The TTC Board will meet on April 20, 2017. Items of interest on the agenda include:

• The monthly CEO’s Report
• Repair of SRT Vehicles
• Disposition of Bay Street Bus Terminal

This article has been updated with a commentary on subway and surface route performance statistics presented at the Board meeting. (Scroll down to the end of the CEO’s Report.)

CEO’s Report

Ridership

Discussion will inevitably turn to ridership numbers. For the first two months of 2017, total rides are running below both budgeted values and 2016 actual, although this takes a bit of explaining.

January 2017 had proportionally more days with no school traffic because of the timing of the Christmas Break, and this reduced monthly travel compared to 2016. February riding was above 2016’s level, but the running total for the year is still below 2016. This situation has continued into March 2017.

With the budget assuming a modest rise in ridership, the TTC is running below budget on revenue, although this has been offset by some expense savings partly due to the mild winter, and partly by some expenses falling later in the year than planned. The latter are not real “savings”, only a matter of accounting for which the reckoning will arrive eventually.

This is a difficult environment into which the TTC’s still-undelivered “Ridership Growth Strategy” will have to be considered whenever it shows up. As long as we have a Mayor and Council desperate to rein in spending, and the TTC faces extra costs in 2018 to operate the TYSSE to Vaughan, how there will be a “growth strategy” remains a mystery.

Management is particularly concerned about the possible effect of the elimination of the federal Public Transit Tax Credit (PTTC) as of July 1, 2017. This credit reduces the cost of a Metropass by 15% for anyone who has taxable income, although it is worthless for those who pay no tax. Toronto’s monthly pass is priced higher than in most cities relative to the single fare (token or ticket) rate.

The adult pass at $146.25 is equivalent to 48.75 token fares at $3 each. This is a higher multiple than the benchmark of 40 fares that many riders think of as the typical cost of commuting. With the 15% tax credit, the effective multiple comes down to 41.4. A further discount is available through the Monthly Discount Plan (MDP).

	Cost	Single Fare	Multiple	Multiple with credit
Adult	$146.25	$3.00	48.75	41.4
Adult MDP	$134.00	$3.00	44.7	38.0
Senior/Student	$116.75	$2.05	57.0	48.4
Senior/Student MDP	$107.00	$2.05	52.2	44.4

The TTC estimates the potential effect of the loss of the tax credit as follows:

• Average monthly pass sales for 2.5 years before PTTC implemented: 174,000
• Average monthly pass sales for 2.5 years after PTTC implemented: 250,000
• Annualized change in pass sales: 900,000 (est)

The TTC then assumes that if 75% to 100% of this increase were lost, then there would be 675,000 to 900,000 fewer passes sold. The TTC then assumes that this represents a loss of 7 rides per pass (the difference between the multiples with and without the credit. The imputed loss of revenue is calculated at $1.84 per ride, well below the adult fare.

	Low	High
Lost Pass Sales	675,000	900,000
Lost Rides	4.7 m	6.3 m
Lost Revenue	$8.7 m	$11.6 m

This is a rather odd way to calculate ridership losses. First off, people who now have passes are used to the convenience they bring. It is by no means certain that the 900,000 increment in annual sales consists entirely of people who still take only 49 trips/month. Those who were lured to the pass by the tax credit would soon find how convenient a ride-at-will ability was, and push their usage higher. Such riders are less likely to abandon the pass even without the tax credit. Passes will be a harder sell for new converts, but that is a different matter.

Another way the TTC does this calculation is to assume that a perceived 15% increase in cost will, via the expected negative elasticity of fares versus trips, cause less riding. However, there is a difference with passes compared to single fares in that the decision to take transit is made every time one pays a single fare, but once one has a pass, there is no further barrier to travel. (As an aside, the increase is actually 17.6%, the ratio of 1.00 to 0.85.)

The actual usage of Metropasses is tracked by a weekly diary study that samples pass-using ridership. I reviewed these data in 2016, and it is quite clear that the average weekly use of a pass is well above the level equivalent to 49 fares (about 12/week). Comparatively few passholders fall below this level, and that suggests that although PTTC may have enticed people to buy passes, once they did so, their riding habit increased above the break-even point.

On the subject of equity, the TTC observes:

Given the high cost of the Metropass, the PTTC will have generated a higher relative tax-related benefit to TTC customers compared with customers of other transit agencies. As a result, there is a very real possibility that the elimination of the PTTC may have a disproportionately negative impact on TTC customers. This negative impact may be even more pronounced for customers at lower income levels for whom the PTTC may have been important in being able to afford purchasing a Metropass. [p. 54]

That statement is valid to the extent that lower income riders have taxable income, but it ignores the fact that to use a tax credit, one must have tax payable. Also, lower income riders are more sensitive to current outlays – “can I afford to pay for a pass today” – and a future tax credit will not pay today’s bills.

