What Toronto transit planners can learn from Vancouver
The Eglinton Crosstown is delayed again. Here’s how things could’ve gone differently.
During a bit of a recent media craze, we heard that Toronto’s Line 5 — the Eglinton Crosstown — has a ton of construction-quality issues that are holding up commissioning.
To recap: The project was meant to be done in 2020, and it still likely won’t be open until well into 2024.
Everyone knows big projects are usually completed late, so some delay shouldn’t be surprising to us at this point. That’s especially true because the best comparables — other rail rapid-transit projects in Ontario and Canada — have also been going overtime on construction. Here’s a sample:
- Evergreen Line SkyTrain Extension: Six-month delay, opened 2016
- Toronto York Subway Extension: Two-year delay, opened 2017
- O-Train Line 1: One-year delay, opened 2019 (clearly rushed and poorly executed)
- Edmonton Valley Line: Projected opening 2020, delayed three years
What’s also key to note is that the Eglinton Crosstown is more complex than any of these projects. It has more tunnel, several interfaces with at-grade GO corridors, a connection to Line 2 of the subway at Kennedy Station, and two interchanges where it crosses just under Line 1 of the subway (these have generated a lot of delay).
But all that said, the fact remains: The Crosstown is super delayed. And that makes it worthwhile to take a macro look at the problems to see if there are things we can do better in the future.
Why do so many transit projects get delayed?
If you’re familiar with the infrastructure-construction space, you’re probably aware that it’s very common to lowball budgets and timelines to try to win political approval and the like.
Of course, this lowballing comes at a cost. The resulting delays not only create animosity toward the project, but also increase the budget because you have to pay out more hours to workers and project staff than originally intended. (For more on this topic, check out the fantastic recent book How Big Things Get Done.)
That said, while Toronto is having all of these problems, another city across the country is in a very different position. When you compare Vancouver’s recent transit projects to those happening in Ontario, the results are rather shocking.
The Canada Line built for the 2010 Olympic Games was completed on budget — which was a fraction of the price of current projects — and ahead of schedule. And even the Evergreen Line referenced above was delayed only about six months, despite having major subsidence issues resulting in large sinkholes when boring the project’s tunnel.
Clearly, something is going on in B.C. (more likely multiple somethings) that is leading to better projects, and more and more I’m urging people in the rest of Canada and the world to pay attention. This is a great case study in how to build and plan better projects.
How did Vancouver complete the Canada Line on time and on budget?
The easy and oft repeated answer is that the project was completed on time because it was a key component in the aforementioned 2010 Vancouver Winter Olympic games, but that’s beside the point.
Winter Games are not an engineering and planning solution to get subways built. Instead, the pressure of an event ensured a critical project would be completed on time — by using smart construction methods, experienced contractors, and extremely careful management. Lower-risk approaches were taken, which reduced the chances of both delays and cost overruns.
Here are some of the techniques that were implemented:
Tunnelling
One of the obvious ones is cut-and-cover tunnelling. The Canada Line project more or less used cut and cover where it was easily practical — notably on the wide Cambie St. on the southern part of the tunnelled section in the city of Vancouver.
Cut-and-cover works were completed fairly quickly (remember, the whole project took three years), but even still, public and media backlash has meant that it will likely not be used again anytime soon. Cambie St. business owners successfully sued for disruption (though the Canada Line was eventually granted an appeal), and the media has poured gasoline on the fire for years.
But it seems pretty clear to me the disruption on Eglinton Ave. — where tunnels were created using tunnel boring machines (TBMs), and works have been going on for several times as long — has been far worse.
If citizens and businesses are going to scream if you do the construction method that is faster and less expensive — which makes it no longer politically acceptable — that’s a really bad outcome.
Stations
Stations are one of the most costly and time-consuming parts of building a transit line, given their scale and complexity. So the Canada Line massively simplified these facilities as well, all while providing the same capacity as the Crosstown.
Now, I get asked endlessly — how can the capacity of the stations be the same if the trains themselves are so much longer on Eglinton?
Well, the rapid transit system capacity depends on several factors. And while Eglinton’s trains are longer, the Canada Line trains — which are traditional subway trains rather than trams — are wider, have a more open layout, and more doors allowing for more frequency.
