visually, spatially, AND tangibly different than pavements for other modes. It takes more than paint.

    I know, NACTO references the “premium” treatments in use in places like Amsterdam, but I believe a published “standard” should specify the safest treatment, and allow for less costly alternatives as exceptions. I’m not an absolutist who will accept only the best regardless of available resources, but I do object to representing half measures as best practices. The Moody Street example “Two Way Cycle Track at Sidewalk Level” is downright dangerous and should be presented as an example of what is unacceptable. Has anyone who worked on the NACTO guidelines actually walked or biked there since the street redesign and bike/streetcar implementation? The pedestrian environment is miserable, bikes are forced to weave around and change grade/pavement constantly (street, “sidewalk level,” street, “sidewalk level,” oops look out for those streetcar tracks), and the streetcar is still pokey and unreliable.

  22. Congested downtown locations *should* have fewer lanes for general vehicles. Transit and bikes move more people more safely and using less space.

  23. Or at least limit automobiles to 20 MPH.

    Then we save money on all the segregated infrastructure construction and maintenance costs, save lives, condense land use, etc.

    Re-engineer the car, not everything else.

  24. If the downtown is that congested, the fewer lanes for cars, the better. Based on the coverage from StreetsblogChi, it really sounds like they need to take another lane away so that buses can pass each other.

  25. The problem is the alternative doesn’t scale very well. Put in very frequently running buses with closely spaced stops and you’ll literally have constant lengthy conflicts with pedestrians crossing the bike lane every block or two. Just because something might work OK 80% or 90% of the time doesn’t mean it’s the only answer. We need to look at streets on a case-by-case basis. It’s sort of like when people see protected bike lanes as the only answer. They’re fine when they can run alongside a natural barrier like a river or railway or park but they’re an epic fail on streets with many intersections.

  26. Get rid of the ability to supply urban grocery stores too. What about the majority of people who can’t ride bicycles, including plenty who can’t walk very far either, just ban them from the city too?

  27. If we limit cars to 20 mph we also limit trucks and buses to that same ridiculously-slow speed, which will require twice and many buses and trucks to haul the same amount of freight, and greatly raise the cost of goods shipped by truck too. You want to double the cost of groceries that would be one way to do it.

    Denver’s RTD is already on the ragged edge of bankruptcy thanks to blowing 5 times as much as they should have on the DIA station, and do keep in-mind that residents of the north side of the city do not have either the trains that we all paid for, nor even bus routes that were promised to us 15 years ago either, thanks to a bizarre case of grandiose overspending on FasTracks.

    Considering that there are far less bicycles on our roads than cars, in-fact no more than one bicycle for every 15-20 cars, and Denver is projected to grow by another million or more people over the next 20 years while at the same time RTD is nearly 80% short on funding projected needs as well as transit promises already made I am thinking that if we must make cuts anywhere why not on bicycle infrastructure funding? How about we adopt the same strategy as Seattle and move away from trying to accommodate bicycles on streets to building off-street bike lanes instead, which are safer?

    I would recommend a thorough reading of DRCOG’s Fiscally-Constrained Regional Transportation Plan for the 7-County Metro-Denver area first, as both CDOT and bicycle needs funding is also in critically-short supply even though our economy is as hot as it has been in 20 years.

    Either that or have one bicycle-friendly artery for every several powered-vehicle-only arteries? The City of Cleveland has done this on their Chester / Euclid / Carnegie / Cedar corridor east of downtown, which has center-riding BRT running on Euclid Ave along with the bicycles, as well as a single traffic lane in each direction, while Chester and Carnegie, six lanes each, are entirely for cars, buses, and trucks, with no bikes allowed?


    2040 Fiscally Constrained Regional Transportation Plan, Denver Regional Council of Governments, 2015: https://drcog.org/sites/drcog/files/resources/2040%20Fiscally%20Constrained%20Regional%20Transportation%20Plan.pdf

  28. Too bad that Denver’s early town planners didn’t leave enough right of way for boulevards as center-riding transit is far-safer for bicyclists than is right lane-riding public transit, also according to NACTO too.

Leave a Reply

Your email address will not be published. Required fields are marked *


NACTO Previews a Progressive Design Guide for City Streets

One of the interesting developments to come out of the National Association of City Transportation Officials “Designing Cities” conference (currently in its second day) was the announcement of a wide-ranging new design guide to be released next year. NACTO’s “Urban Streets Design Guide” will show how streets of every size can be re-oriented to prioritize […]

FHWA to Engineers: Go Ahead and Use City-Friendly Street Designs

The heavyweights of American transportation engineering continue to warm up to design guides that prioritize walking, biking, and transit on city streets. On Friday, the Federal Highway Administration made clear that it endorses the National Association of City Transportation Officials’ Urban Street Design Guide, which features street treatments like protected bike lanes that you won’t find in […]
What's keeping cities from rolling out changes like this faster? NACTO wants to know. Photo: Nathan Roseberry (CDOT) via NACTO/Flickr

NACTO Wants to Find Out How Cities Can Design Better Streets, Faster

The National Association of City Transportation Officials, representing more than 50 urban transportation departments across the United States, is known for street design guides that prioritize walking, bicycling, and transit. Now the organization is turning its attention to the nuts-and-bolts of how city bureaucracies can implement these designs in a timely manner, so meaningful change can happen within our lifetimes.

NACTO Beats the Clock With Quick Update of Bike Guide

Once again, the National Association of City Transportation Officials has proven what an agile, modern coalition of transportation agencies is capable of. It was just a year and a half ago that NACTO released its first Urban Bikeway Design Guide and today, it’s released the first update to that guide. NACTO’s guide is far ahead of the industry standard, […]

Washington DOT Officially Endorses NACTO Street Design Guide

One state down, 49 to go. The Washington state Department of Transportation is the first state DOT to endorse the Urban Street Design Guide put out by the National Association of City Transportation Officials’ (NACTO). The manual provides instruction on creating treatments like protected bike lanes, transit-priority streets, and parklets, which aren’t included in the […]