If One Bus Is Good, Seven Must Be Better

Recently, I received a note from a reader about bunching on the 29 Dufferin bus.  This is one of two routes on which increased supervision has led, or so the TTC claims, to improved service (as discussed in my recent article about the Customer Service plan).

I’m wondering what TTC is doing to prevent bus bunching as this seems to be a big problem on many routes…

I wrote to TTC customer service but of course I didn’t hear back from them. I would be interested to know what measures are in place to space out the buses. It seems to me that buses are just rushing to get to the end of the line; they basically do not care if they have 2-3 buses traveling together or if the last buses have no passengers.

I took this video as an example, there are 7 buses passing Bloor stop in under 3 minutes … such a waste from TTC. Just imagine being the “lucky” one that just missed all 7 buses. How long they will need to wait for another bus?

The TTC talks a good line about service management, but a casual look at any of the routes where real-time monitoring is possible (through the open data interface to GPS vehicle data) routinely shows bunching even at 7am when there cannot possibly be “traffic congestion” effects.  Two basic questions about bunching emerge from all of the service reviews I have done:

  • Why are vehicles allowed to leave termini very close together rather than regularly spaced?
  • Why are vehicles entering service from yards or from short-turns not spaced between through runs so that even headways are provided?

This does not require millions in high technology to implement, only the will to manage the service, something the TTC once did regularly with no more than route Inspectors on the street.  With the tools now available for vehicle tracking, it should be much easier as all vehicle locations are available online.

I plan to review the Dufferin bus operations again in the Spring (my last review was published in January 2008) once the effects of winter weather are not a concern.

Short turns disrupt service near the ends of routes, and on Dufferin this can show up as poor service to Liberty Village thanks to turnbacks southbound at Dundas.  On King, short turns eastbound at Parliament and westbound at Roncesvalles deprive riders on Broadview and on Roncesvalles (which have demands to some degree independent of the route downtown) of service.

If the TTC is going to strive for having 99% of the service advertised actually on the street, it must also strive to have that service on the route, not sitting in a short turn, or running in a pack followed by a large gap.

31 thoughts on “If One Bus Is Good, Seven Must Be Better

  1. Of course, the TTC could always do what OCTranspo did a couple of weeks ago: block the ability for personal devices to track GPS information. That solves a lot of problems — for them.

    Steve: Difficult though it may be, a skillful use of any sort of real time next vehicle arrival system will reveal the approximate location of any transit vehicles. The stupidity in this is that anyone who is really dedicated can find this information will a little effort, and blocking access is simply a collective ass-covering exercise by transit operators and management who don’t want the public to know what is actually going on.


  2. I would say if there was a reason for drivers to use their cellphones while driving this is as good as any … rather than hiring new route supervisors they should be hiring a developer or two to write an easy to use cellphone app to tell drivers to slow down.

    Steve: The TTC’s vehicle monitoring system CIS already has the ability to send text messages to operators, although these are manually generated. There was also a scheme to include real-time map displays of route status on vehicles, but that won’t show up until, at best, a refresh of the aged CIS hardware, a project currently in the deep freeze thanks to budget cuts.

    There should definitely be a public roadmap to move to a more automated system, starting with the right-of-way streetcar routes – since they should be the easiest to implement it on. There should also be publicly available data sheets on “% of service that is evenly spaced” for each line, and analysis of what the choke points are across the city.

    There should also be an analysis of end-points of the system – there is no reason why this can’t be controlled to the second – and with the GPS system it can be managed to ensure drivers are following the rules. A report on “% of vehicles leaving end-points within 10 seconds of ideal” would help to give us this information.

    Steve: The TTC was supposed to make archival CIS data available by the end of 2010, but like so many initiatives, it seems to be buried in their focus on the budget and, possibly, suspended because of the change in the Mayor’s office. It would be delightful (as I am sure you would agree) to be able to download info for all routes on a daily basis and produce these sort of stats without having to ask the TTC and be told they don’t have any staff free to do this sort of work.

    If there are any drivers out there: how do you currently figure out when to leave an end-point, especially if there is no route managers in the area? What could be done to help you do your job better? If an app was made that told you when to leave (even if it wasn’t made by the TTC) would you use it? Could you – given the current firings of drivers who used cellphones?


