Toronto’s Budget Committee has asked staff for many briefing notes on details behind various programs. Among these requests was for TTC to detail the level of service that would be operated on all streetcar routes after the 204 new Flexitys have been delivered, and how this would be improved with the addition of a proposed 60 car order.
The TTC has responded with a report that details how the cars would be used.
Long–Term Peak Headway Projections for Streetcar Routes
With the initial 204 car order:
- Peak headways would widen by varying degrees on all routes except 501 Queen where existing AM peak frequency would be maintained using the larger cars. The biggest change would be on the 502/503 services changing from 12′ to 14’30” on each of the routes which, in theory, provide a blended service.
- Capacity increases in the AM peak would be greater than in the PM peak.
With the additional 60 cars:
- Peak headways would return roughly to current levels. Capacity, relative to today’s service, would be considerably higher than today, except anomalously, on 501 Queen.
The report notes that additional cars will be needed for routes to serve the waterfront, but gives no indication of the service levels or fleet requirements for these routes. Because the report shows only headways for existing routes, not vehicle allocations, it is unclear how many of the 60 cars go for service improvement and how many for new routes. The percentage improvements on existing routes are high enough that it is possible that cars have been double-counted for this purpose. I will follow up on that issue with the TTC.
Updated at 12:22 pm: At Budget Committee, Andy Byford confirmed that the only “expansion” covered by the 60 cars is the Cherry Street spur south from King Street.
Updated at 3:00 pm: The TTC has confirmed that all of the 60 additional cars would be allocated to “legacy” routes with none reserved for expansion. As to the spare ratio they would design for:
“Spare ratio of 18%. We expect that that we will be able to reduce that when the fleet is settled in and we have confidence in the performance and reliability but, until then, this is our going-in assumption.” [Email from Mitch Stambler, TTC]
The ratio for the streetcar fleet today is about 25% (not including cars out of service due to cold weather) with roughly 200 of the 247 in the fleet scheduled for the AM peak.
Steve Munro 
Though I know that the Bombardier contract says that the extra 60 streetcars have to be ordered in 2015 to get the same price, we all know that Bombardier has been VERY (in fact, embarrassingly) late in actually filling the first order. Is there any new date by which we need to place this extra order? Will this be one part of any ‘penalty fee’ talk that is, or should be, happening.
Steve: The actual trigger is the delivery of the 60th car in the original order which is now expected sometime in 2016. Penalty fees are a separate issue.
LikeLike
It seems blatantly irresponsible for council not to follow through with the 60 car addon. If I recall correctly, they’ve basically said “Yeah…so, talk to us again in 5 years once the current order is completed”, right? I’d rather see them say yes right now, and spend these few years gradually funding the addon. Hopefully late penalties we can get out of Bombardier are significant enough to pay for, at least, most of the cost so that council is more likely to support it…
LikeLike
What is missing from the TTC submission is any indication of how many of the peak streetcars are presently at or above the 77 & 108 loading standards. I could easily see members of the budget committee and council looking at the summary and saying 50% increase in capacity is plenty, why go to 80%?
LikeLike
What’s the latest word on how they’re going to serve that Cherry Street extension once it opens? Will alternate 504 King cars be signed 504 CHERRY? Or will we get a dedicated route to either Broadview station, or operating through the downtown and looping at Spadina, Adelaide and Charlotte?
Steve: No word. I don’t think a car running just to Spadina makes much sense, and the last thing we need is more cars trying to make turns around Charlotte Loop. At least to Dufferin Loop, although most of them would short turn at Bathurst or Spadina 😉
LikeLike
Report:
The question I would have, is: At peak, will this be enough? The capacity increase on King at 50% might almost cover it, at 60% (extra cars) well I think that might be ok, depending on how much more development there is along the route without more support. If they do not do the East Bayfront, and increase service on Queen by more than 20% or 30% west of core I have no problem seeing these cars past their loading standard even as they go into service. I suspect that headway on King will still be too long at 2 minutes by 2019 and Queen will need to be at or below the 3:53 (maybe 2:45) if they allow development to continue at its current rate west of the core (assuming it continues to the north). Given where we are delivery wise assuming that Bombardier gets moving, we are looking at the last 60 in what 2021? so 6 years from now? How large is the pent-up demand, and how fast will that area go into high gear in development once there capacity on the cars again.
Steve: I don’t think that the TTC and City have done a detailed study of future demand, and suspect that the 60+ cars scenario is an arbitrary return to current allocations (ie a 264 car Flexity fleet replacing a 247 car CLRV/ALRV fleet).
LikeLike
Is there any thought for increased headway on the 502 Downtowner during mid-day from 16′ to be more inline with its evening or weekend 22 Coxwell’s headways? 12′ would be just okay, while 10′ would be better.
Steve: That’s a whole separate subject. The daytime headway on the 502 is actually much worse because of irregular service and short turns.
LikeLike
There really should be some land use projections done, and some serious modelling in order to look at the demand through the area. If we are to assume that the east and west areas should absorb growth as it spreads, it would not be hard to imagine in the next decade another 40k residents being added in the area immediately north of Liberty Village, plus whatever redevelopment in and around the Ex, and say 80k east of the core, as the area either side of the Don, and just beyond is redeveloped. These residents’ trips would of course have to be added to the streetcar network, or as cyclists and pedestrians. It does not seem like much in the context of the entire city, however, adding say an extra 120k residents (am I being too conservative Steve?) in the area from say the Park in the west to Pape in the east, would have a dramatic impact on a transit system that will need to absorb the increased employment happening in the same area and has no spare road capacity.
I have no problem seeing an East Bayfront LRT running a 2 minute headway, and being full even with the new cars. This on top of a King and Queen Line running similar headway.
LikeLike
Will the Cherry St line extend south of the railway tracks?
