There was a verbal presentation on the Queen route at the May 21 meeting. The slides from the presentation together with some descriptive text will go online on the TTC’s site (I’m not sure where yet) with luck by the end of this week. Here are the highlights.
In January 2008, the Commission directed staff to prepare:
- a monthly report on unplanned service cancellations
- performance measures for service
- proposals for site-specific traffic improvements
- a study on splitting the route
- a strategy to measure customer satisfaction
(See minutes of the January meeting, item 10, for the complete set of motions.)
To avoid confusion about which voice is speaking here, I will save all of my comments to the end. Statements made from this point to the beginning of my comments are the TTC’s, paraphrased for this article.
Operational improvements:
1. Focus on the customer’s view of service.
- no short turning to put operators on time
- short turn only for service regularity, to fill gaps
- although this was effective in December, the status is “still educating supervisors, operators”
2. Service Assistance Crews
- additional standby operators and cars are spotted on the route to fill gaps and to relieve operators with the intention of more reliable service
- this change was staged into use in December and January with two crews for the morning, and two crews for the midday and afternoon, one in each end of the city
3. Workforce
- ensure that there are enough operators to run all of the scheduled service
- this is achieved both by training more operators on streetcars, and by the temporary closing of the 512 St. Clair line
4. Vehicle Assignments
- consistently have available and use ALRVs for runs on Queen to ensure service stability and more-even loading
- clear rules and monitoring are now in place to ensure that this actually happens
5. More Route Supervisors
- on-street supervision is preferable to impersonal instructions by radio or onboard text display
- one supervisor was reassigned to Queen in January, and six new supervisors were assigned in April working weekdays covering the period from 0600 to 2000
- two are at the Neville and Long Branch terminals, and four are on the route at various locations
What Are The Results?
A graph of schedule compliance tracks the percentage of runs that are within two minutes of the scheduled headway. This has crept up slowly from the 20-25 percent range to 30-35 percent. Meanwhile the percentage of headways that are off by 10 minutes or more has dropped from roughly 10 to 5 percent. Similarly, the percentage of bunched streetcars (defined as headways of 1 minute or less) has fallen from about 10 to 6 percent.
The absolute number of short turns fell from about 2,500 in October and November to about 1,900 in December and 1,300 in January. February and March saw a rise back to the 2,000 range due to bad weather, and by April the count was down to about 1,600. The TTC now has a daily report of short turns and the reason for them. On a sample day, April 18, there were 81 short turns all attributed to Traffic Congestion or Blockage of the line.
For January-April 2008, over 99% of the scheduled service (measured in vehicle hours) has operated. However, TTC staff are not comfortable with this metric because it lumps the entire day’s and route’s operation together.
Monitoring the Route
The TTC now produces daily graphs summarizing the headways at various points on the line so that they can review what worked or didn’t work in line management on an ongoing basis.
The old signpost-based Communications and Information System (CIS) (which I have described in some of the early articles about route performance) will be updated this year, probably by the fall, by conversion to use the Global Position System (GPS) data now available on all vehicles through the automated stop announcement system. The management displays for CIS control will be changed to a map-based format with considerably more data about each car, and the TTC is considering what would be involved in making such displays available to supervisors in the field.
Traffic Management
The TTC is reviewing the entire route for locations where traffic controls will improve transit operations for presentation to the City. This work is not yet complete, but recommendations to date include:
- Bathurst to Roncesvalles: No parking in the counter-peak direction during peak periods
- University to Victoria: No stopping any time
- University to Spadina: Eliminate 16 parking spaces on the north side of Queen.
