Is It Time for Muni to Consolidate Bus Stops?

Picture_8.jpg1 California line bus stop map. Does it need so many stops? Click for the full PDF.

As the MTA Board considers solutions to reduce its enormous $129 million deficit, bus stop consolidation should be at the top of its agenda.  Bus stop removal is one of the easiest engineering solutions to speed up service, but the political fallout could be significant without a groundswell of organizing and support from those who would benefit from service improvements. 

The MTA’s current bus stop spacing policy is the following:

1. Passenger stop spacing should be approximately 800-1,000 feet on motorcoach and trolleycoach routes except where
there are steep grades, and 1,000-1,200 feet between stops on LRV surface
lines.

2. On streets with grades of over 10
percent, stops should be spaced 500-600 feet apart. On streets with grades of
over 15 percent, such as on Castro between 22nd Street and 24th Street, stops may be spaced as close
as 300-400 feet.  There are no special grade guidelines for surface rail
(Muni Metro and Historic Streetcar) because the technology limits operation
well under 10 percent inclines.

Bus stop spacing is usually balanced by density, topography and the
guidelines you see above. Said Aimee Gauthier at the Institute for
Transportation and Development Policy, based in New York:

Ideally,
bus stops are determined not by distance (like every block) but by
actual passenger volumes. In general, bus stops should be spaced so
that they are a reasonable walk between stops (that differs according
to who you are talking to and the context [uses, activities around
there], but usually it is a within a 10 – 15 minute walk). 

Many stops in San Francisco, however, do not fit the above criteria, including instances of single blocks with multiple stops.  If Muni eliminated stops now, or at
least began an experiment, trying out one line, it could save them
precious running time costs and improve service and reliability.

Muni actually revises bus stop spacing [PDF] more frequently than one might think, though in a recent interview, MTA Executive Director Nat
Ford told Streetsblog San Francisco they aren’t planning to release system wide bus stop consolidation
recommendations until after the budget process is over.

Walk SF Director Manish Champsee analyzed MTA data on several lines and his determinations strongly support consolidation:

On
the 14 Mission if we were to remove every other stop (i.e. only run the
14L as opposed to the 14), MTA could provide about 20 percent more service
than they currently do and save about $5 million.  To give you an idea
of what $5 million is, MTA is proposing a $5 increase in the adult Fastpass. This would yield $7 million per year.  They are also
proposing an extra $5 increase in the senior/youth/disabled pass which
would bring in $2.5 million per year.

Under a declaration of fiscal emergency,
MTA could act now without a lengthy environmental review process and
potentially drawn out political debate to eliminate redundant stops on
lines like the 1-California, and any number of lines, to save millions.
By some calculations, eliminating 10 percent of bus stops could save
Muni at least $14 million a year.

SF State Geography Professor Jason Henderson thinks the whole approach is problematic:

Optimizing the system is a good thing, but MTA is going about it the wrong way. The cutting of stops and the cutting of routes is endemic of the way the MTA is hamstrung because they don’t want to raise the price of driving while lowering the conveneince of driving. Instead, from all I can tell MTA is focused on how to raise the price of transit fares and lower the convenience of transit.  MUNI needs to expand capacity significantly, not contract.

A couple years ago, Mayor Gavin Newsom vocally supported stop removal, so there is reason to believe he will come out strongly in support of the MTA for consolidation when the agency releases its proposal. Without a vocal and organized constituency, however, voicing support for service improvements and time savings through stop removal, the issue will not be popular politically.

The Board of Supervisors would not be able to reject a consolidation proposal outright, but you can imagine the political reaction when constituents bemoan the loss of the stop they have taken for granted for decades. No supervisor wants to be seen as creating hardship for the poor people or the elderly.  Should make for some dramatic hearings, if we get that far.

  • Is that boarding and alighting data available to the public?

  • friscolex

    I’ve always thought it totally ridiculous that so many buses stop every block– and sometimes twice on one block! I understand the grading and accessibility issues, but I’d happily walk another block for my stop if it would make my commute shorter.

  • DaveO

    Now this is something I can get behind. Remove a bunch of stops and increase the visibility of the ones that remain. For LRV, convert all remaining stops into accessible platforms.

  • jim

    Jeffrey,

    Muni stop and boarding data can be found at
    http://www.sfmta.com/cms/rtep/tepdataindx.htm

  • Pat

    Maybe explain the graphic you use for your headline?

  • Jim: the radius is proportional to total boarding+alighting at that stop, the green angle is the fraction of boarding, and the yellow angle is the fraction of alighting.

  • Er, that was for Pat. To Jim, thanks for the link to the TEP data.

  • “they don’t want to raise the price of driving while lowering the convenience of driving. Instead, from all I can tell MTA is focused on how to raise the price of transit fares and lower the convenience of transit.”

    Word! I can reluctantly stand by a Muni service reduction if it is part of a larger plan that reduces the public services to private motorists by just as much. Why should responsible transit riders be the only ones that feel the pain?

    Remember that stop consolidation is not an alternative to service cuts, it is a service cut. If you want to aver that faster trip time should take priority over passenger convenience, you need to explain that argument. IMHO it’s far from given.

    If, on the other hand, you want to argue that convenient local bus transit (which we pretty much have now) should be complimented by rapid or express bus service, that’s another argument completely. That increases the choices available to transit riders.

    Rather than consolidating stops, why don’t we explore the idea of running fewer local buses and more express buses along some lines? People who prefer a short walk and a longer bus ride will still ride Muni, and people who prefer the opposite will still ride Muni. More people using Muni… isn’t that what we want, after all?

