This is the first of a series of articles reviewing service quality on short routes. A fundamental problem across the TTC network is that service is unreliable with bunching, gaps, missing vehicles and crowding all contributing to making transit less attractive than it could be.
Many routes that get a lot of attention are quite long, and there is a raft of standard explanations for their problems. Traffic congestion and construction are chief among these, along with road accidents, ill passengers and “security” incidents. However, there are severe problems with service reliability on short routes where most of the standard explanations simply do not apply, and where the TTC should be able to maintain service like clockwork.
These routes are short enough that the source of problems is easily spotted in the tracking data for TTC buses. The two most common problems are:
- Buses are missing from service probably because no operator is available to drive the vehicle, and a near-embargo on overtime leaves scheduled work unfilled.
- Where buses are missing, service is not always adjusted on the fly by Transit Control to space out the remaining vehicles and the result is large gaps where missing vehicles should be.
- Some operators simply prefer to drive in packs even though they are reasonably close to their schedules. At times, pairs (or worse) of vehicles will make multiple trips close together showing that there was no attempt to space service.
There is a distinctive difference between missing and bunched buses in the data.
Where a bus is missing, headways will widen either where that bus should have been, or overall if the remaining service is spaced out. Where buses are bunched, there will be corresponding short and long headways where two or three buses arrive together followed by gap much wider than the average headway.
In worst-case situations, which happen too often for comfort, most or all of the route’s buses run together in a convoy. These are eventually broken apart, but the convoy should not have been allowed to develop in the first place.
In brief, there are times when nobody is minding the store and riders suffer. In the two months of data reviewed here and in Part II to follow, these problems are not one-off instances, but repeated events.
On 22 Coxwell, the service on the weekday shuttle between Danforth and Queen is usually well-behaved, but come evenings and weekends when the route extends east via Kingston Road to Bingham Loop at Victoria Park, the service can be very erratic.
In this article, service will be shown at three locations:
- Coxwell south of Danforth, southbound. This is the service shortly after it departs from Coxwell Station.
- Coxwell north of Queen, northbound. This is the daytime service shortly after it departs Eastern Avenue for the trip north. On weekends and evenings, this is a mid-point of the route.
- Kingston Road west of Bingham, westbound. This is the evening and weekend service after it leaves the eastern terminal.
To save space, the charts are presented as galleries which readers can open at any page and scroll back-and-forth to make comparisons. Full sets of charts, including illustrations not included in the body of the article, are linked as PDFs after each gallery.
In the text describing the charts tracking vehicle movements, I refer to buses by the colour of the line rather than the run number because this saves readers from having to translate via a legend. Each day’s colour allocations are independent of the others, and they occur in order of vehicle numbers in the underlying data. For example, the “pink” bus is a different run number each day depending on the vehicles assigned to the route.
For those who have not encountered these charts before, there is an introductory primer. For those who want to know how the underlying machinery to produce these analyses works, there is a detailed article about methodology.
Continue reading →
Steve Munro 