The 2013-2022 Capital Budget includes plans for the conversion of streetcar services from the existing CLRV/ALRV fleets to the new low-floor light rail vehicles (LFLRVs), related projects to bring the infrastructure up to the required standard for the new fleet, and continuing system maintenance.
Updated Nov. 4, 2012 at 8:30 pm:
The headway shown for LFLRV operation on Queen was incorrectly calculated in the spreadsheet. This has been fixed.
Fleet Plan
The version of the plan linked here is adapted from the TTC’s information by me to show the likely effect on service capacity and frequency for each route.
The first page gives the TTC’s fleet plan. 204 cars in all will be delivered, and at the end of the decade there will be a modest surplus relative to the projected requirements. Additional cars for Waterfront Toronto projects appear separately in the Capital Budget with money for Cherry (5 cars), Waterfront East (6), and Bremner (4). There is an overall budget line for the Port Lands, but no specific mention of vehicles. The proposed dates in the budget for delivery are based on the original plan for waterfront transit expansion, but these dates are clearly not going to be met. Whether cars will be bought for Cherry Street in the next few years, or if the TTC will simply absorb this short extension into the base fleet, is hard to tell. Probably this will depend on whether other Waterfront Toronto LRT projects go forward.
There is a considerable provision for increased capacity on the streetcar routes (discussed below), but no mention of redeploying CLRVs onto currently overcrowded lines as they are released by LFLRV route conversions. In the short term, the primary use for those CLRVs will be to replace the unreliable ALRVs on Queen and King while the LFLRVs displace CLRVs from Spadina/Harbourfront, Bathurst and Dundas. By the end of 2014 when the TTC should have a year’s worth of new cars on the property, the excuse that service cannot be improved because there is no spare equipment won’t wash any more. The Commission will have to address the issue of service capacity on all routes, not just those that benefit from LFLRV conversion.
The second page of the chart shows the implications for service capacity and frequency based on the number of vehicles allocated in the Fleet Plan. It presumes that running times will not change with the new fleet. (This chart is entirely my work based on the Fleet Plan.)
The first step is to convert the existing mix of CLRV and ALRV service to its equivalent CLRV capacity. Each ALRV is presumed to equal 1.5 CLRVs, and each LFLRV is presumed to equal 2 CLRVs. For example, Bathurst now operates with 13 CLRVs, and is planned to have 9 LFLRVs. The ratio of vehicles is 0.69, but when the capacity of the new vehicles is taken into account, the ratio is 1.38. However, with fewer vehicles on the line, the headway will change from 4’00” to 5’47” (probably rounded to 5’45”).
If the TTC cannot achieve double the capacity for the new cars even with all-door loading, then the capacity estimates will scale down. This will not affect headways which are based on the number of vehicles, presuming no change in running times.
Queen is a special case because it operates with ALRVs, but once the 31 cars on the line are converted to their CLRV equivalent, 46.5, the remainder of the calculation works the same way. The capacity of the new service will be 1.42 times the current operation, and with the number of LFLRVs almost equaling the current number of ALRVs, there will be little effect on a slight improvement in the scheduled headway.
King actually operates with two blended services. One is a base headway of 4’00” which I have treated separately for the calculations. Overlaid on this is a 4’00” headway of trippers that make only one trip to provide a wave of 2’00” eastbound service for about an hour during the AM peak. Replacing the base CLRVs with LFLRVs at a ratio of 1.3:1 (the reference value used for the Fleet Plan) gives roughly a 50% increase in capacity (twice the capacity per car, offset by a 25% reduction in the number of vehicles). The new combined headway would be about 2:40.
The worst-hit routes will be Dundas and Carlton where the added capacity will be small (about 25%) and the increase in headways substantial. Even St. Clair only receives 28% more capacity, but with its shorter existing CLRV headway, the LFLRV headway is just under 4′. It is clear that although the fleet plan claims a 1.3:1 replacement rate for CLRVs, some routes do better than others. This is of particular concern on routes that will have wide headways such as Dundas where a single short-turn will create a gap of over 15′. The TTC will be unable to achieve its “three minute rule” for headway adherence. Gaps and bunching will be more critical with wider headways, but whether the TTC will continue “business as usual” in line management remains to be seen.