  • Gregorio

    Please oh please let this happen! It would speed up the lines without significantly affecting convenience, and it would make for a bus ride where you’re not constantly lurching back and forth. Can people REALLY not stand to walk and extra 500 feet?

  • theo

    Rather than consolidating stops, why don’t we explore the idea of running fewer local buses and more express buses along some lines?

    Josh, many of the lines that would massively benefit from stop consolidation aren’t downtown commuter lines, and they also don’t have enough ridership to offer express service.

    Nonetheless, 95% of riders on those lines would experience increased service after sensible stop consolidation. For the 5% who would be minorly inconvenienced, sorry, but ultramajority rules.

    Check out the maps for a line like the 6 Parnassus or the 23 Monterey. Some stops are so close together and underutilized it’s embarrassing. You can barely see them on the map because there are so few passengers.

    http://www.sfmta.com/cms/rtep/tepdataindx.htm

    Muni has the lowest average stop-to-stop distances of any metropolitan bus service in the US, possibly the world. Also the lowest average speed. You think those might be related?

  • Can people REALLY not stand to walk and extra 500 feet?

    When they consolidated some of the 38-Geary stops a few years ago, where at both ends of the same block, someone was quoted in a story complaining that half a block was too much. I think the root of the problem is we design for the lowest common denominator instead of the greatest public good or the greatest long-term viability.

    Decision makers and politicians don’t want to be the bad guy who is makes it harder for a 90 year old with a walker to get from point A to point B so that able bodied 30-somethings who opted to sell their cars can get to work faster. But the fact of the matter is it’s able-bodied 30-somethings that generate the revenue that keeps the city running and we’re doing a terrible job helping them get to and from work so that we can afford things like Muni.

  • Matt

    “I think the root of the problem is we design for the lowest common denominator instead of the greatest public good or the greatest long-term viability.”

    Hear hear! There’s something to be said for TOO MUCH minority power. Why do people tolerate fewer stops on a train or subway but not on a bus? It makes no sense! Let’s eliminate the stops and finally have sensible buses.

    Now if we could only find a way to punish the jerks who hog up the Bus Only lanes 🙂

  • greg h.

    This is a fantastic idea. I never thought I’d hear anyone seriously consider this. I’m dizzy with false hope.

  • rzu

    Let’s not be afraid to call stop consolidation what it really is: a service cut. The governor’s slashing of operating subsidies has made service cuts necessary, _unless_ the MTA is willing to agressively go after funding. And we know how good they are at that…

    On the other hand, this may be a better way to handle service cuts than say, ending non-owl service early or running less frequent buses. I believe that what is being attempted here is to let the MTA know that there are other options to save money. There is no magic budget fix, but ideas like these can cut back on costs with less negative consequences. And if it came down to an option between service cuts and fare increases, I know many folks would opt for stop consolidation over a $60 fast pass.

  • david vartanoff

    Try this. If a stop is lightly used the bus RARE:Y stops there. Thus other than Muni’s minimalist signage and denying someone a parking space it costs nothing to retain. If OTOH even the emblematic two stops in a single (probably extra long) block case each has hundreds using the stop every day that is the market–serve it! The local to ltd split is the real issue. Part of the utility of local bus service is being able to do multiple stop errands along commercial strips. Expecting a local to be a limited is silly. Any well used route should have both services as the ridership patterns dictate.

  • Interesting idea. Maybe Muni could take out some of the 18-46th avenue stops on 46th avenue. I’m starting to notice passengers who board at one stop and they get off AT THE NEXT BLOCK.

  • I’m aghast at the ‘screw the minority’ attitude that some people are touting in this forum. Isn’t that the attitude that causes public transit to get screwed in favour of private cars, time and time again?

    In general I’m in favour of stop reduction, but we need to remember that for some people, walking an extra 1000 feet is actually a really big deal. That arrogant Republican ‘suck it up’ attitude is as short-sighted as it is inconsiderate; some people just can’t walk that block uphill to get to the stop. They just can’t. And most of them paid into the system when they were 30-somethings like me, and it’s unjust to cut them off from the world just because they can’t be worker ants anymore. I’m doing my best to take care of my body, but one day 1000 feet will be a big deal for me, too.

    I don’t understand why the idea of more limited stop services is being called ‘silly’ and whatnot. It sounds like a win-win compromise to me! Why shouldn’t there be a 22L and a 44L? What’s so silly about more 14Ls, and even the occasional 14X outside of peak times?

    And it’s high time we push for legislation to allow photo enforcement of diamond lanes and parking in bus stops. The other day I rode the 14 from the bottom of Bernal Hill back to 14th Street, and I counted no fewer than 13 vehicles parked in bus stops! That’s 13×$250=$3250 worth of revenue that could have been raised just in those 20 minutes, and the trip would’ve taken less time if Muni operators and passengers didn’t have to negotiate their way around SUVs that are too big for this town.

  • theo

    Try this. If a stop is lightly used the bus RARE:Y stops there. Thus other than Muni’s minimalist signage and denying someone a parking space it costs nothing to retain.

    No. Say only 1 of 100 passengers gets off at a lightly used stop. For night buses with 10 passengers, only 1 in 10 will have to stop. For fully loaded rush hour buses with 50 passengers, 1 in 2 will be forced to stop to let one person off.

    When buses are empty, those stops are irrelevant. When they’re full, they matter — they inconvenience as many passengers as possible. That completely sucks.

    Besides, the driver always has to slow down to see whether there are passengers waiting at a stop or not. This costs 5-15 seconds per run even for stops that are never used.

  • theo

    I’m aghast at the ‘screw the minority’ attitude that some people are touting in this forum. Isn’t that the attitude that causes public transit to get screwed in favour of private cars, time and time again?