  3. I agree that, normally, buses should be made to leave the termini regularly spaced. Sometimes, though, they should maybe run in a team if there has been a long gap in service and there are more people waiting along the route than one bus could carry. This could also be handled by short turning before the terminus but this may not be practical if the buses all have lots of passengers.

    Steve: A big problem with buses running in packs is that there are cases where skip-stop operation would serve everyone better, but current operating practices forbid it because it might cause complaints. The TTC trades one type of customer inconvenience for another on this one. There are also times where express operation could be used to close up gaps, but I have rarely seen this done, and not at all recently.


  4. Steve,

    I use the southbound 29 regularly on weekday mornings. I can assure you that even at 7:30-7:45am, traffic congestion is not an issue, but there is significant bunching (buses almost always come in pairs or three-somes).

    My destination every morning is the Exhibition GO on the Lakeshore West line, which is served only by the 29D route variant (which are about a third or a fourth of all southbound 29 vehicles, in my experience). Skip-stop operation actually hurts me here, because if a 29D passes me by in favour of a plain-vanilla 29 just behind it, I’m left hoping to “catch up” with the 29D — if that fails, I walk (more often, run) the ~1 km from the end of the 29’s route to Exhibition GO.

    Short-turning the 29 or 29D hasn’t been a big issue for me — what are more significant are the suspensions of the 29D branch that last weeks at a time. It seems that anytime there is anything going on at the Exhibition grounds, they turn all the buses back at Dufferin Gate. So from June to Sept 2010, between all the events going on (G-20, the CHIN picnic, the Ex, etc.) the 29D was mostly non-existent (it would make a brief appearance for a few days and then disappear for a few weeks). This wouldn’t be as much of an issue if there had been advance notice, but there was none — no mention of the route suspensions on the TTC website or 393-INFO, no printed notices on bus stops, and TTC trip planner still recommending the 29D as the best way to get to Exhibition GO. When the 29D finally re-appeared in mid-September it was a welcome surprise!

    I actually contacted the TTC complaints line about this. The folks there were friendly, but basically said “keep complaining so that management is encouraged to work this out with the CNE.”

    A bit of a rant but I think this illustrates both the specific issues of Route 29 as well as communication breakdowns and “customer service” issues with the TTC.


  5. Why do we need “route inspectors”? Don’t TTC buses have radios? If they do, then surely drivers should be simply contacted directly when someone monitoring the GPS data notices a problem.

    Steve: In theory, yes. In practice, the central management didn’t work out too well because the folks looking at the computer screens didn’t have a good feel for what was happening on the street. Yes, they might have done a better job, but at the point the decision to move back to on street supervision was taken, the monitoring system still didn’t use GPS, and the info it displayed was generally a mess especially when service was badly disrupted with short turns and diversions.

    I had to deal with this in attempting to analyze tracking data that had all sorts of spurious info that I had to filter out to get reasonable results. A common problem would be that a car would be reported as westbound in Parkdale when it was actually eastbound downtown. Not the greatest of monitoring tools, and it took the TTC ages to get around to fixing this, long after many other cities were using GPS.


  6. I agree with Jordan that it is important to take into account the route variants when using skip-stop (and short turning) so that people aren’t skipped by the variant they are waiting for.


  7. Well, the 7 buses in a row is more proof of the incompetence of TTC management. Notice too how many drivers are too lazy to pull into the curb properly leaving the rear doors away from the sidewalk. No wonder people do not want to use the rear doors.


  8. Steve writes: “A big problem with buses running in packs is that there are cases where skip-stop operation would serve everyone better, but current operating practices forbid it because it might cause complaints. The TTC trades one type of customer inconvenience for another on this one. There are also times where express operation could be used to close up gaps, but I have rarely seen this done, and not at all recently.”

    I’ve been on Finch East buses that have traded off stops between them. And recently I’ve been on Queen cars, both in the morning and evening, which bypassed stops where people were waiting because the car was jam-packed. (And by jam-packed, I mean people doing gymnastics using the rear stairwell stanchions to levitate above the treadle steps.)

    Steve: I am making a distinction here between passing stops because nobody can get on, and pairs of vehicles alternating which stops they serve in order to spread out the load and reduce their running times.


