The City of Toronto will hold three public meetings regarding proposed changes in rush hour traffic restrictions on Queen, Dundas and Carlton/College between Parliament and Roncesvalles.
- Wed. June 17 at 6:30 pm: Mary McCormick Community Centre, 66 Sheridan Avenue
- Thurs. June 25 at 6:00 pm: Scadding Court Community Centre, 707 Dundas Street West
- Mon. June 29 at 6:00 pm: City Hall, Committee Room 2
Work underway by City Transportation Services and the TTC includes:
The project team is currently:
- Analyzing GPS-based data for the streetcar routes along Dundas Street, College/Carlton Streets, and Queen Street West
- Looking at how existing peak-hour restrictions function along the routes
- Finding opportunities to improve streetcar travel times and reliability. For example, where streetcar travel times increase sharply after current restrictions end (e.g. 6pm), the project team is considering extending the peak period by a half-hour or hour on each end to improve the flow of transit, general traffic, and cyclists
- In a few cases, where no clear benefit is observed from an existing extension, peak hours are being reduced
Additional measures to improve transit operations, in various stages of implementation, include:
- Re-timing signals to give more green time to the priority direction
- Active Transit Signal Priority (extended green or shortened red when transit vehicles are detected and meet certain requirements)
- All-door boarding / Proof of Payment (POP)
- Greater enforcement of restrictions
[From the “Overview” tab]
Disclosure: I have acted as a consultant to this project in the analysis of TTC vehicle tracking data.
Steve Munro 
Well – re-timing, I assume also means the direction with the greater load of traffic – so well this would be good for all no?
Active Transit Signal Priority – hopefully this will be specifically done so as to co-ordinate with the position of the vehicle relative to where it is supposed to be, ie not provided to those vehicles running early, but consistently provided to those vehicles running into a gap, to at least help them hold if not close the gap.
All-door boarding – should be working on this everywhere.
Greater enforcement of restrictions – yeah – hurrah, will they really succeed. This to me has been an issue forever.
LikeLike
It is great that the City & the TTC are (finally) looking at using actual data to manage, and *gasp* actually try to improve streetcar service.
LikeLike
This is great news, when I ride TTC streetcars (almost exclusively), it drives me mad to see a a line of cars turning left allowed to block literally hundreds of transit users on a streetcar. Steve, what metrics does the city use to determine what left turn restrictions should be in place? I can’t imagine that if all road users time is considered equal, that a streetcar should ever be blocked by a few (usually single occupancy) vehicles. How does the city justify allowing a few cars to block and slow hundreds of streetcar riders?
Also, on a related topic, why is the King street transit lane un-enforced? From my experience the King car is among the slowest routes in the city, yet much of it’s length is a transit lane full of (again mostly single occupancy) cars.
Love your blog!
Steve: The metric varies by location because in some cases, banning a left turn would simply shift the demand somewhere else. Equally, adding an advanced green phase can encourage left turns and be counterproductive. And so the answer is “it depends”.
King Street was not enforced from the day the signs went up because, frankly, the “reserved lane” never made sense in light of actual demand in some sections and what was allowed to continue in the curb lanes.
LikeLike
Interesting. One of the worst congestion spots on the 506 westbound in AM peak is approaching Parliament on Gerrard (with the Parliament/Gerrard intersection being the primary issue). Hopefully they expand the study by a couple of blocks.
Steve: Jarvis was originally the east limit of the study, and then it was expanded to Parliament. Obviously the data exist for other sections of the routes.
LikeLike
The only requirement I can see is that the vehicle is in service. Streetcars and buses should receive a green light to keep them moving.
Steve: Note that the text cited above is from the City’s website. I don’t agree with the approach they and the TTC seem to be heading toward.
LikeLike
All these seem like good solutions that ought to help but (as noted above) the test will be in enforcement. (I also wonder why they ‘pretend’ that King has a ‘transit priority lane’. As you not it is not enforced so maybe they should just give up on it. Like the “broken window theory” about crime/vandalism, I suspect that if one law (or bylaw) is ignored and is not enforced then similar ones are also ignored.
LikeLike
Along Dufferin, the effect of left-turn and parking restrictions is pretty stark. Navigating the street on a Saturday (with parked cars taking up one lane, and lines of cars waiting to turn left taking up the other) is far more stressful than the same trip during the weekday afternoon rush.