Essentially, Eglinton’s stations — using Keelesdale as an example — are typically larger and more intricate than they need to be for the number of passengers who will conceivably use the system. The multiple entrances and other flourishes are nice, but they are nonessential, and other systems do fine without them.
The Canada Line, meanwhile, appears to have started from the trains to determine the minimum acceptable size for all other infrastructure, so each Canada Line station has only a couple of fare gates and one entrance.
The result is that passengers will arrive in frequent but small bursts, and the actual flow into and out of the station will be less “spiky,” reducing the need to size things for infrequent rushes of passengers coming off a big train. It’s essentially a tradeoff between more sophisticated operations (enabled by automated trains) and more complex capital construction.
Project structure
Of course, while it may be suboptimal to build a high-capacity subway for use with trams, just because the Crosstown’s design isn’t ideal doesn’t mean it should take longer and cost more to build — so what’s the issue with that?
I think a lot of the problems here come down to project structure, timeline, and design. The Crosstown started its existence as a TTC project, but was eventually “adopted” by Metrolinx and run as a P3 process — which has gotten it a lot of bad press.
But the Canada Line was also a P3, so that’s not necessarily the problem in and of itself. The difference is that the Canada Line was a very open project. The government gave a rough specification for what it wanted — 25-minute travel time from downtown to the airport, a certain capacity, etc. — and let the bidders go at it. One successful bidder would build the system, buy the trains, and operate it for 30 years.
By comparison, Eglinton was much more constrained and complex. Tunnels were a separate contract done before the stations and maintenance contracts (meaning using cut-and-cover to build cheaper, shallower stations wasn’t in the cards), and trains were chosen beforehand and purchased separately by Metrolinx.
The project is also going to be operated by TTC, rather than the project’s builders — which will probably create conflict if maintenance done by a third party is not up to the agency’s specs.
And I haven’t even mentioned the legal action that occurred when Bombardier couldn’t deliver the trains for the project on time.
The period of time the project has taken to construct and plan is also frequently blasted in the media, and on social media — but, as I just mentioned, it was clearly not optimized for quick completion.
Do current Vancouver transit projects live up to the Canada Line legacy?
Scheduled to open in 2026, the Broadway Subway is not a perfect project, but I find it fascinating to compare it to the Crosstown — both Eglinton and Broadway are dense urban streets with lots of traffic, and both subway projects need to intersect an existing subway line.
That being said, despite having years of inflation and potentially negative influence on it, the Broadway Subway still seems to be better designed than Eglinton, and this is reflected in its timeline — which is only about five years from start of tunnelling to line opening.
For one, the phasing of the Broadway Subway seems more logical: Station boxes will be excavated prior to tunnelling (on the Eglinton project, station headwalls were the extent of pre-tunnelling station construction). This means final station construction can begin the second the tunnel boring machine passes through each site — shaving years off the project.
The stations on Broadway are also all being built cut-and-cover under decks, while several on Eglinton used the costly and risky mining method.
Even just comparing the station designs, Broadway’s, despite being built for 60 per cent more capacity (SkyTrain Millennium Line is designed for 25,000 passengers per direction per hour, while Eglinton is designed for 15,000), are much simpler.
Check out this image of Mt. Pleasant Station on the Crosstown:
And then Mt. Pleasant Station on the Broadway Subway (funny that both have a station with the same name, I know!):
Notice how the Vancouver station has one entrance and a small emergency exit, while the Toronto station has two large entrances. The Vancouver station is also a little bit shorter in length.
Probably most interesting to me is how the station in Vancouver manages to remove a whole vertical level from the design. Passengers descend from street level to a mezzanine and then down to platforms.
Meanwhile, in Toronto there is an additional sub-mezzanine level. There also appears to be way more back-of-house space in the Toronto station design, and this is all for a station that is designed to move less people.
So, at the end of the day, I think it’s important to remember that Toronto isn’t unique in delays on transit projects, but there is clearly much to be learned from successful projects in Vancouver and elsewhere.
This story is an edited version of a post from RM Transit, transit analyst Reece Martin’s Substack email newsletter. You can subscribe to the newsletter here.
Code and markup by Kyle Duncan. ©Torontoverse, 2023