  9. Living in London, there are a couple of obvious differences that seem to resolve most of the problem of bunching (although it does exist in some routes):

    1) Many buses run on common paths for some portion of their route but not the entire route. This allows many routes to cover higher demand areas of each route by overlapping and then spreading out during lower demand areas. As a result, buses do bunch but many users at a stop many opt for different buses which automatically distributes the load amongst several buses.

    2) Buses are flagged by the users at the stop based on demand. If a bus is driving past a stop, it will not try to pick up passengers unless explicitly flagged down which means that routes which are bunching will automatically rotate positions of the lead bus to distribute the load.

    3) During high traffic periods, routes with a larger number of buses on service are based on a frequency requirement (i.e. a bus will be at this stop every xxx minutes) whereas for a smaller number of buses on service it is based on timing (i.e. 9:20, 9:35, 9:50, etc). I presume that drivers who are on heavy demand routes then have the ability to control the bunching at the route terminus whereas low demand routes can control the bunching by pacing themselves at each stop (assuming sufficient lag is built into the route).

    Steve: Toronto’s network is a grid, and there are relatively few locations where many routes serve the same corridor. These tend to be pinch points caused by geography, or approaches to major terminals where many buses connect to the subway. There are large parts of the network served by only one route. As for controlling bunching at terminals, the TTC has a big problem with buses (and streetcars) leaving terminals in pairs or even trios even though the headway on the schedule may be four minutes or more. I leave it to your imagination why this happens and why it is not corrected.


  10. The usual complaint I hear about the TTC follows a standard phrase:

    “There never is a bus and when it does come, it’s full and there are 6 others behind it.”

    The longest lineup I have ever seen was 12 on Finch East at Bayview about 5 years ago. A truly awesome sight to behold coming over the hill towards us. Cars having to snake past them — expresses jostling with non-expresses to see who was going to stop where. Almost made me forget the 12 minutes I waited for them to show up (others had waited longer).

    Whenever I see a bunch up of buses, roaming like they are hunting in packs, I get that now familiar feeling of both anger and resentment as I know somewhere around me, one more person is getting that much more frustrated and taking that little farther step to taking a car and/or hating the TTC and transit in general. Yes, the subway stations look like the longest line of dirty bathroom tile in the world. But, the real killer of transit support in the inner suburbs is those bunches of buses. And I am convinced they are one reason why Ford did so well out here in that election among apartment dwellers. Heck, all those buses together even look like a train. Just paint them all brown and gravy like and you could have handed out Ford buttons easy.

    Steve: And don’t forget how flexible buses are that they, unlike streetcars, can go around obstacles and each other.


  11. I’ve always thought the TTC should install traffic signals at route terminals, timed to only allow a bus/streetcar to leave after a certain time, say 75% of the planned headway, since the previous vehicle departed on that particular route. It wouldn’t solve all the problems, but spacing out vehicles properly at the terminal is the low-hanging fruit.

    The biggest challenge is giving operators the information they need (either at terminals, or on-board) to decide how long to stay at the terminal to maintain the headway, assuming they care about providing a good service.


  12. Steve said, “the central management didn’t work out too well because the folks looking at the computer screens didn’t have a good feel for what was happening on the street.”

    Part of that problem, according to a TTC operator I know, is that these folks were somewhat young and inexperienced with the streets. Promotion to inspector positions is based on seniority, but when these moved to computer screens at desks, many of the senior operators used their right of first refusal. This meant that the people in those positions often did not know much more about the territory that what a Google map could show them.

    This leads to a Lawrence East operator being told to short turn at Manse Road and use Coronation and Homestead to get back to Lawrence to head west. Looks fine on a Google map, but misses the point that going the other way around the loop gives one a traffic light to turn left onto Lawrence.

    Having route inspectors on the street is a step in the right direction.


  13. Here is a KISS solution to the 7 buses in a row nonsense. Issue instructions to all operators that no bus is to leave end points or timing point ahead of schedule (unless instructed otherwise). Make Bloor Street a timing point in both directions. Monitor the situation constantly. If after say 30 days the situation does not smooth out, go to step two. Remove one of those 7 buses and establish a management “starter” located at Bloor Street. This supervisor will control buses leaving Bloor. That one bus will never be missed but the improved flow of buses will be noticeable and service improved at NO additional cost. Step 3 expand this to other routes and keep doing it until things run better everywhere. Since the Operators have proven they cannot be trusted to do what is right, remove their jobs and replace them with supervisors. This is good management in any event since that is what the TTC should be doing, Manage.