Overall, I have a philosophical problem with the concept of free or very cheap parking on major arterials.
Steve: Yes. When I have analyzed the Dufferin bus, there are severe problems in some areas on weekends, but we concentrate all debates re congestion on the peak period weekdays.
LikeLike
They have most likely copied this from the details of the transit priority signalling used on VIVA routes in York Region. The “certain requirements” are that the bus is or just about to be running late. Vehicles automatically send out a signal when they are in that condition that will extend the green cycle if they are approaching a green light or shorten the red cycle if they are approaching a red light.
That requirement works well for a service that is on a 5 or 10 minute headway, whether “late” is based on an absolute schedule or on headway. However, I suspect that Edward’s suggestion of “vehicle in service” is more appropriate for more frequent service.
The only other requirement would involve sorting the need for priority on both roads that have transit vehicles approaching.
LikeLike
Actually, I mean for any vehicle that is in service, not being used as a training vehicle or not in service. Even if the vehicle is running on time, a red light – or a number of red lights – may cause the vehicle to become late. Frequent service or not, the vehicles on every route should be given priority over other vehicles on the road, period.
LikeLike
A while ago I was chatting with one of the drivers on King and suggested that it would be a good idea if they started enforcing the priority lane. He replied that the police had tried to do that, but unfortunately there is no place before the tracks to pull offenders out of traffic. During the blitz they ended up with a long line of cars in the curb lane, which, of course, forced all traffic into the streetcar lane and made the congestion far worse than it had been; therefore the enforcement attempt was quickly abandoned.
LikeLike
@Edward Brain – I would not want to extend signal priority to a vehicle running hot. These vehicles not getting priority should allow the driver to drift back on relative schedule. For a vehicle running into a larger gap I could see stretching the signals further to help. The scheme will need to reflect operating conditions for the route, but an effort should be made to prioritize transit vehicles running into a gap over crossing ones running early.
Steve: There are a number of exceptions to the rules that make a “standard” algorithm tricky. For example, a car that is going to short turn doesn’t really have a schedule, and it should probably be given priority especially where it turns off of the route. This can be a significant delay to following through service. Then there are cars that are either from other routes, or on diversion, or whatever, that have nothing to do with the prevailing headway on the main route the signal faces. (Never mind situations where there are two routes on the same street.) It is vital that headways be managed, not timetables, but the TTC/City seem to be aiming at timetables which are utterly meaningless in many situations.
LikeLike
To me this would mean that there should likely be a number of them, and to make the system work well, the signal control system would need to be aware of a great deal more than the mere presence of a car/bus, and when another had been there.
LikeLike
Steve, are you in effect saying that the TTC is only really giving lip service to moving to headway based management even on the higher frequency routes? I would have thought the time to make the break across the board would be when you were implementing a new signal control system, in combination with a new vehicle tracking system. It should provide you with a broader set of tools, and hence make a proper jump a lot easier. This is a spot where real service and efficiency may be foregone because of dancing to the wrong tune. Properly spaced service, provides more real capacity and a much more attractive service – hence would generate more ridership, with the same number of drivers and vehicles. This is the type of real efficiency that should be being pursued, and moving in the correct direction in the first place would make a huge difference.
Steve: The problem is as much on the city’s side as the TTC. It is “easy” to look up the time a car is due at point “x” and decide if it is “late”. It is much harder, on the fly, to determine the appropriate strategy depending on the condition of the line, prevailing traffic, weather, diversions and, of course, headway. I think that both the TTC and city are looking at too simplistic a model, one which might not actually behave as desired., or at best will have little effect. For example, if everything is “late” then there is no schedule to hold cars for.
The fact is that real consideration of every portion of every route {and at a wide variety of times of day, and traffic conditions} is required to really get it right, but properly done signal priority can be of substantial help. Understanding which direction and segment, requires more help and getting it right likely involves active signal management, not a passive scheme that merely hands out priority based on the presence of a vehicle, or even how long, or schedule is an issue, loading on vehicles (before and after) loading at stations that are ahead etc, should all be part of the consideration. Falling back on schedules, makes me think that none of this will be considered. This is like pretending there was no baseball game for the subway or GO, and running 5 minute headway near the end of the game – just because it is “off peak”. Getting it right, and being able to dynamically adjust, will make transit much more attractive, to all kinds of events that otherwise can be hugely disruptive.