Track and Overhead Plans
The plan for track replacement is almost identical to the 2012 version published a few months ago on this blog. Changes include:
- New track installation on Queen’s Quay West (omitted from previous list)
- King & Sumach intersection moved from 2013 to 2015
- Spadina Station Loop added in 2014
- Several projects, mainly in yards, added in 2015 when there is an embargo on road construction that could affect the Pan Am Games
- St. Clair from Bathurst to Tweedsmuir (the portal to St. Clair West Station) added in 2016
- Humber Long Branch loop and CNE Loop added in 2017
What is striking in this plan is the decline in tangent track replacement projects now that the TTC has caught up with the backlog of poorly build pre-1993 track. Much of the work in this plan involves intersections, as well as short stretches of tangent rail used mainly for diversions and short turns.
There is still no sign of a project to rehabilitate Adelaide between Victoria and Charlotte.
The TTC expects surface track to last 25 years or more except at locations of high wear such as carstops and curves. Track built between the late 1960s when the TTC stopped welding rail, and especially since a move to untreated wooden ties in poured concrete, would only last 10-15 years with the low end of the range in high use areas.
Queensway Reconstruction Project
The TTC proposes to extend the reserved lanes on The Queensway east to Roncesvalles in 2016. This will require an formal assessment (probably a “Transit Project Assessment”) which has not yet been launched. The changes include:
- Widening of The Queensway to provide east-to-north turn channels at Glendale and at Sunnyside
- A new traffic signal at The Queensway and Sunnyside
- Curbs beside the streetcar lanes
- Relocation of the eastbound stop at Roncesvalles to the far side of the intersection (this may be challenging considering that Queen is only four lanes and the sidewalk is not very wide)
- Transit priority at signals at Glendale, Sunnyside and Roncesvalles
- Removal of the east-to-south turn channel at King and Roncesvalles to make room for an east-to-north left turn lane
Also scheduled for 2016 is replacement of the south exit track from Roncesvalles Carhouse and the replacement of the King-Queen-Roncesvalles intersection.
Overhead and Power Distribution
The TTC has a multi-year plan scheduled to finish in mid 2017 to rebuilt all of the streetcar overhead for compatibility with the LFLRV fleet. This will involve stringing heavier gauge wire and changing all of the suspension to be pantograph-friendly. Once the CLRV/ALRV fleets are retired, further changes would occur to take advantage of the absence of trolley poles and shoes. (It will be interesting to see the heritage fleet converted to use pantographs presuming that these cars have not yet been scrapped or sold in a fit of budgetary pique.)
The tangent wire will be largely converted by the end of 2015 with intersections planned out to 2017. That poses a problem because of the wandering nature of streetcars in Toronto. Although the last route won’t convert to LFLRVs until 2018, diversions and short turns take streetcars to unusual places. Partial system operation with pantographs could run into snags with unexpected routings into unconverted territory.
A long-standing program to replace overhead support poles will be largely completed in 2013 as will the program to replace feeder cables. Some of this infrastructure is well over 50 years old.
All underpass troughs will be rebuilt to new standards by 2015. The first major example of the new style is in the King Street underpass between Sudbury and Atlantic. This has also been used at Queen & Dufferin, McCaul Loop, and the reconstruction of underground sections of Spadina/Harbourfront.
Two new substations will be added to beef up power on the outer end of the Queen line, one near Neville Loop and one in Humber Loop.
The matter of streetcar track switch controllers has still not been settled. The project has now been complicated by the appearance of subway-level standards for a surface project including:
- switch position detection and indication (so operators know how a switch is set even in the dark, rain or snow),
- switch mechanical or electrical locking, and
- use of higher quality “vital relays” for what is considered a safety application.
How this will all work for track sitting in the street, a much more hostile environment than a subway, remains to be seen. Preliminary design is completed, but further evaluation of alternatives will be done prior to a system selection. The budget shows spending mainly in the years 2014-17. I cannot help wondering how street railways and surface LRTs the world over manage to operate, or why the TTC has taken so long to fix the existing system that has misbehaved more or less continuously since it was installed about 25 years ago.
