News   Dec 20, 2024
 2.6K     8 
News   Dec 20, 2024
 1K     2 
News   Dec 20, 2024
 1.9K     0 

Ottawa Transit Developments

Well, this tweet kind of sums it up. There is a bit of hyper focus on things right now... the moment the countdown clock moves for any reason twitter explodes whereas in other cities people would shrug it off. Still, there's been a delay virtually every single day of the past 10, and even though they are almost all minor that seems high. The other thing is that in KW the whole transit network doesn't grind to a halt when the train has an issue, but in Ottawa at the moment delays cascade problems throughout the bus system as well, as they are very tied together. OC Transpo kind of bungled their way into things, and because there's confidence issues right now anything less than Japanese-like efficiency at the moment will not be accepted by the public. Eventually things will calm down.

 
Last edited:
I’d like to see reliability statistics comparisons between Ottawa and Kitchener-Waterloo.

We hear very little about KW ION LRT problems, so I’m pretty curious how the situation compares.

This data can be important for other LRT projects beginning construction such as Finch, Hamilton, Hurontario... and even Ottawa Stage 2, too.

I don't think we're likely to see official stats until at least the 1 year mark. That said anecdotally the ION did have a number of mysterious delays in the first month or two after launch, but not as many people were taking trains, so it wasn't as noticeable. Transit riders in KW don't have a complaint culture which can have good and bad aspects to it, there isn't the overwhelming negativity of other cities but there's also little momentum toward improvements. Complaint culture seems to correlate heavily with commuters and white collar professionals where being late really matters, while students and seniors are comparatively more relaxed. The Confederation line seems to have a lot of commuters so that may be a big part of things. Since ION service started to mature in the fall delays have been minimal and pretty much always due to collisions, which is an unfortunate reality of street running LRT. Therefore popular consensus (aside from boomer car culture types) is that it's drivers' fault. Confederation line delays can't be attributed to cars so people are more likely to blame the system itself and Rideau group.
 
I don't think we're likely to see official stats until at least the 1 year mark. That said anecdotally the ION did have a number of mysterious delays in the first month or two after launch, but not as many people were taking trains, so it wasn't as noticeable. Transit riders in KW don't have a complaint culture which can have good and bad aspects to it, there isn't the overwhelming negativity of other cities but there's also little momentum toward improvements. Complaint culture seems to correlate heavily with commuters and white collar professionals where being late really matters, while students and seniors are comparatively more relaxed. The Confederation line seems to have a lot of commuters so that may be a big part of things. Since ION service started to mature in the fall delays have been minimal and pretty much always due to collisions, which is an unfortunate reality of street running LRT. Therefore popular consensus (aside from boomer car culture types) is that it's drivers' fault. Confederation line delays can't be attributed to cars so people are more likely to blame the system itself and Rideau group.
Routes such as Hurontario and Hamilton are at least higher-order than the street section of ION LRT.
-- Much like Spadina streetcar except much longer curbs between traffic intersections, and bigger/raised subway-style platforms.
-- No left turns allowed except at traffic signalled instersections.

Transit priority systems work extremely well on such systems:
The Eglinton Crosstown is likely to be the first LRT in Toronto, Ontario to utilize a "properly" modern predictive transit priority system. Otherwise, the intersections will clog the whole service (including trains in the tunnel). They have to, or else -- it's critical to a hybrid subway+surface system.

How GOOD transit priority works
-- Better transit priority (that can be GPS-based, or hook into new signalling systems that feeds realtime positional & speed info)
-- Once an LRT train departs the previous stop, it has a deterministic arrival time at the next traffic light 0.5km later. (curbed lanes with no obstructions & no LRT stops)
-- At that next intersection -- the pedestrian crosswalk countdown timer begins automatically to clear pedestrians safely.
-- The intersection goes red a few seconds before the LRT train arrives. The intersection is safely clear in advance of a full-speed LRT train.
-- The LRT train whooshes through the intersection at a speed of up to 40-50kph (typical speed of a median LRT).
-- The LRT train decelerates to the stop on the far-side of the traffic light (unpredictable dwell time, but that doesn't matter; bear with me, continue reading)
-- Once the doors close and the LRT starts accelerating again, it's yet another 0.5km to 1.0km continuous curb straight to the next traffic light, no LRT station to stop at beforehand.
-- Good systems predictively time the green light to the speed currently occuring (e.g. slow moving in bad weather, versus fast moving in good weather).
-- It's predictable if it's a curbed lane clear of cars and clear of LRT stops/stations until right after the next traffic light (aka "far side stops")
-- This makes transit priority (automatic green lights for LRT) much more intelligent than for yesteryear TTC streetcar systems.
-- Properly designed, the lights can be green-in-advance for an LRT practically 80-90% of the time, for a full speed whoosh-through.
-- For a properly engineered transit priority system, they become almost as train-speed-preserving as crossing gates.