    Actually, in SF the majority of workers take transit rather than drive — we’re a “transit first” city you know. I’m all for that, as is every other progressive Democrat posting here.

    You know what really screws up public transit? The idea that it should be run as a money losing charity taxi service for people who can’t walk 100 feet.

    We already have a service for that and it’s called paratransit. For every potential paratransit user, we already spend thousands of dollars yearly ensuring they have adequate mobility. Tell me why we should piss away thousands more and spoil public transit for the other 99% of the city just so a few hundred people can have a regular bus stopping at their front door.

    I want to live in a progressive city that actually works. There is no one size fits all solution for the goals of public transit and paratransit — they must be separate services if they’re going to work at all.

    I don’t understand why the idea of more limited stop services is being called ‘silly’ and whatnot. It sounds like a win-win compromise to me! Why shouldn’t there be a 22L and a 44L? What’s so silly about more 14Ls, and even the occasional 14X outside of peak times?

    A reasonable question. There’s nothing silly about the 14L, that’s is why it’s being discussed. On the other hand, the 44L would be pointless. Limited service buses have to run often enough that it’s worth waiting for one (or at least timing your departure). If they only run every 30 or 40 minutes, you’re almost always better off taking the local, and limited buses will end up running empty and wasting resources. That’s why you currently only see limited buses on the busiest lines (9, 38, etc.)

  • Paratransit is very expensive, and the ADA says that government needs to provide mainstreaming accommodation as much as possible.

    This meme of bus stop consolidation has fashioned itself into group think, and whenever group think trumps empirical scientific arguments, we get ourselves into situations where simply because we believer that we are correct and just, that our ideas deserve for that and that alone to prevail.

    See the bike plan EIR for an example of how group think works.

    Can we please dispense with the notion that because there are a handful of blocks, City Hall and Mission High School come to mind, with two stops, that such instances represent anything close to the general case? Were we to consolidate those instances, there would be no measurable time savings to the system period. Following that logic, we could just have two stops on a line, both terminals and get really fast service. Science ignores the outliers, the endpoint extrema in analyzing systems behavior as non-representative.

    I want to see specific proposals for consolidating stops that show the mechanics of how it would work on any given line and show projected savings. The MTA did not provide anything of the sort during the TEP process.

    Here again are the maps of “as the crow flies” stop spacing, which means that when the bus turns, the hypotenuse rather than the legs is measured, and that end of line data should be ignored for similar reasons:

    http://cybre.net/pub/muni/stops

    Colder colors mean stops closer together; warmer colors mean stops further apart. The system comes very close to the 700-800′ average stop distance, based on the maps.

    The data are there for your perusal. It was like pulling teeth to get the GIS and line stop data out of the MTA. Please show me a line where you could get more than a 5% stop consolidation and how that would be beneficial. Please do not use the 14 line as an example, as the MTA proposes limited service which would play well with the local service.

    This is not about screwing the minority, rather about privileging better off commuters who are deemed as “transit choice” riders by screwing the generally poorer and older transit dependent populations who depend on the system to get around the City for their daily routine. I would imagine that there would be fertile fields for an ADA lawsuit if existing accommodations which served disabled passengers were eliminated given the expressed rationale of the MTA.

    Three extra blocks for someone with chronic arthritis, is the same as 8 blocks for someone without mobility impairment.

    As an example of bad science, during the TEP, the MTA made the error of assuming that the automated passenger counts on segments of a line represented individuals rather than a variable cloud of riders. That is, if the #6 line over the hill btwn the Haight and UCSF only had 50 APC ticks, then that represented only 50 people. Turns out, that those 50 ticks represented 300 people because not all of them used the service every day, perhaps three times per week. They made themselves known at community meetings, ah tell yew what!I see a similar methodological error being repeated here again as gospel absent evidence.

    The antidote for groupthink is science, not more group think; keep your beliefs in church.

    Like your third grade math teacher said: show me the numbers and show your work.

    -marc

  • theo

    Marc, nice work with the stop interval maps.

    You claim to see 800 foot spacing on those maps. I look at them, and I see a sea of green 500-700 foot spacing, which is unacceptably low except in the steepest areas of the city. If one of those stops were removed, it would add 1 minute’s brisk walk for a healthy person, 2 minutes’ walk for an infirm elderly person. They would easily save that time on the rest of the bus route.

    Look at, for instance, the lines I mentioned above (6 and 23). It appears to me that about 1/4 of stops could be consolidated, the result being 1100-1200 foot spacing. If stop locations were changed, lower spacing could be achieved.

    I have traveled to or lived in many cities, and not a single one has so many bus stops. SF has gone too far in the direction of promoting universal access to mainstream transit service, to the detriment of the entire transit system.

    I see the problem nearly every day when I ride Muni. Full streetcars stopping to let 1 or 2 people off, who then take off at a healthy pace in the direction of the next stop.

    It has gotten me to the point where I pre-consolidate my stops by alighting at the same time as other passengers, in order not to inconvenience the passengers still on the bus. So don’t tell me about “privileging better off commuters.”

    No one wants to go too far in the other direction and eliminate thousands of stops. But on most of those lines, a 10% reduction, which is all that is realistically being contemplated, would have minimal negative effects.

  • theo

    Also, Marcos, I disagree very strongly with your characterization of the TEP counts as “bad science.”

    Passenger counts are a valid measurement of a variable — transit trips. They are weighted towards frequent Muni riders, many of whom depend on its service to get to their daily jobs on time. From your rhetoric, I get the impression that you don’t think commuting is a valid use of public transit and that “better off” commuters should all buy cars, with the obvious negative externalities.

    Number of individual passengers is a valid measurement of another variable — taxpayers served by the system, including occasional riders.