Waterfront Transit
Both the East Bayfront and Port Lands LRT lines are now shown as post-2022 projects (beyond the scope of the current 10-year budget). This could change if funding became available sooner, but the big sticking points are the reconstruction of Union Station Loop and the rising cost TTC estimates. The East Bayfront line is budgeted at $294-million including work at Union Station, and the Port Lands at $188m. Developers in East Bayfront continue to complain that the promised level of transit service is not going to be operating when their buildings come on stream. Funding for waterfront transit is unlikely to appear from any of the governments now in power.
Steve Munro 
Streetcars: what’s the latest year where the TTC could extend the order without any gap in production (and associated re-start costs)?
Steve: I suspect that Bombardier wouldn’t object up to about 2016. There is the timing of production for the Metrolinx cars, but from the point of view of sourcing components, both fleets have much in common and it’s not a question of restarting from scratch.
LikeLike
The increased headway has me fairly disheartened, I can hardly wait to see what the gaps will look like. More and more I’m beginning to feel the TTC has a policy of driving people away from streetcar routes. Either that or they really, really don’t care. These headways are the peak period headways on Dundas and Carlton? What will off-peak look like?
If the TTC were to follow this to its logical conclusion we should just have one giant streetcar on each route … who cares if you have to wait three hours, the reliability from all that extra running time is great! and with 45 doors look how fast boarding is! Route supervision will be so much easier! What an increase in productivity! The metrics on the CEO’s report will be stellar.
Steve: And even with only one car, there would still be pairs of Route Supervisors standing at the same corner monitoring the service.
LikeLike
The relocation of the 501 eastbound stop at Roncesvalles & Queen to the far side of the intersection will be problematic. Unless the south-east building at the King & Queen intersection is removed (highly unlikely), I would rather keep (and improve) the safety island.
Steve: I agree. There is going to be a vigourous debate in this neighbourhood as we seem to be making changes more to benefit eastbound motorists, including making it easier for them to access residential streets parallel to The Queensway, while doing violence to the character of Queen & Ronces itself. I was eating brunch in “Easy” just a few hours ago and am baffled at the idea of a farside carstop here.
LikeLike
A few things:
It seems odd to be putting a car left turn from Queensway to Roncesvalles (this is prohibited now), while keeping Glendale and now signalizing Sunnyside as well. I thought Sunnyside was currently being used for left turns because Roncesvalles was prohibited? There seems to be at least one too many (full-fancy signalized with umpty lights) left turns being considered here.
What is “Humber Long Branch loop”? The pretty much unused loop at Humber for Long Branch cars?
Which then lets me notice that Long Branch loop is not on the list. They rebuilt half of it (the west end), but not the other half. Anyone getting off at the loop gets to walk over mud, gravel, and indifferently-laid patches of asphalt, to cross from the north side of the loop to the shelter and sidewalk.
Steve: See my comments about the Queensway project above.
LikeLike
I continue to be baffled that the TTC isn’t working on upgrading Union Station Loop right now.
The station is under construction, and both Spadina and Harbourfront are converted to buses and not running to Union. Instead, the TTC opts to inconvenience its riders again at a later date.
This isn’t a funding problem, is it Steve?
Steve: It is both a funding and a bureaucratic problem. First off, the waterfront projects fell off of everyone’s radar because Waterfront Toronto was paying for everything. However, the TTC’s estimated cost of actually building the improvements, notably Union Loop, went through the roof and well beyond what WFT had budgeted. Also, there is the question of whether provision for the “Bremner” car (which I regard as a TTC fantasy that will never be built) should be made now or at some future time. Stir in a co-ordination problem where TTC engineering was using the pre-rebuild Union Station as their reference for the loop upgrade, and you can see how all of this ran aground.
Council seems finally to be taking an interest because of concern by developers, but, yes, the Union Loop project will be completely separate from the Union Station rebuild and the Queen’s Quay project which shuts down the 509 for most of a year.
Thank you TTC. This is another fine example of how we can fantasize about billions in new subway lines, but miss the obvious projects on our existing system.