Steve: This is shown in future plans, but there is no budget yet for adding a second span under the rail corridor. The existing one isn’t wide enough for the road plus streetcar lanes, cycling, etc.
LikeLike
So … no allocation either for a potential return of the 507 rerouted to Dundas West Stn. then.
Steve: The TTC has yet to acknowledge that this might be a good idea.
LikeLike
Steve out of curiosity, why is the headway so long on the 502 and 503, and would these not make more sense as a single route into the core with a shorter headway??. I know they do not go to exactly the same place but the would not running more service on Queen route make more sense? If you ran the one on a 5 minute headway (regular), would that not make better overall service? If it intersects a very frequent King streetcar, would not a transfer at a decent shelter not work as well for those headed to (from) a destination on King?
Steve: Yes, they would make more sense as one route. Once upon a time, both of them were very frequent, but with the gradual decline of service, having two routes leads to more irregular service than if there were one.
LikeLike
Perhaps we could kill two birds with one stone, and run the Cherry service over the 503 loop south down Church, west along Wellington, north on York, and back east on King. Then 502 could simply run to McCaul during peak at a higher frequency (though I suppose that running the Cherry service on Wellington doesn’t mean that one has to eliminate the current 503 service).
LikeLike
I remember when the old Kingston Road streetcar ran along King Street to loop inside the Roncesvalles streetcar yard during rush hours. Basically, allowing for extra service along King Street.
There still was a Downtowner back then, to either the McCaul loop or Bathurst Station, depending upon the time of day.
Unfortunately, with our current streetcar shortage, basically since the CLRV’s and ALRV’s replaced the PCC’s, it will be a while before we can even think about returning to those days of better service.
LikeLike
Good luck with that. The “Not-invented-here” syndrome will enter into this. Note that the redesigned Dundas West Station still has only two tracks/platforms for streetcars. No room to add third track that might be used for Long Branch route and/or College.
LikeLike
How does the proposed spare ratio at 18% compare to the spare ratio that the TTC had when the CLRV order was completed?
Cheers, Moaz
Steve: This gets tricky because of the varying number of PCCs remaining active over this period. A better reference point is 1990 after all of the CLRV/ALRV fleet had been delivered, and most of the rebuilt PCCs were in service. The November 1990 and January 2015 fleets and service allocations by route were:
Route 1990 2015 511 Bathurst 16 12 506 Carlton 38 32 502 Downtowner 10 7 505 Dundas 30 19 509 Harbourfront 5 7 504 King 44 35 CLRV 7 ALRV 20 buses 503 Kingston Rd Tripper 9 6 507 Long Branch 7 508 Lake Shore 3 501 Queen 34 ALRV 31 ALRV 5 CLRV 512 St. Clair 29 25 510 Spadina 12 CLRV 3 LFLRV Total 222 204 Fleet CLRV 196 195 ALRV 52 52 PCC 12 LFLRV 3 Total 260 250 Spares 38 46 Spare % 17.1 22.5The spare factor is calculated as the number of spare cars relative to scheduled service.
1990 was just before the recession and ridership crash, and the streetcar system has never returned to service levels seen at that time, in part because the Spadina line soaked up the then surplus equipment leaving nothing for growth (let alone declining reliability) when it opened in 1997.
Today the TTC is unable to field enough streetcars to meet scheduled service. It will be interesting to see whether the fleet stages a miraculous recovery from cold weather, or if bus substitutions on some routes continue indefinitely pending arrival of more Flexitys.
LikeLike
Perhaps someone could “suggest” this to the TTC before they start digging up the station again in the near future.
Steve: The scheme to rebuild Dundas West comes from Metrolinx as part of their “mobility hub” studies, but I doubt we will see it actually built. Also, regardless of which layout is used, I don’t think you can fit a third track into the available land. Any “507” service would share the “504” platform as both services would run south on Roncesvalles.
LikeLike
Steve looking at the 1990 comparison to me I find the degree of reduction in service on Bathurst & Dundas shocking and the small size of the increase on Harbourfront equally surprising. How much of the Harbourfront is being carried on the inner portion by Spadina cars? I remember riding Bathurst cars in the late 1980s and they always seemed full, what happened?
Steve: When Bathurst was converted to ALRV operation, the headways really fell apart (similar to Queen) and riding took a big nosedive (as on Queen). That level of service then becomes the “new normal” for the route. Very sad. Dundas has suffered cutbacks over the years, and now there are no spare cars for improvement. Again it’s a new base level from which the line will have to grow again.
The line that really held its own was King, but that’s thanks to all of the new demand both east and west of the core. That sort of change will come to the routes further north, first to Queen, and the TTC is really not prepared for this.
LikeLike
Hey, nice to see a “streetcars, streetcars, streetcars” thread on this blog instead of “subways, subways, subways” for a change!
Though I live in Caledon in the Bolton area, I regularly have business in Toronto, I am a Toronto native, and I endeavour to park the car (for free) & use public transit. My favourite mode of public transit is the streetcar, and not subway or bus. One reason is comfort, another is convenience.
Btw, the idea of connecting Caledon (Bolton) to either the airport or Toronto (Rexdale) by LRT is quite popular here (but not more highways), as said in public meetings about Caledon’s future. Bolton wants public transit!
Here is an observation – Most of the streetcar lines run east-west. They tend to be packed at crush capacity, say, between Ossington and Parliament. Further out, the passenger load is certainly lighter. This is mid-day I am talking about, not peak period!
So, I am thinking that if 80% of the passengers are only interested in travelling through the core area, there ought to be streetcar routes dedicated to serving this particular demand.
For example, a Queen car ought to be able to short turn at Dufferin, without having to go all the distance to Roncesvalles or Humber Loop, or even Brown’s Line. A College car ought to turn at Ossingon. A Queen car at Shaw and back up on Bathurst.
