stop spacing: risks of multiple patterns

In a recent post on stop spacing, I quoted an eloquent defense of very closely-spaced stops based on the needs of mobility-limited persons.  This view is unfortunately in tension with the need to move stops as far apart as possible to increase the speed and reliability of operations, and thus attract more passengers.

I was surprised at how many comments suggested that the answer is to provide a mixture of local-stop and limited-stop or "Rapid" services.  This is absolutely the right thing to do on the extremely major streets where you can afford very high frequency (say, every 10 minutes or better) on both patterns.  Most New World cities have just a handful of these streets.  Examples include Mission and Geary in San Francisco, Broadway and 41st Avenue in Vancouver, Western Avenue in Chicago and Wilshire in Los Angeles.  Key features of these streets are (a) very high demand supporting two frequent services and (b) relatively long trips, so that speed advantage of a rapid stopping pattern outweighs the longer walking time it may require. 

But if you can't afford high frequencies, overlaying local-stop (every 200m or less) with Rapid or limited-stop service (every 800m or more) can be really unsatisfying.  Should you wait 11 minutes for a local at your stop, or walk to a Rapid stop 400m away where the next bus comes in 14 minutes but might be faster?  Those are the uninspiring choices presented to a customer when the frequencies are only, say, every 15 minutes but two patterns are being offered. 

When you consider the major streets that support frequent locals plus frequent rapid services, Seattle's long and busy Aurora Avenue might come to mind, but in fact, King County Metro abolished that pattern a few years ago, creating instead a single stopping pattern so that they could run the highest possible frequency.  That's the key.  Especially for trips of under 10 km or so, waiting time easily overwhelms in-vehicle time in determining door-to-door travel time.  So in those cases a reasonable "compromise" stop spacing — not as close as senior/disabled advocates want, nor as far apart as speed advocates want — is actually the fastest at getting everyone where they're going.

Another approach, which I advocate looking at, is to accept that the constituency for very closely-spaced stops may also accept poorer frequency.  If you look at the part of a route that is halfway between two Rapid stops, and thus most dependent on the local stops, and you then subtract all the people there who are willing to walk 400m to the rapid stop, you end up with a fairly small number of people.  So perhaps locals should be less frequent than rapids.  Transit agencies sometimes try to be neutral about this, carefully calibrating local vs rapid service based solely on current ridership.  But in fact, transit agencies have a strong reason to prefer rapids: faster service is cheaper service to operate, because transit vehicles complete their cycles is less time, and we pay drivers by time, not distance.

But it's definitely not adequate to say that we can resolve the conflict between close and wide stop spacing simply by running two separate lines on the same street.  We can in a few places, and if public transit had a lot more money we'd do it in a few more.  But transit agencies need a stop spacing policy that works for the more ordinary street, where you can afford maybe 10-15 minute frequency on just one line.  That means just one stopping pattern, so we have to pick one. 


28 Responses to stop spacing: risks of multiple patterns

  1. Jonathon September 13, 2011 at 4:11 am #

    Ottawa, given that is has one of the most developed BRT systems in the world, has a very odd approach to this. In Ottawa, even though there is a large network of express routes, there is no such thing as limited stop service. The term ‘express’ only refers to a transferless trip downtown, rather than having to wait for a local bus to take you (along the exact same route) to the nearest transitway entry where you would then wait for another bus. Extremely frequent transitway routes also use local streets to get to some destinations; the 94/95 uses a long stretch of Woodroffe Avenue to get from Barrhaven to the busway, but despite a frequency of 5 minutes or better, and it being the only link from Barrhaven to central Ottawa, it makes every local stop along Woodroffe.

  2. Michael Sypolt September 13, 2011 at 4:32 am #

    The notion of running the rapids more frequently is well done in the northern suburbs of Toronto in York County along Yonge. The VIVA rapid route runs every 15 minutes off peak (7 days per week) and 7.5 minutes during peak times. The articulated buses stop every kilometre and operate to approximately midnight every night between the northern terminous of the Yonge Line of the metro and Newmarket. The local 98/99 routes operate less frequently (approximately half as frequent as the VIVA. These routes also operate night service until about 3am every night. There was definitely ridership to support the service, despite the suburban character of much of the corridor.
    What I thought was most successful was the visibility of the VIVA branding at the rapid stops, countdown clocks at stations, and a capable dispatching system to ensure quick deployment of additional buses or otherwise adjusting service when vehicles are running late or there is unmet passenger demand. When I visited on a sunday in May this year, the effective headway with all the dispatching help was closer to 10 minutes than the advertised 15 minutes.