LikeLike
Though I appreciate why the City in general (and TTC in this case) are trying to avoid any major on-street roadworks in 2015, I note that the Pan-Am and Para-Pan-Am Games run from July 10 though August 14. The TTC (and the City) often start projects in mid-August so wonder why they are planning to budget for at least some on-street work once the Games are over.
Steve: The Pan Am Games have some sort of mythic value for many politicians who are terrified they will look bad if even a gnat gets in the way and this can be blamed on Toronto. They issue a blanket embargo and nobody thinks about the option of fall construction.
LikeLike
The City project to lay new watermains and sewers between Spadina and University is supposed to be finished this fall and I assume that local resident and businesses will expect that be the street (which is in terrible shape) will rehabilitated and possibly improved in 2014. If the City are going to “do” the street I suspect the TTC will be told to replace or remove their rails “now”.
As noted in other posts, the TTC are busily erecting new poles for overhead along Adelaide and last week on Adelaide just east of Victoria I saw them cleaning out an on-street switch that has not seen a streetcar in a decade or more.
Steve: One point about overhead poles: I noticed on the feeder upgrade map that there is a line running east from Bathurst along Adelaide, and it is possible that some part of downtown needs this (including new poles even without tracks). I will have to walk the street to see where that feeder actually goes.
LikeLike
Are there any plans to save one or two CLRVs for the heritage streetcar fleet?
Steve: “Heritage” is a word unknown to the current administration at City Hall. By the time the CLRV fleet is coming to an end, the situation may be different. Having said that, they will be very hard to keep in working order as the electronics are difficult to source for spare parts.
LikeLike
I have a similar concern as Jamie. It’s great that the new streetcars are handicap friendly, and larger than the ALRV, but because the total number of streetcars will be less than the current number of CLRV and ALRV, there has to be lower frequency. It’s great that when it comes I might get a seat, but what’s the point if I have to wait too long to get one. I am more likely to choose another mode of transport (even an all bus route) if I am going to wait too long to get a streetcar.
Whoever came up with this idea made a very unwise decision.
I think the days of the heritage cars on the TTC are numbered – I doubt they will convert them to pantograph operation (which would make them look bad in my opinion) so get out there and get your pictures of them while they are still on the roster.
Steve: PCCs and even Peter Witt type cars with pantographs are not uncommon. Here are Peter Witts in Milan, Italy. Here are Pittsburgh PCCs with a pantograph. Both cars were in revenue service at the time the photos were taken, not part of a heritage fleet.
LikeLike
Interesting, thanks Steve. The headway on Dundas with new vehicles is particularly surprising. 8 minutes is a ludicrous amount of time to wait given the TTC’s typical schedule adherence and the demands on even that relatively placid route. Here’s hoping that gets revisited.
More broadly, I am hoping the arrival of the new cars prompts a basic rethink of TTC streetcar operations. All door boarding plus the removal of some redundant stops plus aggressive line management could totally transform the service–having seen the speed and reliability of mixed-traffic trams in places like Leipzig, the TTC’s insistence on running, essentially, buses on rails is mystifying.
Two questions, though. First, would it make sense to do a rebuild of a small part of the CLRV fleet to keep around as ‘surge’ capacity? This seems to be what they have done with Brussels’ somewhat Toronto-like tram system, with new Bombardier cars the main vehicles but some older ones thrown in to fill gaps.
Second, obviously going back to a plan for 204 cars means Council will have to approve a restoration of the capital budget cuts demanded by the Ford administration last year, which reduced the figure to 189. Is your sense Steve that Council is on board for such a restoration? Or could this get screwed up in the budgeting process?
Steve: I don’t think the TTC has actually reduced the order, just done some budget shuffling to make an accounting problem go away. By the time they would have to decide what to do about this, Ford should be just a bad memory. If he is still around, we have much bigger problems.
LikeLike
This implies no more 508 type services from Long Branch to downtown via King? With all the proposed development coming up (Christie Bakery, Parklawn, 4 points Sheraton) this is going to be problematic. As it is right now, 501 eastbound in the mornings is standing room only by the time it gets to Windermere.