Sixty years ago or so, the streetcar network was a lot more extensive than now. There were streetcars on Ossington, Dovercourt, Dupont, Harbord, Hallam, Lansdowne, Dundas to Runnymede, Rogers, St. Clair East & Mount Pleasant, Eglinton, Bathurst between Bloor & St. Clair, Coxwell, Parliament, Bay, Church, and more.
Of course, the two main streetcar lines, Yonge Street and Bloor-Danforth, were replaced by the subway 60 and 50 years ago, respectively.
Streetcars were the preferred mode of transit back when the urban landscape was not dominated by vehicles using the internal combustion engine. If you look at old photographs of Toronto from the 1920’s and earlier, there were very few cars. Installing streetcar tracks on a road meant huge progress – no more bumpy rides in the mud, an affordable way to get from A to B.
So, the real reason why we have a streetcar network in Toronto, as shrunken as it is, is because it is a relic of the times when streetcars dominated and automobiles did not. Except, that over the decades, the automobile is starting to lose its current dominance. Throughout the core area, increased density makes car usage inefficient, and that is reflected in increasing restrictions and more expensive parking. Lots of people in the downtown do not have cars. However, they can rent one for a weekend out-of-town trip when desired. Or, occasionally, take a taxi. This is very cost efficient.
So, the trouble with the streetcar network as it is now, is that the routes do not match demand very well. We need lots more streetcar capacity in the central area. A big problem is that the tracks from yore are gone.
I recommend restoring streetcars to Harbord, Bay, Church, Dupont, and Ossington, for example, for tripper service.
The streetcars need more flexibility to maneuver. For example, a Harbord car could go either north or south at Ossington, depending on passengers’ need. Short turns could be more “logical” in execution, less arbitrary.
But, I also recommend putting streetcars back on Bloor and Yonge. On Bloor, a line from Castle Frank to Dundas West. And another east on Danforth to Main St. This is not meant to replace the subway service in any way, of course not. Rather, streetcars are on-street transit, able to stop at many places in between the subway stops. As well, it would add to the network. Toronto has matured and grown since the two main subway lines opened, it has been a half century. On Yonge, I bet a streetcar running on the surface would be a runaway success. Count me in!
LikeLiked by 1 person
So what is the compelling argument that would suggest that TTC should exercise the option to purchase 60 additional train sets? No comments posted to date [resent a compelling reason.
I would say yes. But.
Steve: The compelling arguments are quite simple. The city foresees substantial growth in population along the streetcar lines. Already there are new midrise building going up along the eastern parts of the Queen, Carlton and Kingston Road lines. Queen West is about to see an explosion in population as the towers between Dovercourt and Gladstone are occupied, and development won’t stop there. King is already overloaded. More development is in the works.
Then there is the small matter of the eastern waterfront which will have a population considerably larger than Liberty Village. The TTC’s report does not show any cars allocated to services in the waterfront, but they will be needed.
Basically, the streetcar system has seen no investment in new equipment for two decades, but it is carrying far more riders, a trend that will continue.
LikeLike
Hooray! Someone started a sandbox that I want to play in.
I’ve been thinking about the East Bayfront line and where it might be expanded to go, and one idea that occurred to me was to run an extension up Parliament to Bloor. At it’s north end the line could enter a tunnel onto the lower deck of the Rosedale Ravine section of the Viaduct, which would take it right into Castle Frank alongside the subway tracks. Some cars could run a full route onto Harbourfront and Spadina giving an inner U comparable to the outer U lines on King and Dundas. It might even offer a small amount of relief at Yonge/Bloor for people who see it as an alternative route to points on the east side of downtown.
I know the chief argument against Parliament is the narrowness of the street, but it can’t be any worse than parts of King or Bathurst.
Steve: Actually there are several problems, not the least of which is how close Castle Frank Station is to Bloor-Yonge. Have you ever tried to get on a train there eastbound in the PM peak? Similarly, it is difficult to board westbound in the AM at Broadview, and offloading people at Castle Frank will not change this. The ramp down on Parliament would be challenging given the narrowness of the street and the fact that expansion is almost impossible with St. James Cemetery on one side and the underground structures (parking) of St. James Town on the other. Struturally, merging in a streetcar loop with the existing station would be quite a challenge. It would have to be at the mezzanine level to avoid conflicting with the subway itself, and this would likely run into utilities under Bloor Street.
LikeLike
Steve would it not be more likely that the East Bayfront become effectively a Waterfront East LRT? Running eventually Lakeshore to say Kew Beach area? I would think that this area could really use meaningful improved connection if it is also going to continue to develop, and can Queen Street reasonably carry the load from this area if it does develop any sort of density there or beyond? I have a hard time not seeing real redevelopment bring substantial density to the Leslie and Commissioners area. The quick access to the core that would be there with an LRT along with the outdoor park environment of the Spit area would be great, I would think this would be a desirable area if it had an LRT through it. Somehow, I think this area seems a natural, rather than turning north at Parliament. Besides it allows this service to extend beyond the “Downtown” that way. Although I really hope that it would redevelop as a human scale area that was bike friendly & walkable with lots of local shops, and not just a Superstore.
Is there enough room for 4 lanes, nature and bike trail and LRT in the current road allowances & city controlled space to get to and beyond Coxwell Steve?
Steve: The long range plans show track east on Commissioners from Cherry to Leslie (at the new carhouse). I don’t think it makes sense to continue east via Lake Shore or Eastern considering that the east end of Queen is fairly uncongested. However, this link will not be built for a few decades given the likely timeframe of buildout in the Port Lands. There are far more important things to deal with such as just getting all of the Queen cars to Neville Loop on a reliable headway.
Also, I’m not sure that the running time would be all that different given the less direct route to the core a line through the Port Lands would represent.
We must avoid getting bogged down with streetcar fantasy maps just as much as with subway schemes, and deal with problems the system has today.