  3. Zoltán September 13, 2011 at 4:35 am #

    An exception is where a rapid route is very well anchored, so that a whole lot of passengers are going all the way, and possibly have transferred from further afield. This makes a rapid route do what an express route might otherwise be used for, but additionally serve some of the passengers along the way, and by virtue of that offer superior frequency.
    This could well be seen to legitimate Baltimore’s 48 (York Road) route, which makes limited stops every 15 minutes, overlaid on a local route every 15 minutes. It carries quite a lot of people who have transferred downtown and wish to reach the end of the route in Towson, an large independent town centre around which the suburbs have grown.

  4. Zoltán September 13, 2011 at 4:43 am #

    Apart from this, I’d agree that running both a limited and a local each, say, every 15 minutes or less, isn’t necessarily beneficial to many passengers.
    I’d add an additional criticism – increased anxiety. When it comes to leaving your house and walking to either a limited+local or local stop, and at the former a local bus coming which you can choose whether to board, it’s likely to be uncertain what choice will get you to your destination quicker.
    At times along various corridors in various cities, I’ve walked towards a limited+local stop, and been annoyed as a local that I could have boarded instead then passed me. At other times, I’ve gone to a local stop and been annoyed as a limited passed me without stopping. This isn’t a problem if you know the next bus will be in five minutes, but is annoying if you could be waiting fifteen or even thirty as a result.

  5. Zoltán September 13, 2011 at 4:49 am #

    Finally, before I shut up, I shall give the example of certain areas of Leeds, which maintains strict stop spacing of about every 400m.
    A comprehensive network exists of frequent routes running every 10 or every 7/8 minutes along main arterials. In addition, there are some routes that run every 30-60 minutes, that have similar stop spacing, but penetrate areas away from the main arterials. In that way, those that are more resistant to long walks than low frequency have a service closer to their door than a service making many stops along an arterial would provide. This is a potentially good model for cities that want to widen stop spacing and also consolidate routes on frequent corridors, but need to maintain coverage service on other roads.

  6. Robert Madison September 13, 2011 at 5:58 am #

    The idea of running the locals at less frequency than the limiteds is something that Chicago/CTA tried a couple of years ago. The experiment was that the limiteds would have double the frequency of the local service.
    I don’t know what the results were, but unfortunately the whole experiment had to be scrapped in the round of service cuts early last year, when the limiteds were all discontinued.

  7. Jacob Mason September 13, 2011 at 7:15 am #

    I’m surprised you left out NYC on the list of cities with local and limited stop routes. We’ve been doing it for years, and the new Select Bus Service (BRT-light) is really just an enhanced version of existing limited-stop service, via amenities such as off-board fare collection, dedicated bus lanes, signal priority, and branding. One thing to consider for locations where regular and limited routes coexist is the coordination among them, particularly at stops with both services. As a customer I often have to travel a distance where it makes little difference which service I take, so therefore I take whichever one comes first, right? Not always. Many of the SBS services here have their own stops, which were often down the block and even across the street from the local stops. This removes the double-service advantage of having both. The same is true at some of the subway stations with express and local trains, where different platforms serve express and local lines. In these cases you often end up with a weird situation where people wait somewhere in between the two stops and run to whichever comes first. At my old subway stop at 145th Street on the A/C and B/D lines, 2 subway lines stopped on the upper level and two stopped on the lower level. People, including myself, would literally wait halfway up the stairs between the two and quickly move to whichever train came first.

  8. Zoltán September 13, 2011 at 7:42 am #

    @Jacob Mason
    This is a feature that I thoroughly dislike about LA’s metro rapid, where people some times dangerously run over to the rapid stops on the other side of the intersection when a bus approaches. One can understand that buses shouldn’t be backed up at stops, but has anyone heard of double-length bus stops? (Yes, Leeds has, and uses them at a lot of heavily-used stops). An important aspect of legibility is ensuring that passengers are forced to make a choice only when it is relevant to them, for example when a vehicle turns up.