Steve: The channel in question is a road channel which splits the King Street traffic off from the Roncesvalles traffic signal. There is a small island between this lane and the through eastbound lane. The track layout stays the same.
LikeLike
Perhaps the person who designed the far side stop eastbound on Queen at Roncy is the same genius who designed the far side stop eastbound on St.Clair at Keele!
LikeLike
Toronto Streetcars:
In 50 Years of Progressive Transit, there is a picture of Witt 2502 with a bow collector. (1929). This could make part of the heritage fleet authentically compatible with the pantographs.
LikeLike
Are the ALRVs actually more undependable than the CLRVs? I didn’t thing there was any difference at all. I’ve always assumed that one was just as bad as the other.
Steve: The ALRVs have been problem children, and that’s why you often see CLRVs running where ALRVs are scheduled (and without any change in the headway leading to overcrowding).
LikeLike
Steve wrote:
With the lousy service the TTC operates on (believe me it is pathetic at times) how is anyone supposed to know the difference? Part of the issue here is how the TTC schedules its routes – take the 501 car. Whether or not it is operating with all its ALRVs or not, the schedule is lousy and cars are consistently late – I had to wait for a streetcar and according to the posted schedule three cars should have come and none have arrived. Then the car is near capacity.
I have also seen three ALRVs show up westbound at Kipling and the Lake Shore. Instead of sending one or two back east at that point, all three go through.
With the new cars, the TTC really has to maintain a reliable service, and they can’t do that now, so service will be reduced – I don’t see anything getting better with less cars to work with.
Steve: I am baffled at line management styles. Sometimes gaps are left to fester as you described, other times cars are short-turned in such a way that they actually create parades rather than breaking them up. Yes, the TTC has to fix this problem pronto or service with the larger cars will be appallingly bad. I am sure TTC operations will trot out their usual lectures about “mixed traffic” and ignore the fact they can’t maintain reasonable headways even on streets with reserved lanes.
LikeLike
The current fleet plan does’t make a lot of sense. Scrapping the ALRV’s before the routes using these cars are actually converted to low-floor operation would push the CLRV requirements in the short term. Based on the data provided in the spreadsheet, converting Spadina, Harbourfront, Bathurst and roughly half of Dundas in 2014 would free up about 44 CLRV’s, of which 14 would be scrapped leaving only 30 cars to fill in for 42 ALRV’s to be retired the same year. Service “planning” at its best.
Even assuming the CLRV retirements are postponed (leaving the full CLRV fleet available for service until at least 2015), this doesn’t make sense, unless they plan to in initially replace the ALRV’s on Queen with CLRV’s on a roughly 1:1 ratio, with a significant decrease in capacity. Which of course cannot be done.
I get the feeling these plans are made up as they go along.
Steve: I agree and think that whoever put together the various iterations of the fleet plan doesn’t know what they are doing. An ad hoc plan that looks good on paper as long as you know nothing about the actual effects.
I was comparing the fleet plan from two years ago (published by Steve here) and there are significant differences both in timing of route conversions and the number of cars assigned to each route. Interestingly, Dundas and Carlton, the hardest hit routes in terms of headway under the current plan “lost” five cars each compared to the old plan. Most likely they will change their plans again next year, so knows by how much the headways will actually change by the time the new fleet hits the streets.
And speaking of headway, Steve, there is a small error in your headway calculation for the Queen route. If it uses 31 cars on a 5’10” headway at the moment, and the plan is to have 33 new cars once it is converted, the headway cannot increase to 5’30”, but will actually go down to about 4’51”, assuming no change in running times.
Steve: You are correct. Because Queen is an ALRV route, the calculation for that line is different from the others in the spreadsheet and I had the fraction upside down. The correct LFLRV headway is 4:51 as you say. I will update the article and the linked file. Thanks for catching this.
And one other thing that leaps right off the page: how come the new fleet, supposedly much more reliable than the current equipment, needs a 20% spare ratio, no better than the 30+ years old fleet it is replacing?
Steve: In the short term, I could understand a high spare ratio as the fleet builds up and teething problems are dealt with, but that should not last forever and a lower spare factor should be possible later in the plan. On the bright side, this gives some headroom to put back those missing cars on Dundas and Carlton.