    You could also measure potential riders, i.e. people living or working within a certain radius of a stop.

    These variables all have different uses, and as long as we’re clear about which one we’re optimizing, it’s “good science.”

    Ironically, you are yourself guilty of exactly the methodological confusion you object to. First you claim that a passenger count of 50 reflects 300 people using the service 3x/week (that’s an error, it would be more like 1x/week at that rate). Then you say you’re protecting the “generally poorer and older transit dependent populations who depend on the system to get around the City for their daily routine.” If they’re actually riding daily, they’re being counted daily in the TEP statistics. See how that works?

    I also went to some of the TEP meetings, and it was apparent to me that many of the attendees didn’t use transit nearly as often as they wanted to believe they did. That’s fine — infrequent or potential riders deserve service too — but “one person with unlimited free time for attending community meetings, one vote” is not the fairest or the most efficient way to make decisions affecting everyone.

  • Why shouldn’t there be a 22L

    The 22 is a trolley with overhead wires. My understanding is this means that they can’t pass each other, at least not easily. The line is pretty long, from Potrero Hill to the Marina – a Limited could catch a local unless they are spaced out, perhaps beyond current frequency levels?

  • Thanks for the props, Theo.

    I see 700-800 foot spacings on the map.

    First off, the standards call for 800′ spacing on the flats. If you want to go to 1/4 mile spacing, then that requires refining the existing stop spacing standards. The nebulous proposal as is calls for 800′ stop spacing standards. If you want to change the standards to 1/4 mile, then good luck with that political process, and I want video tapes of those hearings and public meetings! Wear yer asbestos suit.

    Next, the mechanics of increasing the stop distance by 1/7 or 1/8 to achieve the set standard means a full reconfiguration of the all stops throughout the line. And that would only gain you 1/7 or 1/8 fewer stops. If more than 7 or so folks board at each stop, then you begin to see dwell times increase, wiping out some of those gains. So in the real world, we’ll probably see high single digit percentage increases in speed max, a large fraction of one mile per hour, hardly the stuff that can generate the kind of savings required to speed up the line.

    Passenger counts are raw data of boardings and deboardings, no semantics more significant that that can be divined from those data until you go to a communty meeting and find those 50 boardings translate into proximate service for 300 people living up on a hill. That was where the bad science came in, misinterpreting the data based on the desired outcome rather than allowing the data to speak for themselves and letting the cards fall where they would.

    The 50/300 number was based on the #6 line. At our TEP meeting where members of the public came out, that was their story, of seniors who stay in their homes because they have a straight shot downtown via the #6 from their homes near Ashbury Heights. The TEP suggested they try the 33 Godot as a substitute and to trust them because service would be speeded up–waving of the hands!!! I’d never seen any of the folks from Ashbury Heights who came to the TEP ever before at a community meeting.

    My statement at the time was that 50 represented a larger cloud, I believe the language I use was “perhaps 300” people.

    I was on the TEP CAC as a member of the Transit Justice Coalition, I ride the system with and without my bike and I bike without taking the system, for full disclosure.

    Commuting is a valid use of public transit. The fact is that the TEP made the call to screw transit dependent and to appeal to transit choice, and that the goal was to slim down the system so as to optimize commute routes. This is their language not mine.

    My position is that this is a very complex system which has to balance many different needs if we are to create an attractive alternative to private autos. And as a progressive, my position is that we need to take special care to ensure that the less powerful are not stomped by the more powerful solely for their convenience.

    The argument I’m hearing is the same as one that would have us cut payroll taxes for social security and by extension benefits because that money is better spent in the “real” economy by those who “earn” it rather than by seniors who really aren’t that economically productive.

    Politics can be rocket science to the extent that you’ve got to maximize multiple variables, and the set of politically optimal solutions to one problem are generally technically suboptimal. That is what happens when you’ve got to balance myriad conflicting interests and are too cheap to spend real money to solve the problem in the optimal way that other world class cities have–with rapid transit subways. Suboptimal resources lead to suboptimal solutions and in that case, we need to ensure that all get screwed equally.

    I am also not comfortable with “one nonprofit, one vote” that allocates political power according to the ability to secure grant funding. We’ve seen how well that works with bicycle and housing advocacy, that model needs to be dispensed with.

    -marc

  • jim

    @marcos: To address your points:

    1. No one ever said that two stops per block is the general case. The general case is 1 stop per block and this is too many. The Muni guidelines call for spacing 800-1000 ft or 3-4 blocks apart.

    2. You are obfuscating the issue by introducing discussions about the hypotenuse of turns and APC counts into your arguments. These are red herrings and have nothing to do with the excess stopping of the entire Muni system.

    3. For an example of a line that would benefit from consolidation, look no further than the top of the page to the map of the 1 California. Spacing stops every 4 blocks except on the hilliest part of the line would be a tremendous improvement. Let’s say that stop consolidation improved performance by 30%. This would be mean that it take 32 minutes to complete the route instead of 45.

    4. ADA complaints in this case are without merit. You merely have to show that in other cities other than SF bus stops are spaced further apart and that Muni is moving toward this norm. Furthermore, you are presenting a false choice between service for those with mobility impairments and increased stop spacing. They are related but not as you have asserted. By increasing spacing, fewer resources are required to provide a given frequency of service. Extra resources can then be used to provide specialized transportation for those with mobility impairments, either permanent or temporary. By optimizing the system for those with the lowest mobility, the resulting inefficiency ensures that we will not be able to properly take care of those with mobility impairments. This is a point theo made and it is a good one.