  9. anonymouse September 13, 2011 at 9:49 am #

    On the topic of increased anxiety, I actually get to deal with this every day. My commute involves a transfer, from a route that has both rapid and local service to one that runs every half hour. So I get to decide whether to get on the local that’s here now, or wait for the rapid that might pass the local as it’s waiting at a timepoint. The ride is just short enough that it’s not always better to take the rapid (which comes every 15 minutes). And if I guess wrong, I might end up missing the connection and have to wait for another 29 minutes.

  10. david vartanoff September 13, 2011 at 10:56 am #

    first, I deeply dislike the LA separated stops pattern. second, as to choosing between the local or express(rapid,ltd) the various bus tracking services make this easier.
    @Jarrett, Actually many more routes could easily support dual services in base day, with only the local during evenings. Several East (SF) Bay routes function this way. In the case of the route nearest my home, the locals in the evening are time competitive to the daytime Rapids because both auto interference and ridership drop off.

  11. zefwagner September 13, 2011 at 12:01 pm #

    It never would have occurred to me to try to have the limited and local buses run at equal frequencies. One excellent model is Swift up in Snohomish County north of Seattle. It only stops once a mile and has 10 minute frequency, while an underlying local runs at 30 minute frequency. This means for most people it would make sense to walk farther in exchange for a shorter wait and faster service. Those who want a stop really close to them can deal with using a schedule. That’s a good tradeoff.
    One thing to keep in mind is that if demand all along the line warrants frequent service, better to just make the line frequent rather than try this split model. Part of why Swift makes sense is that right now demand overall is quite low (thus the infrequent local), but certain places have higher demand due to certain stores or transfer points. It is currently a very low-density commercial strip, and part of the point of the Swift Line is to encourage development around its stations. Already dense, built-out areas probably would not benefit as much from a split system like this. A compromise would be BRT with half-mile spacing, with frequencies high enough to outweigh the extra walking time.
    Overall I continue to think something closer to quarter-mile spacing on most local services would be better for most people. Transit agencies should obviously take care to examine different areas to see if certain neighborhoods have a high number of mobility-challenged folks, barriers to walking like bad crossings or lack of sidewalks, and things like hills and interruptions in the street grid. Those factors could easily justify shorter stop spacing, but every 500 feet still seems ridiculously close.

  12. Zoltán September 13, 2011 at 12:14 pm #

    @david vartanoff
    RTI can help somewhat to those deciding which bus to board at a limited+local stop, but is of little use to those waiting at a local stop, unless they have a smartphone and are willing to take in information about multiple routes at multiple stops – which then reduces the simplicity of using the system.
    So while RTI is brilliant in many respects, I don’t think that it fundamentally changes the amount of frequency that is desirable before limited stop routes should be implemented.
    In addition, RTI displays at stops are prohibitively expensive if you have too many stops.
    Daytime-only limited stop operation is a sensible thing to do in many cases, and I can think of a good number of examples where that happens. And yes, evening local buses tend to be as quick or almost as quick as daytime limited stop buses.
    The one criticism is that it can compromise simplicity. That can be remedied by using logical numbering systems (like the 20/720 in LA, 8/48 in Baltimore, 38/38L in SF) and displaying operating hours at stops.

  13. david vartanoff September 13, 2011 at 1:06 pm #

    @Zoltan Even a simple phone can acces RTI in the Bay area and IINM Portland. SMS w/a bus stop # works well letting one p;lan how soon to expect a bus @ which stop. As to clarity of route yes #72R as rapid version of #72 for example.

  14. Zoltán September 13, 2011 at 2:29 pm #

    @David Vartanoff Funny you should quote the number 72. The 72 in Leeds is duplicated by the rapid route numbered… wait for it… X6. Unrelated: The 6, that runs on an unrelated road to the North, while the X6 goes to the West. The 72 shares some of its route with the 14, which has a limited stop equivalent that complements the X6, numbered X11. I hope this is all making sense to you so far.

  15. anonymouse September 13, 2011 at 4:06 pm #

    @david vartanoff that simple phone call is never as simple as you’d like to think. And I quote:
    Welcome to the VTA Customer service center. For automated transit information in English, please press 1. {1} For scheduled bus service, press 1, for light rail service… {1} Enter the route number of your choice, followed by the pound sign {5}{2}{2}{#} Choose your direction. For service to “Eastridge Transit Center” press 1, for service to “Palo Alto Transit Center”, press 2 {2}. For next scheduled bus, press 1… {1}. Route 522 service between Palo Alto Transit Center and Eastridge Transit Center, after the time you have requested, leaves from Eastridge Transit Center at 5:27 pm.
    By this point the bus has usually turned up, or else you’ve had time to walk to the next stop. Okay, maybe this isn’t really on topic, but I just really hate touch tone interfaces. They’re so… cumbersome.