LikeLike
Church cannot provide enough demand to warrant a bus service so the odds of it supporting a streetcar are slim. Harbord is very close to Bloor and the Wellesley Bus covers it with room to spare. Do you want to actually run the old Harbord Route on its 11 streets between Lansdowne and Davenport to Pape and Danforth or just on Harbord St? The old Dupont route ran from Christie and Dupont to Bay and Queens Quay.
Before the Yonge Subway opened in 1954 the TTC had a number of “dipper routes” that started off in the north east or north west and worked their way towards Yonge Street and downtown without actually going on Yonge. The Bay car ran from St. Clair and Lansdowne over to Avenue Road to Davenport then over to Bay and then south to provide a route downtown without going on Yonge. Harbord ran from St. Clair and Old Weston Rd. and ended up on Dundas through the downtown. Dupont went from Dupont and Christie down Bay st. The Dundas car went from Runnymede and Dundas to old City Hall. Also the Danforth Tripper ran along the Danforth and Bloor St to Church and then went Downtown.
When the Yonge Subway opened it took over all the loading from Yonge Street as well as the Bay car and the Danforth Tripper. With the change in travel patterns created by the Yonge Subway and then the University and Bloor Danforth the chances of a bus route on most of your routes carrying significantly, let alone a street car route, are slim to none.
Steve: Much more generally, any new surface route must have demand feeding it from somewhere, and ideally with a destination as well. Many of the old routes were directly fed by suburban bus lines (Dundas, Harbord) or they linked residential communities with industrial areas (notably in what is now a large housing neighbourhood around King and Shaw). Just because a route worked in 1950 does not mean that it will work today.
Are you suggesting that a “vote” be held when the car gets to Ossington to see if more passenger want to go north or south? The advantage of running a grid system is that you reduce the headways on all the lines so that while you may force a transfer, the wait for the transfer will be less than the wait for the particular branch that you want.
Steve: Another problem with the premise is the claim that demand falls off on the outer parts of the routes. This is not entirely true depending on which route one considers. King has the most notably complex demand structure and it carries considerable traffic counter peak, as well as operating as both a Roncesvalles car and a Broadview car connecting the ends of King Street to the subway. For example, an outbound King car starts to pick up passengers again west of Dufferin. The TTC’s short turning policy completely ignores this pattern. Additional developments along the other east-west routes will create a similar, although I hope not quite so extreme, situation.
“But, I also recommend putting streetcars back on Bloor and Yonge. On Bloor, a line from Castle Frank to Dundas West. And another east on Danforth to Main St. This is not meant to replace the subway service in any way, of course not. Rather, streetcars are on-street transit, able to stop at many places in between the subway stops. As well, it would add to the network. Toronto has matured and grown since the two main subway lines opened, it has been a half century. On Yonge, I bet a streetcar running on the surface would be a runaway success. Count me in!”
Like the subways under them, the street car routes on Yonge (along with Bay and Church) and on Bloor-Danforth did not get the majority of their passengers from “walk ins” but rather from people who transferred to the streetcars from other lines. Yonge really has a problem generating enough ridership to keep its operation viable and while Bloor Danforth has never had a parallel bus service and I am willing to bet it would be underwhelmed by passenger demand.
The subway and the growth of the suburban areas has changed the TTC from a downtown oriented to service to more of a grid system. While it may be true the the downtown core is the destination of a large plurality if trips, the TTC is no longer a grid system with a large downtown radial system superimposed on top of it. While I love street cars I do not think your routes would help the system.
LikeLike
All true. But personally I think that I’d like to see (post QQE line) funding to extend trackage on Cherry down to Queen’s Quay, and connect the Exhibition and Dufferin loops. Then a nice streetcar loop on King, Dufferin, Fleet, Bathurst, Queen’s Quay and Cherry.
As for the QQE line, I just want it to be built and get revenue service as far as the proposed extension of Broadview Ave through the Lever lands. That development, if done properly (with streetcar & buses, DRL and GO RER for local, medium and regional demand) will really transform the east side of Toronto.
Cheers, Moaz
Steve: The connection down Cherry is far more important than the link through the CNE grounds. They should not be conflated as one project or priority.
LikeLike
I’ve often though that the only way a modified 507 works is if a portion of streetcar service on King is turned back at Roncesvalles. This way the presence of the 507 keeps up a similar level of service on Roncesvalles up to Dundas West station, while service on King sees improved frequencies.
A slightly different option would be to create a new route that ran from Roncesvalles (or Humber?) to Union Station via King-and-Dufferin-Fleet-Bathurst-Queen’s Quay. Of course this would require a track connection between Exhibition and Dufferin loops (and a few other things … and there would be challenges) but if combined with a new 507, it would be a way of improving service on King without having to turn some 504 streetcars back. Because … short turns.
On … I’ll now flip the fantasy streetcar planning to “off.”
Cheers, Moaz
Steve: I find it odd that a GO train every now and then is considered a huge improvement for people in Liberty Village, but a slightly longer walk to Exhibition Loop and the 509 (which would provide the inner end of the service you describe) is not. I think that the whole idea that Liberty Village can somehow be served by a route south of the rail corridor is oversold because much of the population must walk too far to access it.
LikeLike
Maybe the problem is the wording of the contract. The new streetcars were to be delivered between 2014 and 2018 +/-. Maybe the option to but 60 more should have been delayed to 2018. This would have helped the cash flow., etc.
Steve: The contract says “when 60 cars are here, then Toronto must exercise the option”. That is unlikely to happen until sometime in 2016. 2018 would probably be too late based on original dates because component supply contracts would be timed to expire about a year before the final car, possibly sooner depending on how Bombardier is handling its own procurement. You can’t extend an order if the supply chains have been cut.
LikeLike
Well Steve, to be honest getting all the way to Leslie is huge. However, I wonder about the build out of the Port Lands. I suspect that as long as there exists really good transit, the premium that can be had from the Port Lands and the tax impact of having that land in best use, will be substantial. I do not believe in TIF planning, but this is one of the places that transit first planning, will likely yield a nice bonus to the tax base, and more than pay the cost of construction. However, I would still not support the use of TIF on principal.