LikeLike
I have to admit – to me that would make it all city, as the TTC is a city agency, and this clearly needs to be pushed to a level, where there is a will and scope to make it truly work. This is – on both sides – an issue at council, and beyond mere TTC management (although active headway management at dispatch sure needs to be a big part of this discussion,and that TTC management owns).
Also Steve, and this is not an argument with you, but like any truly worthwhile thing, an effective signal priority system will be a hard thing to achieve, and will require on-going active management. This will be true even with a reactive and well written adaptive program. It will require regular visitation, to make sure it works, and continues to do so. This is where real efficiency can be created, it will not just be found lying there in the dirt, ready to go. Real efficiency gains require, time, effort, and investment. That is not mere service cuts, but largely improved service using the same resources. This is what Tory and council should be looking to achieve, and it is not done with the stroke of a pen (although budget cuts are). To even start would mean that they were looking at the route, desired headway, and loading of each car on the route, and a reasonable forecast of loading at the stops ahead (ie what is the gap, are we at peak, is there a special event ahead on this route etc.)
LikeLike
Steve et al.:
Outside of more on-street route supervision, I’ve been racking my brain to think of other ways we could address bunching, that could be done with the existing personnel/arrangement. I know this is a problem where throwing money at it may actually help, but trying to think of other solutions. Two ideas came to my mind that I wanted to run by the group. Apologies if they’ve been covered before.
1) Would it be possible to put a vehicle leaving the line terminus into “Not in Service” giving it time to catch up to fill the gap? Or if the bunch occurs in the middle of line, have some sort of “drop off only” service??? In the case of 2 back to back cars, do you think passengers would be willing to transfer to the rear car, giving the front one a change to go Not In Service and speed up?
2) I don’t think I’ve noticed TTC bus timing points anywhere, but my own route knowledge is limited to the 72, 83, 506 for the most part… I know on other systems I’ve seen these and vehicles wait are specific route points if they are running fast. Maybe the TTC schedules are set in a way that it’s virtually impossible for a bus to be early, but curious if this could work to slow down the 2nd vehicle which is quickly catching up to the first?
Thanks!
Steve: Running express trips in front of locals is possible, and I have seen it done elsewhere. That said, people waiting for a car after a gap are rarely pleased to see a “Not In Service” car as the first of the parade. Whatever tactic is used, the whole point is that someone is actively managing the line in real time, not giving instructions to an operator to “come out on the :45” which might be for a short turn half an hour in the future, with no management of service blending at the merge point. I have been on vehicles where CIS control asks the operator (bus or streetcar) to hold at a point to space the service, but this seems to be the exception, not the rule. Hoping that somehow a signal based dispatching system will do all of this fine-grained work for them is, bluntly, an abdication of the need to understand and manage service.
LikeLike
The idea of a signal based dispatch system – at least in my mind- would be to getting the headway at dispatch close – as a baseline. The signal would then be something could control could alter – closing or increasing the gap depending on conditions. If the signal default at dispatch is set for the decided headway, and nobody does anything: a 10 minute headway route, will at least be getting a green at the dispatch point at 10 minutes. This added to a proper signal priority system should help it stay closer to 10 minutes. There are times where control should be prioritizing or removing same for a run – beyond simple on headway at this point on this run. The dispatch green does not help when the bus is not there ready to go, nor when there is no action taken when an operator jumps a red.
Having a baseline set however, should mean control is doing value added work. Creating a condition where they are tweaking for better service, as opposed to struggling to keep a basic schedule. Dispatch signaling and a sophisticated signal priority system should help move the point where active management starts. Active management should then be able to deliberately create the very occasional procession, where there is a known overload, more closely spaced service for events etc. That is where active management creates very high quality service, with basic standard headway the default, not a stretch achievement.