    5. Remember that spacing stops 1200 ft apart means the MAXIMUM distance between a stop and any point along the route is 600ft which is a 2 minute walk for the average person. Spacing stops 1 per block (300 ft) means a MAXIMUM distance to any point along the line of .5 blocks or 150 or a 30 second walk. Spacing the stops so close together is what is crippling Muni financially and contributes to the biggest customer complaint which is the slowness of the system.

    We are at a crossroads here. We can have an inefficient system that stops every block with continuing fare hikes and service cuts leading to more car driving, or we can have a system that is faster, more reliable and operates for longer hours allowing more people to effectively use the system. Implementing stop reduction costs very little and provides the biggest potential service improvement.

    Consolidating stops is the best alternative for stopping Muni’s downward spiral and ensuring that other livable streets goals are viable. It is unreasonable to ask someone who has the option to drive to take Muni given its current speed and reliability. Until we have a reasonable alternative to driving in place, removing parking for bike lanes and expanded sidewalks will be much more difficult.

    We can only hope that the BOS rejects the Muni budget tomorrow as proposed by David Chiu. This will give the MTA time an opportunity to propose a solution that moves our transit system forward, not backward.

  • @jim

    1. EVERY time that stop spacing is mentioned, as in the featured quote on this site right now, the endpoint extrema case of two stops per block is used reductio ad absurdum. The MTA does this in the TEP and advocates do as well. That is a propaganda tool designed to coerce changes based on a nonrepresentative sample.

    2. When I present data, I point out the known “errors” in those data as a courtesy to those using those data. “As the crow flies” stop distances mean that, nothing more, nothing less. The APC data is an example of how the MTA projects its desires onto data when interpreting it and is used to demonstrate pitfalls and error in analytical methodology.

    3. SHOW ME WHERE YOU WOULD PUT THE STOPS AND WHAT THE SPACING WOULD LOOK LIKE OR ELSE YOU ARE TALKING OUT YOUR ASS.

    4. The courts will adjudicate ADA matters.

    5. Do you see much blue and green on my maps outside of hilly areas?

    6. Show me the numbers, show me the proposals, show me the savings on a real line with real stop reconfigurations and don’t forget to include longer dwell times based on more de/boardings per fewer stops. Merely repeating the waving of the hands is a good exercise in kinesiology, as good as walking 1/4 mile for the bus if you do it enough, but is not a good exercise in making a valid, substantiated case for a course of action.

    I am not opposed to some stop consolidation, but based on existing standards and existing spacing, I don’t see the savings that others claim to see.

    -marc

  • “We can only hope that the BOS rejects the Muni budget tomorrow as proposed by David Chiu. This will give the MTA time an opportunity to propose a solution that moves our transit system forward, not backward.”

    Here we agree.

    -marc

  • Eloise

    I don’t understand why Muni would place a stop on the far side of an intersection. For example, Southbound at 29th and Mission, many buses, especially during prime hours, have to stop at the light before crossing 29th. The light changes a minute or so later, and then they stop again at the bus stop just South of the intersection. Why isn’t the bus zone on the North side of the intersection? I think this could eliminate many delays.

  • “This is not about screwing the minority, rather about privileging better off commuters who are deemed as ‘transit choice’ riders by screwing the generally poorer and older transit dependent populations who depend on the system to get around the City for their daily routine.” -marcos

    I completely agree. I also sense people use “transit choice rider” to mean “transit choice commuter.” Let’s none of us forget that “choice” riders make decisions about what mode to take to the grocery store or their friend’s house too, and the idea of walking (or waiting! not disagreeing that that point!) an extra block or couple of blocks often makes them take the car. Engineering transit systems to only serve commuters makes for transit systems that only carry commuters.

    “I have traveled to or lived in many cities, and not a single one has so many bus stops. SF has gone too far in the direction of promoting universal access to mainstream transit service, to the detriment of the entire transit system.” -theo

    Honest question, do any of those cities have the same level of bus ridership as Muni does?

    In your second sentence I read, ‘SF promotes universal access to functional transit to the detriment of one specific niche of transit riders,’ that niche may be a large one, but I don’t think that on balance it worsens Muni as a whole. Frankly I’d argue that sweeping generalities of its poor performance hurt the brand more than its actual performance.

    “I get the impression that you don’t think commuting is a valid use of public transit and that “better off” commuters should all buy cars, with the obvious negative externalities.” -theo

    Marcos addressed this but I just want to clarify that reducing exaltation of commuters is not the same thing as condemning them.

    “Consolidating stops is the best alternative for stopping Muni’s downward spiral and ensuring that other livable streets goals are viable. It is unreasonable to ask someone who has the option to drive to take Muni given its current speed and reliability.” -jim

    I suspect both of our hearts and even ideologies are in the same place, but I think the presumption that Muni’s responsibility is to reduce service every chance it gets is what is responsible for its “downward spiral.” I’d reiterate that just as it’s unreasonable to ask someone who has the option to drive to endure a long journey, it’s unreasonable to expect them to take the bus – no matter how fast – if they have to walk many blocks farther to the stop than to their car.

  • @john, the reason why there is no 22L, and I amongst other suggested it at the TEP, was for basically the same reasons why stop consolidation is considered low hanging fruit.

    Fillmore is too narrow for express coaches to pass locals. One solution would be to have the express service run parallel to Fillmore between express stops. Imagine express service that did “figure esses” around Fillmore, running a block over and picking up 22 passengers as it crossed Fillmore to resume its express service by taking a right on the next street. This would place transit vehicles on quiet neighborhood streets through some of the priciest real estate in the toniest neighborhoods in the City, so, naturally, this was laughed at and dismissed out of hand.

    Compare this to the interests of low income seniors and disabled in the Tenderloin, whose stops are disposable to facilitate the convenience of commuters and I think the picture becomes clear as to the priorities and values at play here.