  14. The TTC used to run articulated buses on this route in the 90s with the Orion 03.501. The 29 Dufferin bus is an extremely heavily travelled route and running these buses is the solution especially during school hours.


  15. The Scottish radio show sometimes plays a song that is (something like):

    “There were 10 wee wifeys waiting in the queue. (2x)
    And when one wee wifey saw a motorist she knew,
    There were 9 wee wifeys waiting in the queue.”
    down to
    “There were no wee wifeys waiting in the queue. (2x)
    along came twenty buses
    Every one a forty-two!”

    I guess other places experience the same problem.


  16. Maybe if the city looked at proper route management as a way of controlling costs, something might be done. Rob Ford might not care about how long us poor car-less slobs wait for a bus, but if he can run the same service with 5 properly spaced buses instead of 7 buses in a bunch, then then the city can save a lot of $$$$.


  17. Very crappy route management is at the core of TTC’s problems. Here’s an example:

    You’ve got to wonder about quality of service…. I just drove on Bathurst from Davenpoort to Steeles and the first northbound bus was seen north of Drewry. The Bathurst bus is suppose to run every 12-13 minutes, but if there were no buses between Davenport and Finch, how bad is the wait on the route – maybe 30 minutes? I saw 3 southbound buses all north of Lawrence. This leaves only 2 buses unaccounted for, hopefully they were on Bathurst south of Davenport, otherwise good luck to people waiting to go north. I drove southbound about 9 p.m. and the traffic was pretty light, so I don’t understand how the schedules could be so screwed up that they haven’t evened out by 11:30 p.m.

    If the issue is that buses/operators aren’t given enough running time to complete the route, then perhaps the schedules themselves have to be reviewed as this doesn’t seem to be an issue of too few buses.


  18. It’s the bunching of the 320 Yonge night bus into clumps of three, four, sometimes five buses that I don’t understand.

    When I have to do maintenance work on equipment outside the operating day at work, I either work overnight until 3 to 4 AM or work through until the start of the next day’s shift at 7:30 AM (the 7:30 finishes are typically on Sunday mornings), or start at 5 AM, which exposes me to using the Yonge night bus when I finish work and that’s when I’ve observed the bunching and wait times it causes.

    There’s nothing worse than standing around in the middle of winter, waiting to go home after a long night and watching four 320 buses go the other way or have the same thing happen in the opposite direction waiting to get to work. Traffic congestion as an excuse doesn’t stand up when Yonge St’s totally deserted in the middle of the night. The 320 Yonge night bus operation’s obviously not being supervised but clearly needs to be.

    Tonight is one of those nights that I’ll be performing preventative maintenance starting at midnight and based on the work I need to do, I’ll likely be finishing up sometime between 3:00 and 4:00 AM. Maybe I should post my observations when I get home.


  19. “There never is a bus and when it does come, it’s full and there are 6 others behind it.”

    When packs of streetcars arrive at a stop I find it odd that most of the passengers try to stuff themselves into the sardine can lead car. Of course this further delays the lead car, causes passenger discomfort, increases dwell time along the route and exacerbates the gap/pack situation. I don’t see this passenger behavior in other cities – at least not to the extent it occurs in Toronto. It’s the simplest problem for transit managers to fix, yet they do nothing.

    What’s with Toronto passengers and the silly lemming behavior? And why won’t the TTC constructively address this scenario?

    Steve: Passengers cram into the first car because it is usually the one that will not be short turned. If you have already waited 20 minutes for something to show up, the last thing you want is to discover that the 2nd, 3rd, etc cars are not going as far down the line as you want. Boarding one of them almost guarantees that you will be thrown off, and will wait at the short turn point for the next parade of cars 20 minutes behind. People are not being lemmings, they are trying to get home.


  20. If Mike Vainchtein were travelling at close to the same speed as the buses, it’s quite conceivable that he wouldn’t have seen any buses during that drive, no matter how far apart they were. Where should he have seen the first bus if they were running every 12-13 minutes.