LikeLike
I attended tonight’s session at Scadding Court. The crowd was small – maybe 15-20 – with the majority associated with local businesses and arguing for more parking. My favourite comment was
Such a fascinating perspective of reality…
Nonetheless, I thought that the City staff did a decent job presenting the material and fielding questions. There were two that left me stumped, for which you might be able to provide insight, Steve.
a) There was no one from the TTC present and the presenters downplayed the role of TTC management in this process to-date. Do you know if that was a correct statement and if so a conscious decision? My personal experience (maybe we can consider this micro data that could complement the macro data apparently being used to drive the process) is that I’ve spent loads of time on a crawling 505 that is snarled by parked cars on one side and a long line of slowly moving cars on the other. I can’t help but see this as substandard for a high-traffic line. I would expect the TTC to be championing the initiative to rationalize the road space, talking about the possibilities it opens, and then ensuring that the changed restrictions do in fact lead to improved service that can be used to lobby for more. If it is true that it’s merely Transportation Services using TTC data, and not the TTC active and engaged, any idea why this is so?
Steve: As a consultant to this process (some of the charts you will have seen are mine), I am not in a position to comment about relationships between the two agencies.
b) About data driving this process. At times it seems that the data alone did not speak loud enough. I’ll take Queen W as an example; in the afternoon peak the proposal is to leave the same/reduce the westbound restrictions even though the running times go up substantially after parking is allowed (the one exception was Ossington to Dufferin, which had a longer restriction period, further proving the point). I imagine that the fight for evening parking spaces from business and a desire to standardize the corridor led to the proposal for restrictions 4-6 DESPITE the data suggesting that 3:30-6:30 or 3-7 would have a much bigger impact. Am I seeing this correctly that the equation is currently set as a zero-sum battle of “parking for business vs. space for travelers” whereby the former concern has out-muscled the latter, even if it had data on its side? If so, what are our chances and opportunities to advocate to change this equation?
Steve: I believe this is the eternal tug-of-war between the traffic planners, the business communities and the politicians.
Although some improvements on the shoulders of the peaks are possible, this does not deal with the more fundamental issue that there is only so much road space and time available. If we value “parking” more highly, this has implications for other uses of the road. And, obviously, the shoulder-peak changes have no effect at all on peak conditions.
One interesting point to note: some of the worst traffic screwups come from short-term occupancies of road lanes for events like construction, street fairs and media street takeovers. The effects can be felt some distance away, sometimes on parallel streets, and yet this effect is never considered when permits are granted for the events. If, for example, TIFF had to justify and pay for screwing up not just King from John to Simcoe, but the much wider effects of taking over the street and diverting transit, they might think twice about this. This is the same sort of argument that arises with things like parades and long races/cycling events. At least they tend to be on Sundays.
LikeLike
Then that begs the question of who/what a city is for… should we focus on moving traffic? Or should we be a place? an event? etc… ? I don’t know, but I think the rhetoric over the last few decades have landed us squarely in the “move traffic above all else camp”… some things can and should be fought (cabbie blocking a lane to get a coffee), but things like festivals and parades and fairs are huge parts of vibrant city life. Should they give way to longer distance commuters? I don’t know. Should we be charging with more comprehensive data based on the total impact of these? Maybe? I’d certainly prioritize transit movements over general car traffic, but even then there’s a point where local enjoyment of the street should overrule. That’s why we should have parallel service, and a plan to deal with the closures of other services.
Steve: I’m a supporter of TIFF and have been attending for over 30 years, but recently it has given a good example of how accommodating a festival is not just a question of putting up some barricades and having a street party. When they closed King, traffic on Queen worsened and produced big delays on the Queen car. The King car service was a mess because of the combined effect of extra time needed for the diversion (with little assistance from traffic signals or police) plus the slow trip across Queen. When College was closed for track construction at Spadina earlier this year, traffic congestion went up on Dundas. In both cases, there was no provision in the TTC’s service plan to adjust for these effects. One was a short-term situation, the other lasted for weeks.
LikeLike
I think that supporting events and festivals like TIFF is really important to the city – however, it also raises the questions as to how the city and province should make allowances for such events. The concept of large squares – or piazzas in Italy can create space for such events – without requiring the disruption of streets that are generally through streets. If there was a better support transit wise to the Exhibition grounds, and more modernized space set up for it, along with a couple of substantial squares set up elsewhere in the city, that were actually designed to support this sort of event, it might make sense to push for more of these events being hosted in that fashion. Ideally there should be a substantial traffic free ground in the city designed exactly for such events, with indoor and outdoor space that is not just dedicated to parking.
LikeLike