LikeLike
What kind of speed improvement can we expect with all-door loading?
Steve: This has not been quantified. The benefits are not just from all-door loading, but from the absence of a stairway, and a much simpler process of loading and storing large objects such as baby carriages and shopping buggies which will no longer impede movement through the entire car. For stops with only a few people boarding it won’t make much difference because most of the “lost” time goes to stopping and starting the vehicle.
LikeLike
Regarding the farside eastbound stop at Ronces along Queen: The existing nearside westbound stop has the same geometry/layout as the conceptual farside stop in the opposite direction would have, except for the relationship with the intersection being at opposite ends of the LRV. If the existing nearside westbound stop isn’t problematic (and it is the same as most stops along 20m-wide Queen), what makes the conceptual farside eastbound stop a problem? Especially considering that farside is generally considered preferable for signal priority.
Steve: The westbound stop approaches the traffic signal, and traffic queues behind the streetcar. A farside stop will block eastbound traffic on Queen crossing Roncesvalles especially if the sidewalk is built out in the manner of stops on Roncesvalles. This could also require eastbound streetcars (in a reserved lane) to merge with eastbound traffic unless, of course, a “transit priority phase” prevented the streetcars from moving most of the time.
LikeLike
The problem associated with traffic conflict management for getting LRVs across Ronces eastbound from a reserved lane to a mixed traffic lane will be there regardless of what side of Ronces the streetcar stop is placed.
Steve: But it will be simpler if the LRV does not stop immediately after crossing. Also, if the sidewalk is extended to make a Ronces style carstop, it will not be possible to improve capacity eastbound simply by keeping the area clear of parked cars for some distance as a merge zone.
LikeLike
I suspect that on routes where bunching of cars is presently a chronic problem, that the larger vehicles will merely have the effect of a pair of bunched cars of our present fleet. I am curious if the TTC has done any headway estimation by modelling (an average) number of bunched cars as a single, higher capacity car, and come up with the forecast headways with the new fleet. Whether the TTC will be able to prevent bunching with the new fleet is another question. The effect that might have on new headways would be disastrous.
Presently, while bunching of cars is frustrating when excess capacity appears in clumps, it is sometimes necessary when frequent service results in full single cars at transfer points, and passengers left at the curb waiting for the next – almost full – car, when it comes at its forecast headway.
Steve: From my analysis of TTC vehicle tracking data, one of the major sources of bunching is that vehicles do not leave termini evenly spaced. This situation compounds as they move down the line. A similar issue exists at short turns where there appears to be little or no attempt to manage the re-entry of a car. In particular, it is not uncommon for an empty car to come out right behind a full one rather than in front of it where some good might be gained. As for bus routes, there are operating rules that interfere with “leapfrogging” where buses skip by each other avoiding every stop. This does not work on a route with branches (a waiting passenger might be served by the “wrong” bus), but those are exceptions.
LikeLike
@Karl Junkin
I’m not sure I understand your concern. Streetcar stops without safety islands are normally nearside to prevent traffic queue through the intersection behind the streetcar. I think the debate earlier in this thread is whether there is even enough room for a safety island easbound east of Roncesvalles.
LikeLike
Toronto Streetcars wrote:
PCCs in Antwerp use pantographs as do trams in Lisbon.
Nothing unusual or special.
LikeLike
Probably a dumb question, but… if vehicles are leaving terminals in bunches, why isn’t a route supervisor stationed there? They seem to hang out more centrally, like at King and John, where it seems like they’re limited to reacting to service problems that have already happened, and (as far as I can tell, although I’m not a qualified route supervisor) able to do little but decide who gets short turned.
Steve: And they hang out in bunches at King & John, usually two, sometimes four, and on one occasion five.
LikeLike
Steve wrote:
What exactly must be done to make the overhead “pantograph-friendly”? And what further changes might occur once the CLRV/ALRV fleet is retired?
It’s been many years since I had the opportunity to view the streetcar action in Toronto. I remember the switches as being NA or SR, what was done 25 years ago that has now become unreliable?