As a buy-in for automobile owners that hate this concept; consider these systems often give drivers an extended green light right after the LRT train passes by, as a kind of a quid pro quo.

TTC needs to upgrade to these systems on St. Clair and Spadina eventually. The current system in use is a crude barebones transit priority system that has a Homer Simpson zero brain. No intelligence whatsoever, and often mistimed. Such legacy "Transit Priority Lite(tm)" needs to go in the crapper and be upgraded to a much smarter European-quality transit priority system like the above. Even the Calgary C-Train transit priority system is superior to the TTC transit priority system.

This is also why Eglinton West LRT shouldn't have to be completely tunneled; it should still be possible to reliably transit-priority a lot of intersections because of the way Crosstown is already designed with long curbs + far-side stops, making possible reliably-predictive transit priority systems.
 

Attachments

  • 1572993647886.png
    1572993647886.png
    697.6 KB · Views: 382
Last edited:
Routes such as Hurontario and Hamilton are at least higher-order than the street section of ION LRT.
-- Much like Spadina streetcar except much longer curbs between traffic intersections, and bigger/raised subway-style platforms.
-- No left turns allowed except at traffic signalled instersections.

Transit priority systems work extremely well on such systems:
The Eglinton Crosstown is likely to be the first LRT in Toronto, Ontario to utilize a "properly" modern predictive transit priority system. Otherwise, the intersections will clog the whole service (including trains in the tunnel). They have to, or else -- it's critical to a hybrid subway+surface system.

How GOOD transit priority works
-- Better transit priority (that can be GPS-based, or hook into new signalling systems that feeds realtime positional & speed info)
-- Once an LRT train departs the previous stop, it has a deterministic arrival time at the next traffic light 0.5km later. (curbed lanes with no obstructions & no LRT stops)
-- At that next intersection -- the pedestrian crosswalk countdown timer begins automatically to clear pedestrians safely.
-- The intersection goes red a few seconds before the LRT train arrives. The intersection is safely clear in advance of a full-speed LRT train.
-- The LRT train whooshes through the intersection at a speed of up to 40-50kph (typical speed of a median LRT).
-- The LRT train decelerates to the stop on the far-side of the traffic light (unpredictable dwell time, but that doesn't matter; bear with me, continue reading)
-- Once the doors close and the LRT starts accelerating again, it's yet another 0.5km to 1.0km continuous curb straight to the next traffic light, no LRT station to stop at beforehand.
-- Good systems predictively time the green light to the speed currently occuring (e.g. slow moving in bad weather, versus fast moving in good weather).
-- It's predictable if it's a curbed lane clear of cars and clear of LRT stops/stations until right after the next traffic light (aka "far side stops")
-- This makes transit priority (automatic green lights for LRT) much more intelligent than for yesteryear TTC streetcar systems.
-- Properly designed, the lights can be green-in-advance for an LRT practically 80-90% of the time, for a full speed whoosh-through.
-- For a properly engineered transit priority system, they become almost as train-speed-preserving as crossing gates.

As a buy-in for automobile owners that hate this concept; consider these systems often give drivers an extended green light right after the LRT train passes by, as a kind of a quid pro quo.

TTC needs to upgrade to these systems on St. Clair and Spadina eventually. The current system in use is a crude barebones transit priority system that has a Homer Simpson zero brain. No intelligence whatsoever, and often mistimed. Such legacy "Transit Priority Lite(tm)" needs to go in the crapper and be upgraded to a much smarter European-quality transit priority system like the above. Even the Calgary C-Train transit priority system is superior to the TTC transit priority system.