  16. ajedrez September 13, 2011 at 7:51 pm #

    @ Zoltán: I’d just like to make one comment. What should happen is that the local stops are spaced at a distance that if you see one coming up behind you, you can run up to the next stop and catch it. If the stops are already far enough that you can’t realistically run to the next stop, it’s probably better to turn the limiteds into locals.
    As far as waiting at a local/limited stop and having to make a choice, you’re no worse off than if the service was local-only. If the local comes first, it’s basically the same as if the service was local-only, and if the limited comes first, you get a faster trip. Plus, the limiteds have a slightly shorter running time, which decreases the cost for the agency (or they can run more service for the same amount of money)

  17. bardak September 13, 2011 at 8:03 pm #

    There are two Issues I see here. First goes back where this whole topic of stop consolidation from the Portland Transport blog. Zef Wagner was proposing stop consolidation as a way to squeeze a bit service out of portland’s limited funds. Running duplicate service on a corridor is not going to save money so as a solution to the question that was originally posed it does not really help.
    But that being said is still an valid and interesting topic. imho running a local/limited only makes sense if waiting for the limited is always a better choice than taking the local if you are going more than 3 or 4 limited stops. This requires frequent limited service (10min or less to put a number on it), wide stop spacing (1km-ish but exception made based on the route), and a route that already has decent demand(If there is not enough demand to make the local stop a lot relative to the limited you are not going to

  18. Andrew September 13, 2011 at 8:39 pm #

    Mississauga Transit and Brampton Transit in the western suburbs of Toronto also have recently introduced local/express buses. In Mississauga, route 19 (local) and route 103 (express) overlap along Hurontario Street, but the locals run much more frequently than the express, and the express only runs Monday-Saturday until about 9pm. In Brampton, route 2 (local) and 502 (express) overlap along its section of Hurontario Street (part of which is called Main Street), but the expresses run somewhat more frequently than the locals. Route 103 and 502 overlap in Mississauga between Shopper’s World and Square One. The differences in the frequency of local and express buses reflects the differences in the amount of local demand along each section (between Port Credit and Britannia in Mississauga there are many closely spaced apartment buildings and shopping plazas so there is more local demand; between Britannia and the 407 is mostly industrial areas and office parks, and north of the 407 in Brampton is mostly low density residential). Queen Street in Brampton also has local/express service (501 express, 1 local).
    Unfortunately the off peak frequencies on Main Street in Brampton are bad (20 minutes for the 502 and 30 minutes for the 2). The 502 was packed though when I took it, on a Saturday no less – it definitely needs more service.

  19. Zoltán September 14, 2011 at 3:55 am #

    @ajedrez Having to look over my shoulder just in case a bus might be coming isn’t my idea of fun. I’d far rather be sure of what bus stop I need to go to, and what bus I want to board.
    Also, if a local stopping every 200m comes first, it’s not going to get me to my destination as quickly as a single bus route stopping every 400m.
    @David Vartanoff @anonymouse I discounted text services, because they are normally quite expensive, and only give you a single snapshot that’s subject to change with traffic conditions between where the bus is and where you’re waiting.

  20. david vartanoff September 14, 2011 at 9:03 am #

    The quality of RTI by phone varies widely. Emery Go Round is very simple. The # to call and the stop # are on the bus stop sign; you call, you get arrival predictions. quick! I use SMS for AC Transit because it is also quick.

  21. Eric Doherty September 14, 2011 at 9:54 am #

    I agree with everything Jarrett writes here, except for qualifying the bit about how few streets there are where rapids and local service can co-exist well.
    The number of these streets is not fixed, it depends on how good the transit service is and how high the ridership is. Greatly improve both, and you increase the number of arterials that can support B-line style express service. One handful quickly becomes two handfuls, and then you have a network of express routes.
    But my assumption here is adequate resources for the operating expenses, not cutbacks.

  22. Wad September 14, 2011 at 1:49 pm #

    @Zoltan, people in L.A. hate the “Rapid shuffle” of darting across the street to catch the different bus, too.
    Metro started to consolidate local/Rapid stops on the newer lines. There doesn’t seem to be any inclination to do so on the earlier ones.
    Rapid stops were spaced apart deliberately because “BRT is special.” Those separated stops are “stations” in theory. Limited-stop (300-series) buses share stops with locals, and the Rapids did when they were 300-series lines.

