[reinventingthewheel]

First off, I’m not surprised the software has defects - all software has defects! And integration time is when the most surprising ones show up, so I’m sympathetic

I have no sympathy for this, they’ve had every opportunity to mitigate software risks. There are very nearly zero differences in software testing conditions now as compared to several years ago when they could first run trains up and down portions of the route. Even if there are new factors to consider at this point, those responsible should have established, streamlined processes by now for change management. 3 months for the next software revision, how can that be?

 

[ZEBuilder]

The downtown streetcars aren't the same. Most significantly, it doesn't use ATC.

ION had ATC issues for a while, though most of the route they don't use it. Is it the same ATC system as Line 5?

ION opened without ATP to begin with, it was months into operation before they finally got it to work, I want to say late 2019 or early 2020 but it could've been later than that, the signs that indicate the various signaling blocks were up by late 2020. Even now nearly 5 years after the ION opened there's still issues with ATP throughout the system, the issues that exist don't inhibit operation of the system in a major way, so they're being left alone for now.

 

[allengeorge]

Even if there are new factors to consider at this point, those responsible should have established, streamlined processes by now for change management. 3 months for the next software revision, how can that be?

Yeah - 3 months feels over the top.

That said, I’ve no real understanding of what software change management and testing in this space looks like. No doubt it’s highly regulated and documentation heavy. And, my understanding is that they’re testing the integration prior to deploying something to the line, so maybe there is lots of iteration in the background?

I’m broadly sympathetic to the problem, but don’t know the specifics of the situation, and agree that your comments are valid.

 

[ARG1]

How can there be software defects, if we've been using the same cars on the downtown network for 10 years and ION for the last 5? What on earth are they doing?

Insert XKCD about the incompetence of Software Engineers: https://xkcd.com/2030/

 

[allengeorge]

Insert XKCD about the incompetence of Software Engineers: https://xkcd.com/2030/

The amusing thing is that both those products referenced in the cartoon (airplanes, elevators) have a substantial amount of software.

No one would pay for having the same quality standard for LOB software. The testing and documentation requirements are over-the-top.

 

[ARG1]

The amusing thing is that both those products referenced in the cartoon (airplanes, elevators) have a substantial amount of software.

No one would pay for having the same quality standard for LOB software. The testing and documentation requirements are over-the-top.

Not that I disagree (or want to bring the thread even further off topic), but the 737 MAX is recent memory...

 

[christiesplits]

I've read some concerns that the Crosstown, especially on the Eglinton East section, could be close to nearing capacity at time of opening. Is this a valid concern? Also, is there the capacity for additional Freedoms to be added to service later?

 

[T3G]

Is this a valid concern?

No, it's just used to discredit the choice of LRT for the corridor.

 

[generalcanada]

From what I remember the crosstown wont be true ATC in the tunnels? like the human drivers will still drive them?

If it recall correctly the only time the trains run on their own is in the yard,

I question if these software updates can wait until after opening?

Unless these bugs are part of the train protection system built into it

 

[warrens]

Feeling a sense of deja vu here after watching the later stages of Elizabeth Line development. Software was blamed for the significant delays in opening towards the end of the project.

The Crossrail people said at the time (five years ago) that a lot of the software testing they wanted to do couldn't be done because it depended on testing actual hardware which wasn't installed yet due to station build delays.

Of course, Elizabeth Line continues to have a range of problems a couple of years after opening, which is something I think we should expect to happen with Crosstown as well.

 

[urbancog]

In the tunnels, driver will close the door and initiate running but the ATC will control the vehicle between stations:

https://web.archive.org/web/20200109215906/https://blog.metrolinx.com/2019/12/09/how-will-the-eglinton-crosstown-lrts-automatic-train-control-work-we-break-down-every-major-element-in-an-infographic/

 

[nfitz]

In the tunnels, driver will close the door and initiate running but the ATC will control the vehicle between stations:

So similar operations to what Montreal started doing about a half-century ago.

 

[Undead]

I've read some concerns that the Crosstown, especially on the Eglinton East section, could be close to nearing capacity at time of opening. Is this a valid concern? Also, is there the capacity for additional Freedoms to be added to service later?

No, it's just used to discredit the choice of LRT for the corridor.

No there are some concerns if you look at the level of proposed development. It remains to be seen. Yes I imagine we could quite easily add capacity by increasing frequency.

I can see why there's capacity concerns for the Golden Mile segment of the LRT.

Currently 25,000 units are proposed. Let's assume a 0.3 parking ratio on average: that's 17,500 units, or about 31,500 people dependent on transit.

Since this area is likely to be students, young professional singles/couples, let's assume 75% of these people are working (City wide this figure is about 50%).

That's as many as 23,625 people taking the LRT. Let's be conservative and assume half of them use the LRT on any given day. Which puts us at almost 6,000 people per hour per direction during morning peak. That's about the same as the projected peak demand through the central section.

 

[Rainforest]

In the Golden Mile area, the demand will be bidirectional. Some take the westbound LRT towards the OL or Yonge, others take the eastbound LRT towards Kennedy Stn with its numerous connections.

The demand never splits evenly, but if that 6,000 pphpd splits into 4,000 westbound and 2,000 eastbound, that's still within the planned capacity.

And yes, there is room to add more trains and boost the capacity, but the city will have to buy those additional trains and hire the operators.

 

[Bojaxs]

In the Golden Mile area, the demand will be bidirectional. Some take the westbound LRT towards the OL or Yonge, others take the eastbound LRT towards Kennedy Stn with its numerous connections.

The demand never splits evenly, but if that 6,000 pphpd splits into 4,000 westbound and 2,000 eastbound, that's still within the planned capacity.

And yes, there is room to add more trains and boost the capacity, but the city will have to buy those additional trains and hire the operators.

I'm concerned we're going to see trains bunching up on the Golden Mile, street level portion, similar to what we see with the downtown streetcars. Two thirds of the line has the tracks completely grade separated and with trains flying across the line, and then they reach the at-grade, east portion where they have to start stopping at red lights/ intersections.