  21. Work went longer than I expected and I only got on my way at about a quarter to six, so I went to sleep when I got home instead of posting the results of the trip, which is why I’m writing it up now. Luckily, I didn’t need to wait long for a 320 bus but it was crowded. I don’t know if it was the first one after a gap in service but when I got off it at the end of my trip, there were two more Yonge night buses right behind it forming a group of three. I started walking home from the bus stop and a fourth bus followed the group of three about a minute later.

    Given the specified frequency of the Yonge night bus, there were probably people somewhere along the line that had a lengthy wait for a bus. Traffic wasn’t an issue because Yonge St. was pretty much empty as it typically is at 6 AM on Sunday so I’m still interested in hearing how the TTC explains the bus procession I observed.


  22. Jim Hoffman says:
    If Mike Vainchtein were travelling at close to the same speed as the buses, it’s quite conceivable that he wouldn’t have seen any buses during that drive, no matter how far apart they were. Where should he have seen the first bus if they were running every 12-13 minutes.

    Judging by the schedules I should have seen the first bus at Davenport or perhaps between Davenport and St. Clair. If it was slightly late it would have been between Davenport and the subway. It also seems like the bus I saw at Drewry was roughly on time as well. However, I still should have seen a bus somewhere between St. Clair and Drewry, distance wise probably somewhere between Lawrence and Wilson. For southbound buses the first one I should have seen around St. Clair, second around Lawrence and third probably around Finch. This would have provided even service.


  23. While I am not making excuses for the TTC’s lousy route management, in fairness it is not possible to keep vehicles evenly spaced on roads in mixed traffic. There are just too many variables over which the TTC has no control such as illegally parked cars, some traffic lights with different cycle times from the majority on the road, unusually large number of passengers at a stop etc. That not withstanding there is no excuse for some cars to intentionally leave a terminal early or late. The TTC does know via their CIS system if a vehicle is early or late as I hear route supervisors talking to the divisions on my scanner to find out where cars are and if the are early or late.

    It is impossible to keep vehicles 5 minutes apart if the traffic signals on the line have a 2 minute or 3 minute cycle time. Some cars are going to make a light while others are going to miss it. This will cause one to get closer to the one in front and put the headway to 4 and 6 or 3 and 9 instead of an even 5. The vehicle that is closer to the one in front will pick up fewer passengers and make more lights while the one in front will miss more. Eventually they will be travelling together. If you start holding vehicles at intersections to space them out them the passengers on them will get angry because they are sitting and wasting time.

    Mixed traffic service on long lines such as Dufferin, Lawrence east and west, Eglinton etc. are going to bunch up in rush hour when they are on short head ways. It is important to not over manage the route and make the service to slow to be of value for the sake of keeping the vehicles evenly spaced. However the TTC has to get rid of convoys of 4 or more buses on the same route. The imaginative use of skip stop and temporary express service to get rid of gaps would be useful but that would require a shift in operating strategies.

    The other practice that bugs me occurs when there is a gap on a car line like Spadina and a group of cars finally arrive at the terminal. The TTC usually insists on filling the first car to the doors then sending it out onto a line that has large numbers of passengers waiting at stops with no hope of getting on until the third or fourth car arrives. This just slows down the line more as people try to crowd on and block the doors. It would make more sense and speed up service for everyone if they would send out 1 or 2 cars empty to pick up surface passengers. It would not slow down the passengers in the station and since there would not be throngs at each stop it would probably speed things up. Unfortunately the first supervisor who tried this would probably have many passengers complaining that they had to wait on extra car and upper management would slap his knuckles.

    The best example of line management that I saw was at the Union streetcar platform on a labour day Monday in 2009. The 509 line did not have any extras as no one wanted to do overtime on this beautiful day. The lineups for the street car backed up to the subway platform and the street from people trying to get to the exhibition for the air show. Finally a Toronto police sergeant showed up, quickly figured out what the problem was and made the unloading platform into the loading platform for the 510 car. This separated the two groups of passengers , those who were going along Queens Quay and those who were going to the Ex. He did not let any local passengers onto the 509. The problem cleared up in 15 minutes and he ordered the supervisor to keep the service that way. It is amazing what can be done when you look at a problem from a different view point.