  23. Jarrett September 14, 2011 at 2:49 pm #

    @Eric. We don’t disagree at all. But even in Vancouver, which has an exceptional number of streets that could or do support local+rapid lines (Broadway, Hastings, 41st just for starters), there are many more than don’t. A solution that works for a whole frequent network grid has to work for those other streets. I suspect that the options are (a) locals with wide but accessible spacing (400m?) or (b) rapids (800m spacing) plus infrequent locals with very close (200m) spacing.
    Note also that you’re in Canada, where transit funding is stable. US transit funding is extremely volatile, so the ability to lock down patterns is more limited.

  24. ajedrez September 14, 2011 at 6:58 pm #

    @Zoltán: Well, obviously, I wouldn’t consider it an ideal situation, but it’s not that bad. I mean, when I go to school, the closest stop is a local stop, so I have to walk a couple of more blocks to a limited stop, looking over my shoulder in the process to make sure a bus doesn’t pass me. That’s why it’s important that the limited stops aren’t spaced too widely apart.
    And obviously, the limited is faster if you can catch it, but I’m saying that if you are at the stop and a local comes first, the additional time really isn’t going to be great (unless you’re dealing with a whole bunch of traffic lights or something). If you’re going a long distance, you should be on the limited anyway.
    Plus, the advantage of the limited is that it takes the long distance passengers, so the locals are sped up by slightly shorter boarding times at the limited stops.

  25. Wad September 14, 2011 at 8:40 pm #

    @AJedrez, L.A. has a unique approach in the assignment of limited-stop routes. This doesn’t necessarily apply for the Rapids.
    Surprisingly, a Metro 300-series limited isn’t segmented around speed or longer distances. The buses are faster as a consequence of fewer stops, but limiteds are primarily a tool to manage crowding on the underlying local line. A local line must have the conditions of overcrowding and heavy transfer activity in order to get a limited branch.

  26. Ben Smith September 15, 2011 at 6:30 am #

    On heavier used trunk routes, I don’t see a problem with running local buses with 400m stop spacing with rapid buses stopping every 800m, assuming a strict half mile grid road layout. From an urban planning perspective, think of each stop representing an urban village. Locals stop at every village, while rapids only stop at major villages.
    As for differences in speed, I’ve roughly calculated that a local would run at 20km/h average, while the rapid would move at 30km/h average. This means over a 30 minute trip, the rapid would be ahead by 5km, or 6 small or 3 major blocks.

  27. Nathanael September 16, 2011 at 5:10 pm #

    Bluntly, the crucial question is how *many* mobility-limited people there are.
    If there are a *lot* in a given area, or going to a given area, add extra stops to the route!
    But of course this implies an area with high residential or commercial or job density! And indeed, we find that (in places which no longer use the “flagstop at every corner” model) many routes have more frequent stops / closer stop spacing in the highest density areas.
    Otherwise, if there are *few* limited-mobility people, remember that this is *mass* transportation and provide an alternative service (dial-a-ride?) for the very small number.

  28. Paul C September 18, 2011 at 10:18 pm #

    On 41st ave in Vancouver. Because they run the local route with a higher frequency than the limited stop route. Most times you will end up getting a local route drive up before the limited stop route. Which leaves you with the problem of should you get on that bus or wait for the limited stop.
    I’m not sure exactly why. Maybe it is just do to a lack of resources (ie not having enough 60′ buses). But I’ve always felt. Since the majority of the people on a route will more than likely be getting on that route at a major intersection or transit point and that most of those people will be getting off the bus at a major intersection or transit point to transfer. That it would probably be better to run the limited stop route more often compared to the local route. Right now on 41st ave when both are running they have the 41 (local) every 5-6 minutes and the 43 (limited) every 10 minutes. Also since the 41 buses are 40′ and the 43 are 60′. It just seems like it would be better to switch those times around. Although somehow I feel they want to do this but just don’t have the resources to do this.
    As for my comment in my first paragraph. The problem of whether to wait for the limited stop bus or get on the local bus has been substantially solved by Translink implementing a real-time map of the where the buses are on the route. However it is still in beta and has a lot of kinks to work out. I’ve used this new feature, something I’ve wanted to see in the past. And it has helped me a lot on whether I should get on one bus or wait for another.