Steve: The hangers and frogs must be changed so that nothing lies below the plane of the wire. Some hangers, particularly on curves, are an inverted “U” that a trolley shoe can slide through, but which would snag a pan. Similarly, older style frogs are not flat and a pan would not slide under them. At some locations, the way span wires are attached they hang very close to if not below the level of the contact wire. All of this must be changed.
Once there are no trolley poles, there is no need for frogs at turnouts, and curves don’t have to be strung with many short segments.
The NA/SR system has not been used since the introduction of ALRVs. It depended on the distance from the front of a car to the point of contact of the trolley pole being more or less constant across the fleet. If an ALRV pulled up to an NA switch, the front of the car would already be on the switch before the pole reached the contactor. The replacement system uses transmitters on the cars and loop antennae in the pavement. However, between inconsistencies in loop behaviour and flaky electronics in the switch controls, the new system was much less reliable than the NA/SR system it replaced. In a stunning example of TTC’s indifference to the streetcar system, this was “fixed” by imposing a stop-and-proceed order on all facing switches (even manual ones) and by taking a lot of the electric switches out of service.
LikeLike
If vehicles are leaving terminals in bunches, the drivers are not being instructed properly, or management doesn’t care if their drivers do what they are told. It shouldn’t take a route supervisor to ensure that every vehicle waits until 1 scheduled headway has passed since the previous vehicle before leaving—drivers have watches, and sufficient education to use them, right?
Of course, management would first have to admit that headway-based management is sometimes appropriate before they could even think the above thoughts.
Steve: The root problem is that a car is considered to be on an acceptable time if it is +/-3 minutes of schedule. This leaves a lot of room for bunching. As for headway-based management, that’s still an internal battle. One big issue is that the vehicle tracking system does not provide “time since the last car” info to the operators and they have no way of knowing what the headway in front of or behind them is beyond what they can see out the window. Indeed, the original pre-GPS version of vehicle location (only recently replaced) was often hopelessly inaccurate and would give wildly erroneous information about a vehicle’s location and direction, and hence its position relative to the schedule.
LikeLike
If anyone from the TTC is watching this, I’ve built an interface that calculates the headway ahead and behind vehicles based on the nextbus data and shows how much adjustment is needed to center the vehicle relative to the ones ahead and below. I only have it working for the Bathurst 7 bus but it’s trivial to add additional routes.
You can see the Bathurst headway monitoring here.
It can also do schedule monitoring .
I know it’s a bit confusing but instructions are here.
Steve: Thanks for showing what can be done with available data. The TTC is decades behind on using information that is right under their noses. Some of their cherished explanations for poor service go up in smoke when we see how vehicles actually behave. Meanwhile, the tools to allow for better service management, including some by the ops themselves, never quite seem to appear.
LikeLike
The other problem with bunching is that the TTC does not use a drop back system – when a car arrives at its destination, the driver gets off and takes a break while another driver gets on and takes off with the streetcar. Then when the next car arrives, the driver from the prvious streetcar takes over. This then keeps going on and on. This speeds up service while giving the driver a chance to take a break.
LikeLike
Ah, yes, the +/- 3 measure. Thanks for the information about vehicle tracking, Steve.
I guess what I find so galling about the parades leaving the terminal is that no fancy technology, or at least no technology fancier than a wristwatch, is required to do what I suggest. Constant headway monitoring would be nice, but surely it would make a huge difference if drivers would at least break up convoys at terminal points.
LikeLike
Amit’s solution is great, but I agree totally with Isaac — the delay of adopting new technology is just another excuse when the bunching-from-terminals problem is entirely solvable with the technology and staff already deployed.
The +/- 3 minutes is a good example of a dangerous KPI: surely the 6-minute range is in place to allow for things that crop up along the route (traffic conditions, bad luck at lights, extra-long dwell time at one stop, etc.). Managing to +/- 3 at a terminal isn’t acceptable, because it’s likely to lead to greater variation further along the route.