    -marc

  • n

    Eloise: This is either sloppy placement or for a future time at which Muni will be getting signal premption for transit vehicles. The system (already in place for the T-Third) detects a vehicle approaching and keeps the light green until it passes through.

    To add some scientific clarity to the discussion, cost savings can only happen for a route only if enough time was saved to eliminate one run and maintain the same headway. On short routes, eliminating one stop would not likely allow Muni to cut a run. You just give your drivers more time to nap, but you don’t save any money. However, the longest routes and the routes with highest frequencies would be good candidates. The golden formula is something like this:

    Time savings needed to eliminate a run = (1/ # vehicles needed to staff a route) X # minutes per round trip

    For example, it takes 18 Muni vehicles to staff the #24 at the peak hours and 111 minutes to make a round trip (this assumes 10% break time for drivers). So if Muni can cut 6.16 minutes off of a #24 round trip, it can eliminate a run during peak hour. This equates to a rough cost savings of $223,000 per year for just eliminating a run during the peak hours.

  • theo

    Imagine express service that did “figure esses” around Fillmore, running a block over and picking up 22 passengers as it crossed Fillmore to resume its express service by taking a right on the next street.

    Marcos, this 22’S’ suggestion is bizarre, and I don’t even know where to start. Bus turns are quite slow, especially left turns, for which buses usually have to wait for all traffic to clear the intersection. Also, none of the streets parallel to Fillmore are optimized for fast traffic flow — Webster has a divided section, but only between Hayes and Geary. North and south of that, it’s almost as slow as Fillmore. Steiner is slow the whole way. I question whether it’s even possible to run your proposed route as fast as the normal 22.

    A 24L/22, running along Divisidero at high speed then extending into the Marina along the route of the 22, might conceivably draw enough people from the 22 corridor to speed up trips on both lines. But this would require service expansion and money that’s not available.

  • theo

    The presumption that Muni’s responsibility is to reduce service every chance it gets is what is responsible for its “downward spiral.”

    Josh, the whole point of this stop consolidation discussion is that it’s not service reduction. For most passengers, it works out to a service enhancement (faster trips), at the same time it saves Muni money. If it comes down to consolidation vs. actual cuts in service frequency or to lines, and it may, I’m going to choose consolidation every time.

    Frankly I’d argue that sweeping generalities of its poor performance hurt the brand more than its actual performance.

    That’s concern trolling.

    I think most San Franciscans have a reasonably accurate impression of Muni. We know, for instance, that it’s not reliably hitting on-time mandates — reducing dwell time is necessary to fix this, and stop consolidation could really help, as will Translink.

    Many of us also know there are dependable lines like the 1-California and N-Judah, and basket cases like the J-Church and the 22-Fillmore. Stop spacing has a role there as well.

  • theo

    “I have traveled to or lived in many cities, and not a single one has so many bus stops. SF has gone too far in the direction of promoting universal access to mainstream transit service, to the detriment of the entire transit system.” -theo

    Honest question, do any of those cities have the same level of bus ridership as Muni does?

    Muni really isn’t that big of a bus system.

    Chicago CTA has 3x as many riders as Muni. Average bus speed is 9 mph, compared to our 8. Stop spacing guideline is also 1/8 mile.

    London is 20x larger, with average bus speed 10 mph despite not running on a street grid. Bus spacing is more like 1/5 mile.

    We really have to look towards Europe here — most US cities use a 1/8 mile spacing, and while it can work better than SF (as it does in Chicago), London should really be the model. They’re running buses 25% faster through a city 3x denser. Clearly, we’re doing it wrong.

  • jim

    @marcos: “Show me the numbers, show me the proposals, show me the savings on a real line with real stop reconfigurations and don’t forget to include longer dwell times based on more de/boardings per fewer stops.”

    The article indicates a real line, the 14, with real stops and estimates the savings. Mayor Newsom, Nat Ford and others seem to agree that the bus stop spacing in SF is a problem. The burden of proof to refute the claim in the article is on you, not me. Perhaps you can find a transit expert that says we should keep our existing spacing and maybe add more bus stops to further minimize the amount of walking done. Show me the numbers, marcos.

  • DaveO

    @theo But Chicago and London also have fairly thorough and extensive grade separated rapid rail systems, something San Francisco lacks. once San Francisco has a subway system all the way out to Ocean Beach , I’ll be happy to support a local bus system that stops every block.

  • @jim, the 14 line is slated under the TEP for two types of limited service and has a high speed rail line below or adjacent to most of it. why would one want to eliminate local stops on the densest neighborhood commercial district in the City if the line were already covered by the 14L, 49L and BART? Is the 14 line representative of most lines in the system? If so, great, make that case. If not, then we can’t generalize from this case and we should look at a more representative case.

    @theo, the idea of a 22L is a good one because there is a need for high speed service on that corridor similar to the 33 AS IF IT WERE A RADIAL LINE SERVING COMMUTERS! Imagine equity for non-commuters. The idea crashes into the world of the sub optimal when it runs into the reality of the configuration of Fillmore and adjacent streets. When staff nixed the 22L for those reasons, it makes sense to look at less desirable options. When compared to the burden of stopping every block to pick up a passenger, I’d imagine that turns in the “white people’s segment of the 22” would amount to less than the added dwell time we can predict if there were fewer stops.

    @n From what it looks like on the 24, eyeballing it, it seems that you’d need to get rid of 13 stops to get that 6.16 minutes at 30 sec dwell time per stop. WIth 66 stops on the 24, that means cutting about 20% of stops. Of course, you can’t just cut every fifth stop, as that would create irregular gaps, so you’d need to reconfigure all of the stops to ensure even spacing.