- Bathurst to Parliament: Prohibit left turns from 0700 to 2200 (currently the ban is from Spadina to Jarvis between 0730 and 1830)
- Spadina to Victoria: Prohibit peak-period right turns at signalized intersections
- City Hall Garage: Prohibit the east-to-north left turn into the garage
- New signal priority at Jarvis, York, Glendale, Colborne Lodge, Windermere and Ellis
- Improved priority at Broaview, Bathurst, Dufferin and Lansdowne
Splitting the Route
Various scenarios are under evaluation for restructuring the Queen route. Considerations in the study include:
- the impact of transfers and the loss of through service for some rides
- changes in service frequency
- operational feasibility with track, switches and overhead
- locations for terminals, layover space and washrooms
- operating cost implications of overlapped operation
- vehicle availability
- implementation timing (Church and Parliament, possible locations for turnbacks, will be rebuilt in 2008)
The schemes under consideration include:
- the existing service (Neville to Humber + Neville to Long Branch)
- all cars to run from Neville to Long Branch
- return to the pre-amalgamation setup with a Neville to Humber service and a separate Long Branch route from Humber to Long Branch
- a Neville to Humber service overlaid by a Long Branch to Church service
- a Neville to Humber service overlaid by a Long Branch to Dundas West via Roncesvalles service
- a Neville to Bathurst service overlaid by a Long Branch to Church service
- a Neville to Bathurs service, plus a Church to Humber service, plus a Long Branch to Church via King service
- a Neville to Bathurst service overlaid by a Long Branch to Woodbine Loop service
The evaluation of these options may be complete in July 2008, but it is uncertain whether it will be presented at that time.
There will be an onboard interview of riders in June 2008 asking if they have perceived an improvement in service quality, and a second survey will be done if and when a route split occurs.
The next report to the Commission is expected in October 2008.
My Comments
All of the preceding was Steve-as-reporter. Now you get Steve-as-transit-advocate.
It is clear that the dust has not entirely settled on the internal view by TTC management of the course (or courses) they will take. Culture change takes time in large organizations, and everyone has to feel that the changes proposed will actually make a difference.
It is essential that City Council address the traffic management issues. The local Councillors (interestingly Chair Giambrone plus Gord Perks in Parkdale and Adam Vaughan in downtown west) will have to decide whether transit really does come first. The changes proposed here are not as invasive or as permanent as the proposed King Street transit mall, and they deserve implementation. The Works Committee, chaired by Commissioner De Baeremaeker, must decide to give the TTC top dog status in the ongoing wrangling with the traffic engineers. Only by showing that Council is willing to work at improving the situation out on the street, to the degree that they can, will operating staff believe that the load isn’t all on their shoulders.
I was particularly struck by the fact that four of the six locations where transit priority signals are proposed are on The Queensway. Here we have the oldest surviving piece of private right-of-way in the city, and the streetcars don’t run the traffic lights. In fact, the travel time from Humber to Roncevalles is now longer than it was before the recent “improvements” which hold cars at locations where they formerly sailed through intersections. Transit “priority” is a very flexible concept.
The whole issue of short turning and headway management needs to be better understood in the field. We still see platoons of cars and wide gaps. I find it impossible to believe that those 81 short turns on April 18 were all due to traffic delay or route blockage. Nothing was reported against construction, disabled cars or equipment problems, passenger disputes, overcrowding or accidents. What a charmed day that was that every single short turn was triggered by our old friend “congestion”.
The operational improvements seem to be slow in coming. We have only reached the one third mark in runs operating within two minutes of the scheduled headway. This corresponds with the analyses I did for December and January, and suggests that there really hasn’t been much overall improvement in headway regularity. Benefits seen by riders may be as much a question of seasonal fluctuation as of any explicit change in management style.
I am particularly amused by the idea that there is a supervisor sitting at Long Branch. This has to be the softest job on earth considering the scheduled headway is only 11 minutes, and short turns (which do occur) make for wide gaps. At least the operators will have someone to chat with during their layovers.
Also, the complement of six new supervisors looks impressive, but we must remember that they cover a 14-hour window. That means three per shift, a much more reasonable number than six (over and above the existing crew).
I look forward to the new, improved CIS, and can’t wait to get my hands on some GPS-based data that will eliminate the tedious process of mapping current route monitoring logs into useable format.
My position on splitting the route has been published before, but for the record it is:
Weekdays until about 1900:
- A Queen service from Humber to Neville with extras as required to operate on the central part of the route
- A consolidated and improved (off peak service no worse than 10 minutes) Kingston Road route with all cars going either to McCaul Loop or to York via King (the jury is still out on the choice here)
- A Lake Shore service from Long Branch to Church via King
Evenings and weekends:
- A Queen Service from Humber to Neville with extras as noted above
- A Lake Shore service from Long Branch to Dundas West via Roncesvalles
After the meeting, Adam Giambrone was scrummed by the media, and among his comments was the observation that the Queen corridor and the Sheppard Subway carry roughly the same number of passengers, but there are far, far more people tending to the subway than to the streetcar line. The TTC needs to recognize the importance of its surface routes. To be fair to the Sheppard line, it does have a lot of infrastructure that the Queen car doesn’t need (that little tunnel for starters), and some of the staffing differences are a direct result of the overhead costs of running a subway line. But you’ve heard that speech from me before.