LikeLike
In this extremely cold weather, can the TTC get sued if a bus or streetcar fails to show up within 10 or so minutes?
Steve: No. There is no contractual relationship by which the TTC has guaranteed any level of service, or service at all.
LikeLike
509 is a rather slow route, despite running in a dedicated ROW. I guess, that’s because of frequent stops and the time spent getting through the formidable Bathurst / Fleet / Lakeshore interchange.
509 works for trips to the Exhibition, and for residents of the waterfront condos. But if you travel from the Liberty Village, mixed-traffic 504 streetcar is usually a faster way of getting to Yonge, rather than walking 10+ min to the 509 loop and then taking 509. I personally tried both options several times, although that was a few years ago.
GO RER service, every 15 min at the Exhibition station, is potentially a better option for Liberty Village. A 10+ min walk is still involved, but after that you should be able to get to Union in 5-7 min.
Of course, a SmartTrack / GO RER / UPX station on the Weston rail line would be even better; located right in the middle of the neighborhood, and providing fast connections to Union, parts of Etobicoke, and the airport.
Steve: The station, if any, on the Weston line is likely to be closer to Queen Street which is of little use to much of Liberty Village. As for GO RER, it will be interesting to see what sort of fare remains in place for this trip as people at other stops further out such as Mimico could reasonably ask for a discount too.
LikeLike
The Transportation Research Board’s Transit Co-operative Research Project “Modernizing Street Car Service In Melbourne and Toronto” says that one way of improving service on King was to use MU operation; this is from around 2002 when only single units were operating on King.
Since the LFLRVs are twice the length of the CLRVs running them is equivalent to running MUed CLRVs. According to the report using MU service reduces the headway variation and improves the benefit of TSP (Transit Signal Priority.) It also states that the minimum useful headway has to be twice the traffic signal cycle time. While I am unsure of the current value the cycle time used to be 80 seconds so running a 120 second headway does not work well. But if it is doubled to 240 seconds this is 3 times the cycle time which allows TSP to work. Even going to 160 second headways, two cycle times, would improve service and the functionality of TSP.
Steve: There is a small problem. The King car needs more service than a scheme designed to optimize TSP would give. The proposed headways with new cars are under 210 seconds.
Also, some of the study work on Toronto that is behind this claim looked at a small section of King in the core, and did not consider the interaction of bidirectional service which further complicates signal timings. It also assumed that it would be possible for the “input” service coming into the study area could be maintained at a reasonably well-behaved headway. We know this is not how the TTC actually operates.
LikeLiked by 1 person
Robert, I am most honoured that you chose to spend time to answer my post. Usually I am totally ignored.
However, I will stand by what I wrote. I take the TTC every week, just like I have for pretty well most of my life, (despite being a non-resident of Toronto) and I know where the service has its shortcomings.
Actual case: I need to go from Rosewell & Craighurst to Carlaw & Danforth, mid-day, mid-week. I am “walking disabled”, meaning that I cannot make it to Eglinton subway station, already walked a kilometer this far. There’s a bus stop at Craighurst & Yonge, the bus goes to Davisville. It is supposed to come every 15 minutes, but usually it is more like every 25 minutes, sometimes never. Subway rumbles underneath. Finally on the subway, do I get off at Broadview or do I get off at Pape? Another bunch of walking. THEN, I need to make the same trip in reverse.
Another case: Very often, I board the College car at the High Park Loop, I am the first passenger. I watch as a good 80% of passengers put something into the fare box, most of the rest flash their Metropasses. College is not a route which depends on feeder routes. Rather, it is the local transit, it is mostly walk-in. Nearing the core, the streetcar is packed. Again, same story on the return trip, but in reverse, the streetcar is packed to crush, but by the time the car reaches Sterling, it is almost empty.
There are lots of places along Yonge and also Bloor-Danforth which are within 100 meters of a subway station. But, also, there are lots of places that are a fair clip from a station. Would Yonge and Bloor-Danforth passengers/residents benefit from closer spaced stops, if they are local and not on their way to the suburbs?
Christie & Dupont? That was where the streetcar loop was. Now you catch a bus that tosses you around before you can find a seat. Gimme a break.
I mentioned Church only with the thought that the street traffic is lighter than Yonge’s, and, being very close by, and with the ambiance of a streetcar added to the street scene, it ought to work out quite well. And Bay, too. Going to the ferry docks by streetcar was much simpler and likeable in the old days, now it is ugly.
So, the bulk of the streetcars’ demand is at the core, and it is local walk-in. The streetcar is successful not just because it provides a smooth ride within a short wait period, but because it stops every few blocks – not at all like the streets where there is a subway underneath.
Therefore, I recommend expanding that which seems to work successfully – more streetcars on more streets in the centre.
LikeLike
This a problem that the TTC seems uniquely unable to solve, local bus service on top of rapid transit lines. There used to be the Avenue Rd. bus which ran on Yonge from Davisville to Eglinton so it ran on a regular fairly frequent headway. When the TTC finally put the 97 back on Yonge from Steeles to Queens Quay but only with a 30 minute headway. The question becomes is there lousy ridership because of the poor service or lousy service because of poor ridership. Most other jurisdictions run surface routes along rapid transit lines to provide local service.
Four former colleagues of mine are walking disabled, 3 with MS and 1 with MD. I have become a lot more aware of problems facing them and the difficulties created by bad planning. There are no accessible stations on the Danforth between Pape and Main. My friend with MD and I tried to go to a restaurant there on GO and the TTC from Brampton. We got onto the subway with little problem but could not get off at Coxwell. For a parallel bus service to be useful it needs to operate at least every 15 minutes and with chronic cash shortages it will not happen. I won’t get into the problem about doors that hit you in the face when you push the button or buttons hidden behind signs and planters.