This is also why Eglinton West LRT shouldn't have to be completely tunneled; it should still be possible to reliably transit-priority a lot of intersections because of the way Crosstown is already designed with long curbs + far-side stops, making possible reliably-predictive transit priority systems.

While I agree with aren't most stops on all the LRTs planned non staggered? Which means one train is always on the wrong side of the light which might decrease the efficiency.
 
While I agree with aren't most stops on all the LRTs planned non staggered? Which means one train is always on the wrong side of the light which might decrease the efficiency.
This is true -- though the proportion of nearside-vs-farside stops varies from LRT to LRT.

The nearside transit priority prediction problem:
Unpredictable dwell times means transit priority signalling isn't always as reliable for nearside stops. By the time LRT is ready to move, the light may be red.

Basic solutions to improve transit priority behaviour include
-- Predicted dwell estimate, especially with standardized LRT dwell time that's enforced 90% of the time (can also vary by time of day, longer standardized dwell for peak crowds and busy stations)
-- Platform crowd sensors are also a possible additional input, though not commonly used
-- Combined with common old standards such as hold-greens / lengthen-reds to fudge the signal timing to the LRT's favour. But they have upper/lower bounds (minimum lengths & maximum lengths). Though this can assist when boarding takes slightly longer than expected deviating away from standard dwell estimate.

In this newer era, it is easier to keep refining transit priority systems. AI and natural learning of historical dwell behavior thoughout the day, week, year, holidays, seasons, also can provide reasonably accurate dwell estimates that are more spot-on than boilerplate numbers. Many of these newer systems are software upgradeable and can gain transit priority enhancements like these. This data can feed (directly or indirectly) into a transit priority system to automatically fudge the stoplight cycling [lengthen/shorten the cycle within safety bounds] to realign a green into an LRT's favour. Improved algorithms massively raises the success of an LRT's transit priority system.

The Crosstown is (reportedly) using system capable of being programmed to do this, if they wished (and Toronto lets them). They might have to export the performance data to an external computer, process the data externally to optimize it all, and re-import new algorithms/timings back into an intersection that has already been upgraded to ‘smart” transit priority, capabilities, but these systems are programmable in many ways (fixed, dynamic, etc).

There is less pressure to do this at low frequencies, but once an LRT needs to do metro-league frequencies and ram the trains through the subway segment at subway frequencies -- you don't want downstream transit priority mis-predicts to congest the trains behind (e.g. trains still coming out of the tunnel and into surface crossing sections). During peak hour & during train incident/delays -- you need trains quickly able to catch up back to their slots quickly. It will be critical for Eglinton Crosstown to have accurate transit priority algorithms vastly superior to anything TTC is currently using with their streetcars. These are really busy artery intersections that Crosstown is going to need to whoosh through.

The goal is to retime the reds/greens so well -- that the intersection might as well be as good as a railroad crossing gate. You might need to start retiming about 2-3 intersections ahead, if the bounds of green/red lengthenings/shortenings is very narrow (e.g. "40 seconds to 60 seconds" range limitation), but more flexible (only 1-2 intersections ahead) if the range is wider (e.g. "30 seconds to 90 seconds"). The crosswalks will enforce a mandatory minimum, since you have to clear very wide artery crosswalks (crosswalk countdown begins), so that makes things a bit tricky. On the other hand, the lengths between the intersections is extremely wide (over 1km of curb in some situations) which makes transit priority easier -- making it easier to re-time the intersection a lot more imperceptibly to cars/pedestrians, well ahead of an LRT.

Complicating this, you have all kinds of a long cycle for major artery intersection, like protected left turn moments. But you can lengthen/shorten different sections of a traffic light cycle (e.g. 5 seconds longer, 5 seconds shorter per stage). That can build up to enough fudging (30 second earlier, 30 second later) necessary for a green light the moment the LRT needs to whoosh full speed through.

Ontario is still very new to this new flavour of "smart" transit priority...
 
Last edited:
The transit signals should be able to hole the trains to prevent bunching. Then there is the problem, as reported earlier, of passengers holding the doors, which lengthen the dwell time or worse cause a breakdown.
 
From link...

November 6, 2019

Thank you for continued patience as we transition to a multimodal service, integrating bus and the O-Train into our daily trips. We acknowledge the many issues that have affected transit users since the much-anticipated launch of the O-Train Line 1.

