Day: November 28, 2024

by

Actual vs Advertised Wait Times

A central part of any transit rider’s journey is the wait for a vehicle that may or may not show up when expected. Even with an app that tells you where the bus is, the news might not be good. Rather than being just around the corner, the bus might be several miles away, and heading in the wrong direction.

The only statistic the TTC publishes on service quality is an “on time” metric. This is measured only at terminals, and even there “on time” means that a bus departs within a six-minute window around the scheduled time. Performance is averaged over all time periods and routes to produce system-wide numbers, although there are occasional references to individual routes in the CEO’s Report.

Riders complain, Councillors complain, and they are fobbed off with on time stats that are meaningless to a rider’s experience.

The problem then becomes how to measure the extra time riders spend waiting for their bus, and to report this in a granular way for routes, locations and times.

This article presents a proposed method for generating an index of wait times as a ratio comparing actual times to scheduled values, and their effect on the rider experience. The data are presented hour-by-hour for major locations along a route to see how conditions change from place to place.

An important concept here is that when buses are unevenly spaced, more riders wait for the bus in the long gap and fewer benefit from buses bunched close together. The experience of those longer waits raises the ratio of the rider’s waiting experience to the theoretical scheduled value. The more erratic the service with gaps and bunching, the higher the ratio of rider wait time to scheduled time. This is compounded by comfort and delay problems from crowded buses, and is responsible for rider complaints that do not match the official TTC story.

There’s some math later to explain how the calculations are done for those who want to see how the wheels turn, so to speak.

Note that this is a work in progress for comment by readers with suggestions to fine tune the scheme.

Continue reading →
by

The Mythology of Service Recovery – November 2024 Update

Back in September, I wrote about the gap between the TTC’s claims of service coming back to pre-pandemic levels and the actual service riders face in their daily travels. See:

I will not repeat all of the information in that post, but we are coming into budget season and the most current info should be available for debate.

When the 2025 budget comes out, we will hear much about service recovery including the obligatory photo op with the Mayor, TTC Chair and other worthies. This will be a sham because actual service today has not been restored to early 2020 levels.

The fundamental problem with TTC claims is that they measure “service” by hours for the simple reason that the primary driver of costs is the labour associated with driving vehicles. Some costs don’t actually vary with driving time, but these are generally a smaller component of the total. (For example, some costs vary with mileage, and others such as garaging are per vehicle.) For budget purposes, the variable that counts is hours.

When comparing pre- to post-pandemic service levels, one hour of vehicle operation does not necessarily provide the same amount of service as in the past. The primary reasons for this are:

  • Buses and streetcars run more slowly today than in early 2020 due to a combination of traffic congestion and operating practices (notably the pervasive slow orders on the streetcar system).
  • More recovery time is included in schedules to reduce short turning. The premise is that if there is enough padding, vehicles will rarely be late enough that they must turn back before reaching their terminals.

The combined effect is that more vehicles (and hence vehicle hours) are required to provide the same service on many routes today compared with early 2020.

I have tracked the changes in operating speed on various routes in past articles, and will return to that subject to refresh the charts in coming months.

A related problem for riders is that thanks to uneven service (gaps and bunching), the average wait for a transit vehicle can be considerably higher than the advertised headway. TTC reports “on time performance” only at terminals where service tends to be (but is not necessarily) close to schedule. The information is averaged over many routes and all hours of the day, and bears little relevance to a rider at a specific bus stop at a specific time.

I will turn to the problem of experienced vs advertised wait times in a separate article now in preparation.

The remainder of this piece updates the September charts with planned service hours by mode to the end of 2024, and a comparison of service levels by route and time of day in January 2020 versus November 2024. PDF versions of the chart sets are provided at the end.

Continue reading →