LikeLike
Steve,
I know you don’t usually do interviews on here…but I’m wondering if you could use your connections to find a line supervisor or two who would submit to a friendly interview (with the TTC’s permission, and maybe Byford present) to see what their strategy’s and challenge’s are (we know a few are technical). I’d love to know why it takes 4 of them to manage an intersection. I’d love to know how they (and their management) evaluate their performance and successes and failures. I feel like it would be good for the TTC to encourage a public discussion about what can be done to improve their tools. It would be interesting to find out what the union is doing to improve things for them (standing out in the cold has gotta suck). Maybe a bit about what their hours are like, how many of them there are and how they decide where to be and how long they just are standing there. I’d also love to hear about what their jobs are like when their are emergencies – I assume they would be moved in those situations…some information about the differences between managing streetcars and busses and the subway…this sort of thing would be very interesting.
LikeLike
George Bell, that is a wonderful suggestion. Perhaps Mr. Byford would be interested as well, to give the public a chance to get to know the different jobs (and people) involved in keeping the TTC running.
I’ve read examples of how the TTC is changing the way it reaches out to the public and shares information. Byford is bringing in Station Managers (including managers responsible for groups of stations — a job he once had in London) and would probably want an opportunity to tell more people about those new positions as well.
I’m not necessarily saying that Steve should be posting interviews with everyone here, but let’s start with the Route Supervisors (since line management seems to be a big issue) and see how it goes from there.
Cheers, Moaz
Steve: When we reach the point that the TTC actually spends money to upgrade its vehicle monitoring systems, and completes the rollout of intelligent devices for field use (now in trial), and we start to see better active management than simply telling operators to short turn so they will be back on time, then we have room for discussion.
LikeLike
Steve said:
If, as it seems, vehicles leave the terminals in bunches it hardly takes much technology to get operators to depart on time (NOT + or – 3 minutes). Amit’s website is amazing and is surely the kind of thing someone at the TTC ought to already have spoken to him about. I would not bet that anyone has though! Does Chris Upfold read this blog?
Steve: Yes, he does.
LikeLike
I have noticed in parts of the system that old (new?) overhead hangers are being left hanging on the span wire next to the contact wire. I am not sure why this is but it seems rather odd.
Steve: These are locations where new overhead was (is) being installed. The second set of hangers was (will be) used for the new wire, although I have noticed in some cases only temporarily.
LikeLike
Should dollars magically become available sometime down the road for additional LFLRVs, what is the combined theoretical LFLRV yard capacity (including the Leslie Barns)? Will there be much room for additional fleet expansion without expanding existing facilities?
Steve: I think that there is room for about 240.
LikeLike
Somewhere down the road, won’t Harvey Shops be redundant with the opening of Leslie Barns? Are there any plans for it? If not, there’s additional storage room.
Steve: I don’t think there have been thoughts that far down the road. Harvey Shops does more than just streetcar repairs, and so it’s not just a matter of closing the doors and knocking down the building. But yes, a fourth centrally located carhouse could be useful if there is further fleet expansion. That’s a long way off. I would just like to see the new cars running reliably on the streets carrying happy passengers.
LikeLike
Went for a walk down Adelaide today, and the pole planting party has continued west (At least as far west as Peter street) Interestingly, the poles are substantially taller than the poles used for conventional spanwire construction, and only on the south side. Makes me wonder if they invisage a cantilevered overhead support for a (potential) eastbound track rather than conventional overhead construction… I suspect the reason the work isn’t yet in the plans is the high degree of time uncertainty associated with all of the on-street occupation by condo developers, though this has cleared up somewhat east of University.
LikeLike
It has only been a few years but cracks and holes have already started to form in several spots around the tracks on the Queen Street bridge over the Don. There are currently slow orders at both ends of the bridge.
It’s purely speculation on my part but I wonder how much of this is due to vibrations caused by poor maintenance of track profiles. I am sure that having a track junction on the bridge does little to help matters. It might make a case to bring back proper rail grinding vehicles.
This is something the TTC needs to get on top of because they can’t go tearing up bridges every 5 years. Here’s hoping the fresh coat of paint on the Gerrard bridge fares a little better than its southern counterpart.
Steve: The problem is with the expansion joints. The ones on the east end were repaired, I thought, last summer, but the slow order signs never disappeared. The ones on the King branch are waiting for better weather.
LikeLike