This September the initial opening of LRT was an overwhelmingly positive experience.

Since October, new and modified bus services have meant significant changes to 100 bus routes where approximately 240,000 daily customer trips changed in some way. We know that with a change of this size and magnitude, there will need to be adjustments based on customer experiences.

During this transition, our system has not been operating with the level of consistency our customers expect and deserve. OC Transpo has a clear objective: to provide reliable and efficient public transportation.

To do this, we are implementing corrective measures, effective immediately. These improvements include:​
  • Freezing fares in 2020 for the EquiPass, Community Pass and Access Pass
  • Additional staff to provide live information and customer support via social media
  • Increased staff in customer service areas
  • Adding 40 extra buses to daily service on routes with the greatest concerns expressed by customers
  • Creation of direct non-stop trips from Tunney's Pasture to west and south end
  • Adding 19 new buses to the fleet in 2020 to provide more reliable and expanded service
Our contract with Rideau Transit Group (RTG) means that they are responsible for the design, construction, integration, testing, commissioning, and 30-year maintenance of O-Train Line 1. The City will hold Rideau Transit Group accountable to fulfill their obligations under the contract.

With winter approaching, OC Transpo is inspecting heaters and defrosters and performing pro-active maintenance.

OC Transpo is dedicated and focused on ensuring customers are taken care of through every aspect of their journey. Our focus is on providing reliable transit service, holding Rideau Transit accountable to maintain the LRT system and improving communications with our customers.

O-Train Confederation Line & Bus Service Update *
 
On the other hand, by TTC metrics Ottawa has an on time reliability of 97%, which is much higher than TTC's Line 1.

The reality though is that there's broken trust at the moment due to the rough launch during the first week. Every single delay, no matter how minor, is immediately reported on and talked about on Twitter as proof the system is garbage. This is a hard thing to come back from, we need 100% reliability for at least a month before people trust again, and that's a very tall order. Lots of people are screaming "bring back the buses" completely forgetting that OC Transpo's bus reliability record is and has been absymal for years, the train is light years ahead in terms of reliability.

Perception is reality though, so I don't know how OC Transpo can fix things in the short term. Long term people will just go back to muttering under their breaths about OC Transpo just like it was in the old days, rather than making a big show about it.
 
Meanwhile in Ottawa, door issues seem to have died down..... Only to be replaced by much worse problems with trains stopping due to various computer gremlins.

Trains are still crush loaded all the time during rush hour, because they can't even run 13 trains consistently, let alone the 15 promised. There has been at least one incident for the past 26 days :(

But it has spawned this rather fun and at the same time useful status website

 
Meanwhile in Ottawa, door issues seem to have died down..... Only to be replaced by much worse problems with trains stopping due to various computer gremlins.

Trains are still crush loaded all the time during rush hour, because they can't even run 13 trains consistently, let alone the 15 promised. There has been at least one incident for the past 26 days :(

But it has spawned this rather fun and at the same time useful status website


The situation in Ottawa has not gone unnoticed by the good folks at The Beaverton!


OC Transpo recommends Ottawa commuters leave for work a week earlier
OTTAWA – Ottawa’s public transit service is advising that area commuters should arrive at any bus or LRT stop a week in advance as a precaution to avoid being late for work.

“We are still making the transition to incorporate the LRT into the system so please leave your homes no later than 168 hours in advance,” said an OC Transpo official. “Thank you for your patience and understanding.”

An unexpected gentle breeze caused doors to malfunction, and contact with sunlight somehow caused OC Transpo officials to prematurely cancel bus routes leaving thousands of passengers stranded.

“We hope that during this chaos, two single people will have their meet cute moment and find true love,” added the official trying to look at the benefits of a frustrating situation.

In addition to departing a week early, transit officials say that passengers will be stacked on top of each other to avoid congestion on platforms to improve efficiency and safety.

When asked by a journalist about how the LRT will perform in snow, the spokesperson appeared to be horrified.

“Oh, no,” said the spokesperson widening his eyes while loosening his tie. “We’ve never thought that snow was a possibility in Ottawa, Canada’s capital.”

At press time, passengers were told to get out and push the 50 ton trains if they wanted to get anywhere.

https://www.thebeaverton.com/2019/1...tawa-commuters-leave-for-work-a-week-earlier/
 

Back
Top