Next Steps
A public meeting will be arranged for an evening at City Hall likely early in June. This will allow the public to comment on the various proposals in a less formal and constrained atmosphere than a Commission meeting, and without having to take time off work. It will allow much better give-and-take on the issues and allow any management or politicians who attend to get a clear understanding of how people feel.
When I have more informaion about this meeting and other future events, I will post an announcement here.
Steve Munro 
My guess is that Adam or someone else on high has seen your analysis and the TTC’s own report does not measure up to your standards. It would be cheaper and more efficient if the TTC just paid you for your analysis and followed your recommendations.
LikeLike
I agree with Dave M. Your report is to the tee and outstanding. The three recommended route changes you proposed will solve the problem. I believe the 502 should run permanent to Bathurst station from Vic park to provide extra service on Queen between Queen/Kingston road and Bathurst. I also think we need a downtown Loop which would be better than using streets Like Church Richmond etc…. This will help the Queen car and other routes. The only location I can think of on Queen is Moss park, south side by the armoury. Then you can run Humber-Moss park route, Neville Park- Bathurst. Run the Lakeshore cars downtown, as you stated, via King during rush hours, to this loop via church. Eventually when they complete the Adelaide track, they can run the Long Branch cars through here during rush hours, using Spadina. Beach car/502 short turns if need be, can use this loop. I had a sneaking suspicion they are trying to sweep away any ideas to improve Queen. This route has Cancer with a probability of recovery of 100%, with no ability by anyone on the TTC, to provide a cure or any treatment. Thanks Steve for using your time to spell out the truth to this route. A consultant would of charged the TTC 1 000 000 or more for this work. Send your findings to Gary Webster and see what he says.
LikeLike
It’s quite plausible and probable that Steve’s efforts are setting the bar quite high, though seeing the presentation did make a positive impression that they are putting time, resource and thinking into it, more than into another east-west project nearby, which was rubber-stamped despite costs, and flawed reports and presumptions.
LikeLike
I agree with the idea of implementing more signal priority. However, it needs to be a modern system, not the current half-assed system which increases delays half the time when the green extension runs out while the streetcar is loading and the light turns red. The system needs to be able to both detect a streetcar from a distance to turn the light green (at lights with no stop or a far-side stop) and have a button allowing the operator to keep the lights red (when there are lights in front of the stop) until the car finishes loading. The TTC needs to hire a decent consultant to do something about this.
Steve: The TTC has decent staff of their own who know what the problems are. The real problem lies at the City where the traffic engineers don’t want to implement true transit priority and thwart the TTC’s requests. Nothing short of political intervention is going to fix this, but Councillors are notoriously unwilling to override the technical staff on issues like this.
LikeLike
Signal priority should not be left just for streetcars, buses should have the same system as well. Not only would buses be faster, but private traffic could benefit from this technology with longer green times.
Steve: Several bus routes have systems in place similar to the streetcar routes.
It’s time for the city to widen our city roads by one lane (Wherever possible) in each direction and make it a transit only lane. Then that’s it, no more catering to the private auto. Express buses can use the other lanes to not block local buses.
Steve: The problem is that the streets where there is really serious congestion cannot widened, indeed, that’s part of the reason for the congestion in the first place. The older parts of the city have buildings out to the sidewalk at major intersections, and there is not even room for a turn lane.
Up on St. Clair, you can see the mess that attempting to shoe horn in streetcar lanes, wider islands and turn lanes has produced.
LikeLike
As always, Steve makes sense.
Here is my funny, but plausible, solution to streetcar bunching. I have always thought that 2 streetcars going the same direction should never physically be on the same block. Regardless of what has happened with short turns I think it is human driver nature to rush together, when that happens 3 streetcars basically become one which is not very helpful. Drivers should be disciplined if they end up driving 3 feet behind another car.
I would add that streetcars in sequence also makes it hard for all other modes of transport to move as well.
