The King Street Transit Pilot study will hold its next public meeting on May 18 to review the proposed design for that street. As background to that study, this article includes a review of travel times over the King Street corridor in recent months.
Important issues raised by these data include:
- Travel time issues on King are not restricted to peak periods or to weekdays.
- Problems in the PM peak are generally worse than in the AM peak.
- Conditions can be perfectly “normal” one day and severely upset on another. Some weekdays are consistently worse than others, but “abnormal” days occur often enough that they are part of the landscape, not rare exceptions.
- Congestion is not confined to the pilot study area between Bathurst and Jarvis, but some portions of King Street see little effect from congestion. A “one size fits all” approach will not deal with all of King’s problems, and could produce little benefit in some areas. Expansion beyond the pilot area, if any, requires detailed understanding of just where and what the problems might be.
This is a chart-heavy article intended as background material for readers interested in what the route looks like today.
The following charts show the average travel times between Jarvis and Bathurst in each direction from October 2016 to April 2017.
- Monday-Friday data have been averaged for each week with statutory holidays omitted.
- Each line on the chart represents data for vehicles crossing Jarvis westbound or Bathurst eastbound during a half-hour period as shown by the legend.
Stepping through the pages, one can see the evolution of travel time averages over the day with the highest values coming in the afternoon peak both ways. The differences between the high values and those during other parts of the day give some indication of the potential travel time saving although this does not really tell the whole story as we will see later.
Even though the data have been subdivided by half-hours, this still masks the degree of variation in travel times that is visible only when one drills down to the daily level with individual vehicles. The following charts show that there is a considerable variation from day to day with notable effects late in the week, but also at other times. For the worse days, the change from times that can be achieved when conditions are good can be quite dramatic. In the charts below, the following effects should be noted:
- The difference between days is sufficient that the high values on certain days are masked in the weekly averages by the “better behaved” days.
- Travel times during the evenings on Fridays and Saturdays are high, and for Saturdays, they are higher than the “daytime” values.
- These are “winter” days with few of the events that can snarl King Street during the peak tourist and entertainment season. 2017 saw a fairly benign winter with much of the precipitation as rain, and with little snow accumulation. As winters go, this was a comparatively easy one for King.
An important challenge for the King Street pilot is not simply to reduce travel times on an average basis reclaiming a relatively small amount of time over the route as a whole, but to tackle those situations where travel times now more then double their typical values. In a future update to this article, I will dive into the specific events behind the days when running times were unusually large.
One recent and quite visible event occurs on Friday, April 28 when eastbound times rise markedly at the end of the afternoon peak. This corresponded to the onset of a 501 Queen diversion around trackwork planned for that weekend. The difficulty of making the east-to-north turn, completely without transit priority, at Spadina has been reported on this site before as a perennial problem. It shows the importance of dealing with non-standard events, not just the scheduled day-to-day operations. (It is also worth mentioning that the effect appears before 6 pm because Queen cars started to divert before the advertised start of the work project.)
The week of April 17 was the beginning of the demolition of the York Street ramp from the Gardiner Expressway. Eastbound travel times appear to be slightly higher during this week, but with the amount of “noise” in day-to-day variations it would be difficult to attribute this change only to work on the Gardiner. In past years, Gardiner effects have been more pronounced when they divert traffic onto King either as a parallel route, or because access paths to ramps have changed.
Looking at the route on a wider basis puts the hoped-for benefits of the downtown pilot in perspective. Here are the running time histories for the route from The Queensway to Parliament. Reading these charts it is important to remember that the day-by-day values show considerably more variation than the weekly averages, and there might be more potential for improvement on the worst days than the averages imply. However, it is important to understand the source of each type of delay, and the problems will not vanish unless the changes are appropriate.
A noticeable change is that there is a drop in average travel times eastbound at the end of the AM peak comparing late 2016 to early 2017 data. When I reviewed this in detail, it turned out to be a change in operating practices by the TTC, not a change in traffic conditions. During late 2016, some bus trippers on King would take extended layovers east of The Queensway eastbound, and this made them appear to have longer travel times. In a future analysis I will filter these out, but will also examine the degree to which these vehicles were used to fill gaps (and thereby contribute to service quality), or simply contributed to or created a “bunch” heading eastbound into the core.
The intersection at The Queensway can trigger backlogs of traffic onto King because of the multi-phase nature of the traffic signals and, at times, the demand for westbound King to westbound Queensway turns exceeding the green time available. This shows up as extended running times later in the day when traffic flow is primarily westbound.
Over the longer distance from Bathurst through Liberty Village and Parkdale, other westbound delays contribute to longer travel times, notably for left-turning traffic headed to the Gardiner Expressway. Although there are some variations in travel times eastbound, they are not as pronounced.
Finally, the portion of the route east of Jarvis sees no congestion at all. (Note that for this analysis, the segment ends at Parliament, but the same conditions apply east to River Street. Past studies included data from the extended period when all King cars operated via Parliament and Queen due to the Don Bridge closure, and I preserve the segment boundaries in the data to allow for comparisons over longer periods than are shown here.)