Steve: The TTC really needs to look at parallel bus services as a supplement for accessibility, and it needs to be reliable. The precise situation described — going to stops in between stations — is a real challenge. The TTC talks a good story about getting folks off of WheelTrans, but this is low hanging fruit, a service that could be introduced tomorrow if only they had the will and common sense. This would also address the missing elevators, at least until they are installed by that project that has no money while we spend billions on Scarboro’s identity crisis.
That is a problem with buses and inadequate schedules. The side to side motion from pulling into the curb and back out does give an unfortunate ride. I wonder if that contributed to the death of the elderly woman on a TTC bus on the weekend?
True but when I ride it or any other car line there is usually a substantial number of riders who get on with transfers; eliminate these and you have a lot less service. King West in the rush hour is probably an exception to this.
Steve: It depends on the direction of travel. There is a very strong (to the point of passing up would-be riders) demand outbound on King in the AM peak with folks going to Liberty Village and other west end office areas. They are transferring from the subway. Inbound in the PM peak it is difficult to get on an eastbound car west of University because they are packed with inbound riders going to the subway. This is a superb example of the problems that arise when planners and politicians only consider what they perceive to be “peak” travel. King & Bay is not the only destination on the planet.
It could probably be made over like Roncesvalles but a walk from Church to Yonge, say at Gerrard, would add the same distance as walking from the subway at College. Passengers do not seem to be willing to take a surface line when there is a subway.
Getting to the Ferry Docks if you have problems walking is horrible. You either wait for the Bay bus or walk back from Queens Quay Centre, not to mention the horror of getting to bay street at Union with all the construction.
Steve: For reasons that defy understanding, the owners of 10-20 Bay Street waited until after streetcar service resumed at Queens Quay Station to close up the elevator and build another one. The TTC appears to be completely insensitive to private buildings that have entrance agreements with the TTC for station access, but who operate and maintain such access as they see fit rather than the full hours of their stations.
The problem many people have when comparing street cars with buses is the fact they probably don’t ride them. Rail vehicles have inherently smoother rides than rubber tired ones because they do not change lanes and have smoother acceleration. This is why passengers prefer street cars to buses but motorist prefer buses, if we need any transit at all, to street cars because they pull to the curb and get out of their road, supposedly.
I misinterpreted the meaning and reason behind your previous email. Good luck with getting the TTC to put better service above subway lines,
LikeLike
Closer to 120 with the 60 extra cars aren’t they?
Studies and computer simulations show that with headways less than 2 traffic cycles it is impossible for TSP to function efficiently, throw in bi-directional service at close headways and forget it. There is almost no way to keep service on this close headway spaced out evenly. The difference between the headway, 121 seconds, and the signal cycle, 80 – 90 seconds, prevents it.
Now if we were to MU the new cars … hmm!
Steve: The planned headways on King are shown as 2’10” with a 204-car fleet and 2’01” with a 264-car fleet both for the AM peak. For the PM peak, the planned headways are 3’23” and 3’02” respectively. I cited 210 seconds (3’30”) as an upper bound for the range of headways being considered, and 210 is less than 240 (4′) which seems to be the holy grail for TSP geeks (ignoring what this means for riders).
I believe that the idea the traffic signals will manage the service is absurd under our typical operating conditions because there are far too many variables, not the least of which is that streetcars entering the “controlled” aera do not do so on a regular headway. However, signals can be sensitive to transit needs and ensure a streetcar gets or holds a green when it needs it. A big issue here is that there is no way for the operator to say “do it now” with the result that green time is wasted on cars that are held up loading at busy stops, and cars are held needlessly by red signals.
Another important distinction is between major and minor cross streets, and those that have or don’t have car stops. TSP is a very different concept at a location where there is no stop, or where the stop is farside (e.g. King & Berkeley westbound, King & Princess). There was a lengthy thread here about getting rid of “unnecessary” transit stops, but the fact that not all intersections have the same requirements was ignored.
The TSP study that has been referred to dealt with central King where all of the stops are nearside, and all of the signals have stops associated with them.
Just because a route dips below 240 second / 4 minutes as a headway does not mean that TSP has no role on that route, especially at the quickly proliferating extra traffic signals that have replaced pedestrian crossings remote from transit stops.
LikeLike
I would hope that we could target the 160 second headway with additional cars sent out to clear the backlog. Say every 3 or 4th headway send out a car 1/2 way through it. That would allow this to start to have a chance if the TTC were to actually dispatch in a timely manner (should be one of the top priorities for TTC management). I would also hope that Toronto would move to a conditional predictive TSP. The only vehicles that really need to get substantial additional help from TSP are those that are actually running behind headway. So give the vehicle that is catching up with the one in front a red, the one that is falling behind the green where possible. It will not work perfectly, and there will be issues with the reverse direction, but well nothing is perfect. A really well designed TSP system coordinated with route information and resolved vehicle location information should be able to improve things (not fix).
Of course you could always really set off a storm with drivers and suggest running the new vehicles as multi unit trains. That would be interesting on King Street seeing 60 meter long trains. Give them the consolation that yes, but it is only once every 4 minutes – and so what if we block a few side streets.
Steve: Please see my reply to a previous comment by Robert Wightman.
Sending out extra cars now and then is nice in theory, but they have to come from somewhere, and the places this might be done are limited. They also, of course, have to actually “split gaps” properly, something that rarely happens on the TTC because supervisors give cars their “due out” times long before they actually will re-enter service, and it’s pot luck if they actually fill gaps properly.
LikeLike
I have mixed feelings on this one. I agree that at least 60 more streetcars are essential to growing the service to meet demand. However, I disagree with giving extra business to Bombardier. Bombardier hasn’t been able to meet their obligations under the current contract; why should the TTC extend that contract when they’ve demonstrated incompetence?