LikeLike
I’m impressed they have daily headway graphs — this makes fine-tuning much easier and faster, assuming the supervisors have the authority to make tweaks without waiting for the next Commission meeting.
The outcome of what you list under “traffic management” will be very telling. There isn’t the ability to hide behind most of the usual excuses (no authority, too expensive, at odds with other agency or level of government, requires lengthy study because it’s irreversible, obstructionist local councillors, etc.) It’s downtown and the stated goal is to favour transit over autos. If nothing changes, it will prove right the critics who say the mayor talks a great game but can’t follow through.
Steve: You can’t pin every issue on the Mayor. However, the support of his office will be quite useful in ensuring that better transit priority is provided.
LikeLike
The Ontario government should enact a new traffic law that says that any far-side streetcar or bus stop MUST be accompanied with transit priority in favour of a larger capacity vehicle. The city council doesn’t seem to have the fortitude to do it themselves.
If your automobile can carry the 132 people that a CLRV can carry, let them have the priority. Otherwise, the streetcar should get first access across the intersection. A capacity of 5 or 6 shouldn’t get priority, even if they are turning left.
LikeLike
There are some indications in the list of route splitting options being evaluated that make me question the TTC’s competency here. The reason for this is that they seem to be jumping directly to “options” without first stopping at “objective.” If I asked them “what is the splitting of the route supposed to address/achieve,” I wonder what their answer would be. The most striking one in the list was “run all cars from Neville to Long Branch,” but running cars from Long Branch to Woodbine Loop was also quite the prize-winner: I can’t help but suspect that some people (not everybody, but some) are just going through the motions, rather than actually making an effort to problem solve, because these options really have no practical purpose being evaluated as they are too long.
Steve: I think every scheme anyone has put forward is in the hopper for completeness, although I am troubled that they didn’t include the 502/503 restructuring in the overall review as this has an impact on service on Queen (or King) depending on the option chosen.
My understanding of the situation is that there is some debate still underway within the TTC about what they might be trying to achieve and what approaches might work.
LikeLike
scotd says:
“I have always thought that 2 streetcars going the same direction should never physically be on the same block. ”
Sometimes the following driver is running ahead of schedule, but often on Queen the lead car is way late.
If there’s a big gap ahead of the lead car, it will be running very slowly with all the extra traffic. People will wait for the following car if it’s right behind, but if it’s far back so it’s out of sight or the destination can’t be read, they will try to crowd onto the lead car. The lead car slows down even more, the crowds grow….it’s a self-perpetuating process.
In addition, it’s pretty annoying to be on a lollygagging following car that is trying to keep three blocks behind a slow lead car.
The other day, I was on a westbound 501 in the PM rush that was running quickly until about Ellis Park, where we caught up to two other cars. We had a five-minute wait at Humber Loop for the first car in the parade to decide to pull around the turnback loop and head for Neville. (Why did it lay over on the westbound track? Who knows.)
After that, we followed a not-particularly-ambitious 508 Lakeshore car, which turned at Kipling (how often does that happen?). After the 508 got out of the way, we resumed brisk operation. The car was still reading -15 on CIS when we approached Long Branch loop, so it’s not like it was running ahead of schedule.
I said to the operator, “I guess if you’re slow enough you get turned at Kipling”.
He replied “That’s not how it works. The lead car is never turned.” (I don’t think we were communicating too effectively. There was no car behind him that I could see.)
LikeLike
W. K. Lis said “The Ontario government should enact a new traffic law that says that any far-side streetcar or bus stop MUST be accompanied with transit priority in favour of a larger capacity vehicle.”
I’d think legislating that at the provincial level is an overshoot. Technically, that would mean that even at a rural intersection that sees a GO bus once in 3 hours, must have traffic lights with transit priority.
While discussing public transit solutions for GTA or Ottawa, let’s not forget that over 90% of Ontario land is, well, a very different environment.
LikeLike
It’s important that TTC get this right – after all, if they can’t figure out a way to interline Queen rationally, the argument that Eglinton LRT makes sense because you can have a fan of interlining routes outside of the tunnelled area takes a knock.
LikeLike
Rainforest says…..
“W. K. Lis said “The Ontario government should enact a new traffic law that says that any far-side streetcar or bus stop MUST be accompanied with transit priority in favour of a larger capacity vehicle.”