    Funny how ideas which sound like wonderful in theory become only viable when they are at best sub optimal as they run into the realities of the built environment and social and political pressures. What, there is no free lunch?

    @theo, London also has a real subway system which carries a great deal of the burden, at least in more affluent North London where the vast bulk of the tube lines are. If there were no reliable rapid transit in North London, then I’d imagine that we’d be seeing more cars clogging London’s twisty streets–look right. Tube starved South London, where the bulk of the bus and heavy rail lines are, is somewhat less dense and congested than the North. Compare North Beach to the Sunset. Mind the gap.

    Good page:

    http://www.london.gov.uk/mayor/transport/facts-and-figures.jsp

    What concerns me about the discussion of stop spacing is the fast and looseness with with terms are being bandied about.

    If you want to revisit stop spacing standards, then that is a conversation. If you want to bring existing lines into strict conformance with existing stop spacing standards, that is another conversation. But this sleight of hand discussion about “blocks” as if that were an easily convertible term is a distraction, as is the incessant reversion to the reductio ad absurdum of two stops per block–note that as soon as I mentioned that, the pull quote was changed on this site!

    -marc

  • bikerider

    This discussion is analogous to what went on in the 1970s with commuter rail systems in the UK and Germany. Large numbers of “underutilized” neighborhood stations were eliminated on the theory this would both save costs and speed up train travel speeds. The end result was disasterous decline in transit ridership. Turns out, travel time to and from the transit stop is just as important to customers as travel time on the train itself. In the case of Germany, those stops have now been put back in, and became a textbook case of how not to run transit service.

    The other thing German rail managers learned how to do was to greatly reduce the dwell time at the minor stops. POP ticketing, level-platform boarding, and high-performance braking/acceleration all helped to keep delay to a minimum — thus they could continue serving the minor stops within same timetable as a skip-stop service running more antiquated trainsets.

    When Muni was folded into the MTA, the argument was that traffic management and bus service were tightly connected, and that an MTA could make the necessary street lane changes to ensure a quality and reliable bus system. That seems to have been forgotten, and there has been no progress on putting in more exclusive bus lanes, boarding areas, etc. Instead, they are going after the minor bus stops, which might be easier politically, but doesn’t really address the root cause: too many cars on the streets are slowing down buses.

  • jdub

    If you look at Marcos map (these are great, btw) for the 1 California outbound at http://cybre.net/pub/muni/stops/001.OUT.html, it shows great variability in bus stop spacing with some 1200 ft stops adjacent to 700 ft stops in the western part of town where it is largely flat. I’m wondering:

    1. Why is there so much variability?
    2. Is it possible that the idea of bus stop removal is worse than the actual consequences of bus stop removal? Otherwise, why would we tolerate a 1200 ft distance between stops in some areas and not others? It seems that if a there is a large distance between stops, we do not see intense pressure to close the gap by adding a stop. However, removing a stop even where there a short distance to the next stop is very difficult.

    We agree that removing bus stops can save money and increase speed if applied properly as n states above. We need to apply stop reduction to those lines where it makes a difference. Doing so could:

    1. save money by not having the extra driver and bus thereby reducing the budget deficit
    2. increase frequency on the line in order to deliver better service
    3. allow the extra driver to service those with mobility impairments. This may be better done by better utilizing the taxi system which I understand is under consideration to be put under the MTA umbrella.

    Bus stop spacing is a tricky business. On the 1 California outbound, there should probably be stops at Webster because there is a hospital there and at Fillmore because it is a major shopping destination. These are adjacent blocks but it still makes sense. On the other hand, the 1 bus stops at both Sansome and Montgomery and many other adjacent streets where the distances between stops is relatively short. Is this line a candidate for analysis to see if we can consolidate enough stops to elminate a run per h’s formula?

    From the article above:

    “Without a vocal and organized constituency, however, voicing support for service improvements and time savings through stop removal, the issue will not be popular politically.” We all want a better Muni system and intelligent stop removal is one of many things that can be done to improve things. Given this fiscal emergency, perhaps this time we will be able to make the politically difficult choices.

  • theo

    @DaveO

    The average speed statistics I gave were only for the bus systems in SF, Chicago and London.

    That doesn’t include Muni Metro, the ‘L’, and the Tube. In all of those cities the bus and light/heavy rail systems carry about the same number of total daily passengers. There may be other reasons why the comparison isn’t valid, but the presence or absence of light rail isn’t one.

  • @jdub, thanks.

    Here is the source code for the app that reads a postgis database of the Muni stop GIS data and uses perl mapscript API to mapserver to generate the map images. Generic city GIS files are available off of sfgov.org :

    http://cybre.net/pub/muni/stops/stops.pl.txt

    I didn’t have the Muni line GIS data, only the stops as points, so I couldn’t follow the line route to give accurate distances between stops when the bus turns, hence the hypotenuse versus sum of both legs presentation defect.

    I could generate tabular data in csv format that shows stops as endpoints and the length of the segment if folks are interested.

    But here is the methodology so that others might test it for accuracy if so inclined. If you use this code for money send me some.

    -marc

  • theo

    @marcos,

    My point is that the London bus system is surprisingly huge, extensive, and faster than ours. The tube is important, but not much more so than Muni Metro/BART within the city.

    Buses are for locals not tourists, so visitors don’t always realize how good they are. (The system’s also improved a lot in the last few years, in part because the congestion charge got traffic off the streets of central London.)

    Nearly every borough in London is equally dense or denser than everywhere in SF except Chinatown. Their system succeeds despite the density.

