The new, improved transit right-of-way on Queens Quay has been in operation for a few months, and it has had its share, and then some, of problems. These included confused motorists, pedestrians and cyclists who could not figure out the new lane arrangement and signals, more than a few autos stuck in the tunnel entrance at Bay Street, and a streetcar collision thanks to an open switch at the Spadina/Queens Quay Loop.
When the design for the new road was still on the drawing boards, a red flag went up for transit watchers with the number of traffic signals, some fairly closely spaced. The “old” Queens Quay’s signals had their problems, and just to get a semblance of “priority” the detectors for approaching streetcars were moved further and further away from the signals in the hope that they would be able to cycle to a transit green before the streetcar actually arrived.
The streetcars returned, but the signals were, at first, on a standard program with no provision for detecting transit vehicles, although this changed in mid-June with the installation of the new, permanent traffic controllers.
Has there been an improvement? This article reviews current and past operations of the 509 Harbourfront and 510 Spadina cars running on Queens Quay.
A Note About File Naming
In this article, as in my previous examination of service on Kingston Road, I have used a fictitious route number “519” which includes data from both routes operating on Queens Quay. This distinguishes files with the merged data from those for individual routes.
Filenames include numeric location identifiers:
- Three digit numbers such as “105” are keyed to direction (the first digit) and location (the next two digits).
- Five digit numbers refer to a route segment or “link” between two points. For example “20605” identifies the direction (“2” is eastbound or “down” in TTC parlance for the route), while “0605” indicates the pair of locations bounding the segment.
This scheme allows filenames to sort geographically in a library even though they include street names.
For a description of the process underlying this article, please see Methodology for Analysis of TTC’s Vehicle Tracking Data.
Running Time History Over The Route
For the purpose of this analysis, the “route” consists of the section from Bay Street to just east of the Exhibition Loop loading platform. Eliminating both ends of the line avoids difficulties with getting reliable data:
- GPS errors can result in “lost” cars within the Bay Street tunnel, and a measurement from the north end of the tunnel would not reliably include all vehicles.
- Layover times at Exhibition Loop can be extremely long, and terminal time should not be included in the travel time for comparative purposes.
These two sets of charts show the weekday running times for:
- February 2010: Mid-winter, pre-construction. This should be a “best case” without summer traffic conditions on Queens Quay.
- June 2011: Early summer, pre-construction.
- May to July 2015: Post-construction including the Pan-Am Games period in July.
The data are subdivided by half-hours throughout the day, and by weeks for the months shown. (Note that “Week 0” in May 2015 contains only the short week of Friday, May 1.)
For westbound trips, running times build gradually through the morning peak, stabilize through the day, and then fall back late at night. The standard deviation values (shown in dotted lines) stay at or below two minutes except when a delay caused a spike in values indicating that travel times over the route are quite predictable within a narrow band of values.
A striking factor is the degree to which running times are longer post-construction, especially from May until mid-June 2015 when the permanent traffic controllers were installed. Even after this, running times are 3-4 minutes longer than pre-construction values, a very substantial increase considering that the “old” values were only 12-15 minutes.
For eastbound trips, the pattern is similar, although the dip in running times following the activation of new signal controllers is most evident only in the morning periods, and late evening running times are highest in July.
Just to be sure what was going on, I also looked at individual portions of the route.
Bay to York
Travel times over this short segment are very consistent, and they have improved slightly eastbound thanks to the removal of the farside stop at York Street.
York to East of Spadina
Westbound service in this segment was about 50% slower with the new signals and right-of-way than it had been before this was implemented. There was a noticeable improvement in mid-June with the new, permanent signals and improved transit “priority”, but this was lost again in July. It is unclear whether this was caused by additional summertime/PanAm pedestrian traffic or by the slow order instituted by the TTC. Until we see this segment operating “off season”, it will not be clear what influences its behaviour.
Note that some of the spikes in the Standard Deviation values are caused by a combination of a small number of high values, typically caused by a delay. These pull up the averages slightly, but because the typical values are small, a few data points with very high values can skew the SD.
Through the Spadina & Queens Quay Intersection
Travel times through this intersection have been affected by various factors since the “new” Queens Quay opened:
- For many weeks, switches were manually operated (usually) by a pointman who spent his/her time running back and forth as needed. This practice was a factor in a streetcar collision where an operator entered an open switch that should have been closed by the pointman who was busy elsewhere.
- Some transit priority phases, notably those specific to movements that depend on the setting of a switch, do not work unless the track switches are energized. The workaround for this was to simply have the “priority” turn phase(s) appear periodically regardless of actual traffic.
- The final version of the traffic signal controller was only very recently installed. A slight improvement is evident in mid-June, but this is not as dramatic as further east on the route.
The intersection is still not operating as speedily in its “new” configuration as it did in the “old” version.
West of Spadina to south of Bathurst/Fleet
This segment operates consistently more slowly in the “new” configuration than it did in the “old” at all times of the day and in both directions.
Through the Bathurst / Fleet / Lake Shore Intersection
Westbound times for operation through this complex intersection are unchanged for the “pre” and “post” periods, but eastbound times are slightly higher suggesting that there has been some change in the signal operations made while the streetcar service was shut down.
West of Bathurst to Exhibition Loop
Westbound times on Fleet Street and into Exhibition Loop were slightly longer in summer 2015 than in the earlier periods, but not by much.
Eastbound times are also similar except during July 2015 when heavy demand and service for the Pan Am Games had a severe effect. This was caused by a combination of delays at Fort York Boulevard and at Bathurst Street as shown in the following detailed chart of operation on Tuesday July 14.
Through the late evening, and particularly when there were multiple cars approaching Bathurst Street eastbound, cars can be seen pulling up to the intersection slowly. This implies that the traffic signal was unable to provide enough green time for streetcars (both the frequent 509 Harbourfront and 511 Bathurst) services.
This is one of the issues that appears to be missed in “transit priority” signals: the need for signals to support unusual conditions of heavy transit traffic, not just the typical volumes that can be fitted into the auto traffic flow. (A related problem is the absence of “priority” for commonly used short turns and diversions, but that’s a subject for another article.)
While the new Queens Quay is beautiful to look at, and certainly an improvement over years of construction, the transit service and priority there are little improved, and in some cases worse than they were with the original setup. One could argue that a mixed use street like Queens Quay was never intended to provide “rapid transit”, but this shows how the best of intentions can actually make things worse.