I’d think legislating that at the provincial level is an overshoot. Technically, that would mean that even at a rural intersection that sees a GO bus once in 3 hours, must have traffic lights with transit priority.
While discussing public transit solutions for GTA or Ottawa, let’s not forget that over 90% of Ontario land is, well, a very different environment”
In my view that kind of outlook is anti transit. First of all, so what if a bus runs every three hours, it has passengers and is getting cars off the road. Second, Yes the landscape is diffrent. Iam actually getting ready for a weekend at the cottage. And I wish we had a service by the cottage, but that is “too far fetched.” Third, A bus running every three hours is a joke, and something needs to be done about it.
Getting back to the subject at hand, any public transit bus, streetcar, train should have the signal system in place, regardless of the headways.
LikeLike
W.K. Lis and Matthew Kemp are proving my point.
We’ve been trained to believe municipal governments are poor and powerless, and therefore lack of progress isn’t their fault. It’s sometimes true, but in this case, it’s simply a case of local political will. That’s not something Queen’s Park should legislate; it’s something where we as the voters have to hold the local politicians responsible.
Obviously the mayor shouldn’t have to get involved to make a bunch of transit priority measures happen. But if they’re not happening, the mayor should worry about the mixed messages that sends to the city’s engineers and planners. If those battles aren’t fought now, they’ll only get harder as detailed design starts for Transit City, and the stakes there are arguably much higher.
LikeLike
Matthew Kemp said “In my view that kind of outlook is anti transit. First of all, so what if a bus runs every three hours, it has passengers and is getting cars off the road. Second, Yes the landscape is diffrent. Iam actually getting ready for a weekend at the cottage. And I wish we had a service by the cottage, but that is “too far fetched.” Third, A bus running every three hours is a joke, and something needs to be done about it.
Getting back to the subject at hand, any public transit bus, streetcar, train should have the signal system in place, regardless of the headways.”
Note that installing the transit priority system, and then maintaining it, costs money. The cost will be approximately same for a busy urban intersection and for a rural road. However, the benefit that system gives to a bus line running every 3 hours will be diminutively small, compared to the benefit it gives to a streetcar line running on a 3 min or 5 min headways.
A bus running to cottages is “too far fetched” indeed (for most locations), but not because the public is under-educated about the benefits of transit. It is too far-fetched simply because there won’t be enough riders. Even leaving fiscal matters aside, simply take into account the fact that 1 bus consumes more fuel and produces more greenhouse gases than just 2 or 3 cars. In needs to replace at least 5 – 7 cars to break even (and then of course the bus will be quickly gaining advantage if its load grows).
Public transit is not a new religion that needs places of worship such as a signal priority system at every traffic light. It is just a tool to do a job, great tool in many cases but not a universal one. An example, a screwdriver is a great tool to handle screws. But it won’t work very well for nails; and if I say so, you probably won’t call my comment “anti screwdriver”.
LikeLike
I especially like Dave M.’s response, the first one. I agree with his observation that your (Steve) work here is well done. It seems surprising that they are willing to provide the information to you. Hopefully they have done some kind of similar analysis in-house?
Of course what is really at work here (I think) is another example of someone who is educated and passionate enough about something to put pride and lots of effort into the work. There’s likely lots of TTC employees in service planning jobs that simply find the whole job ho hum and may not even have the depth of knowledge of data presentation and interpretation to understand what Steve has written here. There’s many people with interest and passion about certain topics that could do an awesome job in the right position but too often this type of person is not the one making the ‘official’ call.
Compliments again to Steve for this work. It is an excellent analysis of what was likely an incredible amount of information and seemingly done the old fashioned way, with a lot of modification and formatting to end up being presentable. Not to berate the point here but I am skeptical if analysis like this exists at the TTC. It’s easy with computers to collect megabytes and gigabytes of data but not everyone has the brains to figure out what they mean.
Steve: The problem is partly jurisdictional within the TTC between Service Planning, Operations and IT. One big advantage I have is that in a way I have a good awareness of the requirements of all three — I am the IT department, the business analyst, the developer and the intelligent client all rolled into one.
The TTC has started to do some of their own analyses although what they have published to date is not is extensive.
LikeLike