    @bikerider,

    Comparing limited stop commuter rail systems to local buses is a badly flawed analogy. Cutting commuter rail stops that are 10 minutes’ walk from each other is not the same as cutting bus stops that are 1-2 minutes from each other. The success of European bus systems disproves your implication that significant numbers of people are going to not take the bus because they have to walk 1/2 or 1 extra block.

  • DaveO

    @theo

    I find that hard to believe, that Bus vs. Tube; Bus vs. L; Bus vs. MuniMetro is the same for London, CHicago and SF.

    quickly googling over the internets, I come across figures that put London’s bus to tube ridership ratio at about 2:1; Chicago’s at 3:1; and SF’s at 4:1. Of course, I’m not including intra-SF Bart, but then again the MuniMetro isn’t really a rapid rail network in the same sense as the Tube or the L. If you have other data, I’d like to see it.

    But your general point remains valid; even though bus ridership is a less critical component to London’s overall transit network than it is to SF’s, it still operates more efficiently there than it does here, and we should strive to at least equal that.

  • “This discussion is analogous to what went on in the 1970s with commuter rail systems in the UK and Germany. Large numbers of “underutilized” neighborhood stations were eliminated on the theory this would both save costs and speed up train travel speeds. The end result was disasterous decline in transit ridership.”

    Then it is also analogous to what went on 4-6 years ago with commuter rail systems in a galaxy even farther away than the UK called “San Francisco/San Mateo/Santa Clara County”. On Caltrain, 3 stops were completely eliminated and multiple stops had service greatly reduced. The end result was record ridership levels.

  • I hope you all are sharing your ideas and solution with the SFMTA. You can contact the TEP group at tep@sfmta.com.

  • “the whole point of this stop consolidation discussion is that it’s not service reduction. For most passengers, it works out to a service enhancement (faster trips), at the same time it saves Muni money.”

    It absolutely is a service reduction, just apparently not for your particular niche of transit use. That doesn’t mean the people that use the current service won’t feel the pinch.

    “Muni really isn’t that big of a bus system. Chicago CTA has 3x as many riders as Muni…”

    I guess we need to define ‘big.’ My question was about bus ridership so I’ll assume that’s what you meant. Chicago has 3x the bus ridership of Muni, in a city over ten times the size. SF is a small city and Muni is a correspondingly small system, but an awful lot of San Franciscans take Muni because it’s just plain possible to do.

    Too be honest, I think bikerider gets it right. We can have this philosophical argument over who Muni should serve and what segment of its ridership should have its resources shifted to other segments, but even removing every stop you wanted to wouldn’t make the difference some proponents claim. Buses aren’t slowed that much by stopping for passengers compared to the delays caused by cars and stoplights.

    “the root cause: too many cars on the streets are slowing down buses.”

  • The difference is that London follows through on a full spectrum commitment to multi modal public transport and San Francisco is engineered from the ground up for cars.

    It is dangerous to build public support in favor of stop consolidation without conditions and public pressure to demand those conditions, because based on the lack of accountability, giving this MTA an inch means they take a foot.

    And absent any concrete proposals for stop consolidation on any given line with modelling that predicts savings, this is all hot air enabling a dysfunctional bureaucracy that has repeatedly done the wrong thing.

    Show me how you will take those 500, 600,700 and 800s and translate them into 900, 1000, 1100 or 1200s without missing a stop at a community facility or a transit transfer.

    -marc

  • bikerider

    John: the Baby Bullet is not analogous. In that project, $100 million capital investment was used to purchase additional trainsets, and install 3rd track to overlay new express service over existing locals. Elimination of the stop at Atherton is not the reason for the “Record” ridership.

    (The term “Record” of course being relative, given Caltrain’s pathetic mode share.)

    And those who use the example of London Transport buses are cherry-picking the data. The huge speedup in bus service is a direct consequence of the Congestion Charging — which significantly sped up all road traffic by reducing the number of cars on the road. As well, London replaced their fleet with modern European equipment, with boarding and alighting through all doors, plus seating configuration more conducive to standees and passenger flow in the bus.

    Yes, London does also have longer spacing between stops, and yes, stop consolidation can show benefits. But stop elimination should come after doing all the other Best Practices.

    Incidentally, London contracts out bus routes to private firms through competitive bid. So if you __really__ want to reduce Muni overhead without cutting routes…..

  • “Incidentally, London contracts out bus routes to private firms through competitive bid. So if you __really__ want to reduce Muni overhead without cutting routes…..”

    KABOOM!

    Translation: you can screw transit dependent and working San Franciscans with crappy service and high fares, but you can’t screw organized labor. Solidarity forever!

    -marc

  • theo

    The huge speedup in bus service is a direct consequence of the Congestion Charging — which significantly sped up all road traffic by reducing the number of cars on the road. As well, London replaced their fleet with modern European equipment, with boarding and alighting through all doors, plus seating configuration more conducive to standees and passenger flow in the bus. Yes, London does also have longer spacing between stops, and yes, stop consolidation can show benefits. But stop elimination should come after doing all the other Best Practices.

    Yes, I mentioned the Congestion Charge fairly prominently when discussing the London example.

    I’m not sure if you realize there is no money for the other Best Practices. We can’t replace the fleet, or the seating, or implement congestion charging. Stop consolidation is the one thing Muni can do to preserve or improve service while being forced to make budget cuts.

ALSO ON STREETSBLOG

Bus Stop Consolidation: The Times Have Changed

|
Flickr photo: erik kuo Does the 14-Mission really need to stop at every block on Mission Street? Does the 21-Hayes? Consolidating bus stops could speed transit vehicles and reduce dwell time, saving service hours that could be used to increase frequencies and add hours of operation. Yet the MTA has avoided the topic for years, […]