  24. Robert Wightman said:
    “If you start holding vehicles at intersections to space them out them the passengers on them will get angry because they are sitting and wasting time.”

    They’re already doing this. I’ve been in buses where they have done just that and on one occasion, this bus has actually been passed by another bus while holding this way. Yes, it is infuriating. But I guess we’re just “freight that talks back.”

    Steve: This is a situation where the hold of one bus to space service really is “for the greater good”, and the operator could help things by explaining what’s going on.


  25. Steve: This is a situation where the hold of one bus to space service really is “for the greater good”, and the operator could help things by explaining what’s going on.

    I have seen exactly this done on the London Underground at the height of rush hour (conditions that make Yonge and Bloor around 9 am look favorable).


  26. Robert Wightman comments, “The imaginative use of skip stop and temporary express service to get rid of gaps would be useful but that would require a shift in operating strategies.”

    The TTC should add more consideration for riders’ desire to actually get where they’re going. Many of the TTC’s coping strategies work the opposite way, for example by increasing running times to deal with headway issues.

    Having one bus or streetcar pause because there’s a huge pack up ahead is one thing. Having a bus or streetcar pause because the running time is set to allow for the slowest operators on the worst-traffic days is really frustrating when you have a good operator and traffic isn’t so bad. (The streetcar or bus moves as if it’s stuck in a traffic jam, even when it isn’t.)

    Some of the route scheduling and management strategies are orthogonal to service speed, but when choosing between a couple of approaches, the TTC should ask “which will get riders where they’re going quicker?” and weight their choice to the approach that yields a quicker trip for riders.


  27. Re: the previous comment on spacing bus service, it’s also worth mentioning that if buses hold frequently to maintain either schedule or spacing (depending on whether a schedule or headway management is being used), they don’t hold much at once. It’s much better to just hit the brakes on a yellow and leave stops slowly throughout the route rather than trying to set new land-speed records until the supervisor finally orders a 5-minute pause.


  28. Why can’t TTC policy be to wait one light at an interchage intersection? This was the case in London England, and it worked very well to stop bunching. Watching a second bus pull away while you’re running up the steps from a subway station or changing from a streetcar to a bus is infuriating.


  29. Buses and streetcars leave the end of a line @ a scheduled time, unless there is a problem on the line. For example, Dufferin buses run every 2-5min peek time, however this is impossible to maintain. Why? Operator experience is a main factor (rookie vs. veteran), rookies are slower, however a vet can be slow due to being disguntled.

    Line supervisor experience, not knowing their job well. Customer lag time, ie. Taking leisure time to board/exit buses and locate fare, blocking operator view. Traffic, accidents, detours, subway delays and configuration of the city, ie. lack of subways, freeways, sensored [??] traffic signalling all lead to this problem.

    Here’s a perfect example. Allen Rd. ends/begins at Eglinton. During rush hour that area is gridlock as far east as Bathust and was far west as Oakwood. Bus service is affected in the area. To make things worse, side streets have signs that restrict vehicular traffic, forcing vehicles to stay on the main road!!! The whole thing is counter productive!

    Steve: Your first mistake is the claim that vehicles leave the end of the line on time. In fact, they are permitted to be off by three minutes either way and still be “on time”, and variances greater than this are not unknown. This sort of behaviour is easy to see in the vehicle monitoring data from TTC that I have published here many times. It is not unusual for vehicles to run in packs followed by a long gap rather than for Route Supervisors to space them out.

    Yes, there are locations where there is severe traffic congestion, and this really screws up routes. Finch West, for example, had big construction-related delays for the Spadina subway extension. However, these delays did not occur with the same intensity at all times of the day, or all days of the month. The service, especially on the outer end of the line, was still erratic.

    Congestion and other factors beyond operators’ control are part of the problem, but not all of it by any means. By the way, the TTC claims to monitor its service and headway on a relative basis, not to the schedule. In other words, it should not matter whether a bus is “on time” but how close it is to the ones ahead of and behind it. That implies someone is actually spacing service, but I don’t even see attempts to ensure that short-turned vehicles re-enter the line to fill a gap rather than to run 1 minute behind another vehicle.


Comments are closed.