Steve: The decision needs to be made after Bombardier has re-established their credibility. The contract does not require the option to be exercised until the 60th car is in Toronto. By that time, we should know whether Bombardier is worth giving an extended contract.
LikeLike
What route number would the Cherry service be given? Or will it be a branch of the 504?
Steve: There are too many permutations of what may happen with Cherry Street over coming years to say. In the short term, yes, likely a branch of 504, but longer term, if the track is connected under the railway to a revised Queens Quay, Lake Shore, Cherry intersection as currently proposed, then there are many options depending on where the service goes. Stir in a connection south via Broadview past the Lever/Great Gulf site, and track on Commissioners linking to Leslie Barns, and there are endless possibilities. At one time the TTC produced a map showing various routes with numbers, but I wouldn’t believe a word of it beyond simply being an example of what might happen. Most importantly, the proposed Broadview link did not exist when that map was drawn, and it substantially revises how one might design transit routes for the area.
LikeLike
Yes, the first and most important thin that needs to be fixed really is management of the line, and actually working a lot harder to get the cars dispatched properly. For the idea of having a car split the headway it would have to be done with huge regularity, ie the car is actually part of the route, and there is enough recovery time allowed in the route, that the car is there waiting to be dispatched before it is required (ie the driver has had a chance to take a bathroom break and other required time).
Also for it to have a snowball’s chance in hell, the “controlled area” would need to be the entire route. Practically, a very good, sophisticated conditional system with communication to the car, full awareness of its position, and whether it has or not loaded, can at best help, not fix problems on the route. It needs to start with a real attempt at dispatch, and needs to include decent traffic enforcement as well. By the way, I would love to see the King car in Multiple Unit operation especially with the new cars, just for the visual.
LikeLike
Hi Steve
Does the TTC have any cost per passenger figures for the streetcar system with the new cars? Do they have any with the existing fleet?
Thanks
Steve: This gets tricky because it depends on how you allocate shared system costs to routes, and how you count “passengers”.
Based on the summary of costs and ridership per route for 2012, the cost per boarding was almost identical: $2.23 for buses and $2.19 for streetcars. This figure is arrived at by summing the daily ridership and assigned daily costs by mode. Note that this does not include capital costs.
Having said that, a “passenger” counts as a “boarding” every time they transfer (except within the subway). Therefore, the cost per trip is higher than the cost per boarding, and will be a mix of bus, streetcar and subway costs depending on which routes are involved. There is also the challenge of which proportion of a trip occurs on each mode as measured by distance or by time. One can concoct many formulae.
For what it’s worth, I believe that back when such things were reported, the proportion of “one seat rides” on streetcars was higher than on buses, but this is a function of route location, not of mode. Most bus routes feed the subway whereas many streetcar riders don’t have to use the subway to complete their journey.
Finally, cost comparisons can be misleading because one cannot presume that simply replacing streetcars with buses will have no effect on cost based on those two numbers. For starters, the bus network operates at a higher average speed (19.7 km/h versus 14.0 km/h) again as a function of the routes served. This dilutes the cost per km of the operator (who is paid by the hour), and then we have to get into the question of how many vehicle km the typical rider consumes. We also know historically that buses on busy streetcar routes tend to run slower than streetcars at least in part because of loading delays. You cannot assume that the system average cost per bus passenger would translate to buses running on streetcar routes.
This is not a straightforward calculation, and it must be done from first principles using disaggregated cost information for each mode applied to specific routes, not system averages.
LikeLike
Oh boy, I wait 3 days to check and everyone’s talking about new lines and the east end and computer studies of TSP! It’s like my birthday came early… 😀
I’ll do the extra 50 cars first, followed by TSP, then step off into fantasy route planning world.
Steve: There will be 60. You have 10 more to play with, as you appear to discover later in your comment.
On the cars – yes, we should get them. There’s lots of talk of “is the ridership fed by the service level, or vice versa?”, but the evidence I generally hear is that it depends if we’re going up or down. A low ridership is generally fed by poor service levels, but high service levels are generally only fed by high demand. That being said I think there is a clear need for additional cars given the known demand increase, and the less-than-acceptable service standards already existent throughout many portions of the operations. I can wait 20 minutes for a Dundas car, I can be passed up by multiple King cars in a row, and both can happen at the same time on queen. I understand that the LFLRV’s are bigger, but that only helps (and only to a point) where demand is outpacing supply – not where service headways are below what most would deem acceptable. Even in some magical fantasy world where there is no ridership increase at all over the next decade or two, they could use the extra vehicles for any new lines, or just providing higher service levels – operators are the costly part, and adding 60 to the order gives us roughly equal vehicle count (there’s a slight increase – I’ll take it).
Steve: Demand can be induced by better service. Back in 1980, the TTC actually tried improving service on a handful of routes to see what would happen. Surprise! Ridership went up.
TSP – There’s a lot not being talked about here, just multiples of cycle times being quoted. That’s not even half the story! If the street was entirely emtpy except for LFLRV’s, single cycle headways are completely maintainable – an exactly spaced line of cars advances through a green to the next stop, all as a unit, load during the red, all as a unit, repeat. But traffic, of course, messes that up. In Toronto traffic at rush hour, I doubt 2 cycles would be sufficient, because it’s not hard to remember times that 2 cycles failed to clear even one car at, say, Yonge and King. I agree with Steve that a “Macro” TSP solution is not necessarily effective or desirable in this case – what we need is a more “Micro” one, where cars can have greens held or closed specifically based on their loading requirements, and lights held locally green through to the next stop. I know the current fleet can’t do this, but do the new LFLRV’s have any provisions for signalling local lights with a “stopping/not stopping” message, or the potential for that in the near future? Other things can be done to improve TSP in traffic, i.e. prioritizing transit vehicles more aggressively or integrating with TTC scheduling (HA! They won’t even turn it on…), but I think traffic congestion and local stop signalling are the keys here.
New routes – I’ll only comment on the east end stuff, both because that’s the area of the city I know best (born and raised in Beaches, then Lesliville, now work in Riverside/Riverdale and commute from QQ west) and because it’s the one that’s been talked about, between the Lever site/ LRT east, and the Cherry spit, and mentions of Parliament etc. As of right now I’d say the city’s development has been heavily skewed to the west end – Liberty Village, Cityplace, etc – but is now moving to the east – East Bayfront, Lever site, Regent Park, etc. So, there needs to be new routing in the east end to account for this, like the Cherry Street spur or the LRT east (which I’ll refrain from giving a route-suggestive name). The planned developments under Waterfront TO are the East Bayfront, Portlands, and I’ll toss in the Lever site (as well as hope for a Broadview extension) given it’s similar location – there is also a massive Daniels led redevelopment of Regent park, and it’s not unreasonable to assume St. James Town will undergo a similar fate in a few decades if Daniels are successful with R.P.
Steve: Note that all buildings in St. James Town are not public housing. It’s a different type of development.
My suggested new routes are, therefore, in order of implementation 1) extending the Cherry spur through the Portlands and back up Broadview as a split of the King route. 2) A waterfront east LRT running along water (ish) to Leslie, then preferably up Jones to Donlands station. My fantasy wants to add one from Sherbourne station, up a tunnel onto Howard, and down Parliament to…somewhere, to serve St James Town and Regent Park, but I can’t find a way to make this work or justify it.
1) I don’t think anyone would argue that the spur as is needs to be extended to reach the Portlands if development happens there, and that a Broadview extension with transit to serve the Lever site makes sense. There’s no real room at Broadview station or along any other corridors to add a dedicated route to do this, and I’m not sure you need to – Splitting every 2nd car to serve Lever/Portlands/West Don lands means a 4 minute headway through that area, and a 4 minute headway for anyone east of Sumach who wants a direct-to-downtown journey (it might make more sense to split off every 3rd car, but demand will tell better than I can). While this may start poorly, a DRL would relieve the commuter congestion on the east of the King route (not as many would transfer to the 504 at Broadview to go downtown) and I think it’s a reasonable balance between meeting needs / balancing expectations.
Steve: As someone who lives in sight of Broadview Station and uses the King/Broadview service regularly, I can tell you that transfer traffic off of the subway to go downtown is minimal. The riding comes from the feeder buses and from all of the local stops between Broadview Station and roughly Sherbourne where loads start to turn over. Passengers begin alighting there for businesses and schools, but the line continues to pick up more people who are destined for the core and beyond. There has been a lot of bilge written here in comments by people who think the King car (and other streetcar lines) would somehow lose all of their demand if they could be intercepted by a DRL. Anyone who knows how the demand on these routes actually behaves should know better.
2) I also think an LRT east line makes sense. I’d personally run it along QQ from Union to Cherry (extending QQ or just having a rail bed between Parliament and Cherry), down Cherry to Commissioners, and along Commissioners to Leslie. I’d love to see it jog Queen to Jones and up to Donlands, to serve Leslieville and the growing east end – or up Carlaw and Pape, to Pape station, which might make more sense. Either way I think this extension might only be warranted if the corridor it runs on is re zoned – certainly a streetcar up Jones is laughable right now, and even Pape/Carlaw would need a little more vertical to warrant one at this point (though I suspect future growth will change that).
Steve: Er .. you have not been paying attention! There are already plans to reconfigure Cherry, Queens Quay East and Lake Shore with QQ going straight east across what is now Parliament Slip rather than veering up to meet Parliament as it does today. Cherry Street will shift one block west south of QQ as a new street bisecting the deep block between “Old Cherry” and the water’s edge. A connecting track via Commissioners to Leslie is planned, although this will require relocation of some Hydro infrastructure. However, once we get further east, going back north to the subway just does not sense both because the streets are too narrow and because I don’t think there would be the demand for what would, after all, be a rather roundabout route to downtown.
Pipe dream time! I’d really like to see a line up Paliament as well to serve the development there. I don’t think Castle Frank is a practical subway track-wise, so I’d run it along Howard to Parliament and then down – serving St James Town, Regent Park, etc. I’d run it south to QQ then into Union – the 509 and my above LRT east are extended to Pape and Ossington (via liberty village) through some magic out west and combined into the “Waterfront line”. At this point 4 lines are trying to serve Union, which is impossible – not sure how this would be dealt with (only waterfront line serves Union, while Spadina loops to Parliament? Spadina and Parliament serve Union, waterfront bypasses? All 4 bypass, and a dedicated link is built between Union/a new Ferry dock terminal (my favorite)?)
Steve: Oh dear oh dear you really have something interesting in that pipe. Any day now you will tell us that Toronto has been deprived of the streetcar network “it deserves”. Could we stick to transit improvements that are not pure fantasy rather than destroying our credibility on fantasy maps? You may have to move to Scarborough for your full talents to flower.
LikeLike
Steve I would like to say, I still have a dream, actually 2. I dream that someday, reliable service reasonably quick service and capacity will actually finds its way at least as far as the Humber in the west and Leslie in the east. I know this is asking too much but well you know!
Steve: What do you have against folks in Long Branch, Mimico and the Beach? They deserve good service too, and not just an occasional express bus at a premium fare.
LikeLiked by 1 person
Well, I have a hope that RER might maybe someday serve them, and well did not want to be accused of dreaming in technicolour. My ultimate fantasy – in the sense of dream only larger is Waterfront West LRT, but well you know, didn’t want to push too far.
LikeLike
Sorry, and well, I think unless something is done in terms of extending East Bayfront, which will be very hard politically, the Beach will need to rely on Queen as you said before, and well, eventually I expect that to be an issue. I know eastern Queen is not congested now, but really believe this is a not yet.
LikeLike