There is a complex set of issues about pass pricing and various forms of discount fares, and these issues have been avoided in TTC management reports for a long period. Nobody wants to “poke the bear”, to raise the question of how “regional fare integration” as proposed by Metrolinx will affect TTC finances and ridership. Other fare structure changes such as a move to two-hour transfers or to capped pricing (a daily, weekly or monthly cap on total fares) deserve a full discussion in public, but this has been avoided by the TTC for well over a year.

To the extent that there has been any debate, it was limited to the simplistic “we can’t afford it” nostrums of “tax fighters” without any sense of the broader context of the effect on transit’s attractiveness or benefit to various classes of riders. Add to this the long-standing bias of TTC management against any form of fare discount and the sense that people are riding “free” and represent “lost revenue”. This is not the environment in which one will find a clear discussion of the available alternatives.

Vehicle Reliability

Subway car reliability figures have bounced around a lot during the past year for both the T1 (Bloor-Danforth) and TR (YUS + Sheppard) fleets. In these stats, what is measured is the failures causing a service delay of five minutes or more. The charts are a bit “spiky” with big changes in high and low values in part because the number of incidents is comparatively low, and values calculated per incident can, therefore, be very different from month to month.

The T1s were manufactured during the late 1990s and have roughly another decade of service in their design life. Whether they will actually operate that long depends on plans for the Bloor-Danforth line and its conversion to Automatic Train Control. If T1s cannot be economically converted to ATC even as a transitional measure to a new fleet, then their replacement may come sooner than their age would dictate otherwise.

The Capital Budget as of 2016 did not include any provision for T1 replacement in the coming decade, but in the 2017 budget this has been moved up to the period from 2022-2028. This project is not yet funded in the City’s plans.

TR trainsets are considerably more reliable than the T1s both because of age and because of design changes. However, some of the low values for TR failure rates dip down into the upper range of the T1 fleet. Oddly enough, a major problem lies with the interlock between the operator’s cab door and the propulsion system, and this has gone through several revisions.

To date, 60 train sets have been retrofitted with Revision E cab doors, which eliminate the broken key/door interlock issues and add a door lockout/interlock bypassing mechanism that allows the trains to remain in service. A fleet check is pending in order to see if the recent increase in Revision E cab doors is related to the issues with the quality of Bombardier’s workmanship. The passenger door system and the brake system have received numerous modifications. Fleet retrofits of the new modifications and validation testing of the proposed
upgrades are in progress and improvements are anticipated in the future. Issues related to the propulsion invertor system with occasional brake holding events are under investigation by Bombardier. [p. 43]

Note that in the charts below, the Y axis values for TR trains are twice those for T1 trains.

On the streetcar fleet, reliability of the old fleets continues to decline, while there are signs of improvement for the new Flexity cars. The ALRV fleet is undergoing a renewal program to tide it over through new vehicle deliveries, but there are ongoing problems with the availability and quality of parts obtained for overhauls of both types of cars.

Flexity reliability numbers are improving, but as of February 2017 had only pulled back roughly level with values for a smaller fleet a year ago. It is unclear whether the drop in reliability through 2016 was caused by issues as new car deliveries resumed, and whether an increase in the delivery rate will hamper growth in the MDBF values through late 2017 and into 2018. The TTC’s position is that there is a variety of factors at work, and the overall numbers cannot simply be attributed to “burn in times” for vehicle components.

As on the subway fleet, the major problem causing service delays lies in the door system.

Of the eight failures reported in February, seven of them were door-related. A plan is in place to create a reference car (“golden car”) to ensure manufacturing and set-up quality is consistently maintained to reduce downtime. [p. 45]

Note that the Y axes are at different scales for each vehicle type.

The Flexity delivery schedule is unchanged and cars are arriving as per the plan. There are 33 vehicles in service, and 4434 is now undergoing acceptance testing. (4401 remains as a prototype/test vehicle and does not count towards the delivery total.) Starting in May, we will see whether Bombardier can ramp up to a faster delivery rate which will double again in the fourth quarter if they stay on track.

Although weekends routinely see full Flexity service on 510 Spadina, 509 Harbourfront, and 514 Cherry as well as a few runs on 504 King, with as much as 90% of the fleet in service, weekdays see a mix of Flexity and CLRV operation on the 509 and 514 lines because more new cars are out of service for training and maintenance. Full service on the low-floor routes requires 30 cars: 14 for Spadina, 7 for Harbourfront and 9 for Cherry.

The TTC has decided to shift its priority for new cars to 512 St. Clair once they complete the roll out on existing routes. St. Clair requires about two dozen CLRVs for peak service, and the TTC’s fleet plan (as presented in the Capital Budget) shows this route getting 18 new cars. This represents a 50% increase of peak capacity with a 33% increase in headways (time between cars). At the planned delivery rate, it is unlikely that the schedule for St. Clair will be adjusted to reflect full Flexity operation until sometime in early 2018.

The bus fleet has shown substantial improvement in reliability over the past year thanks to changes in maintenance processes and a focus on ensuring vehicles are fit for service rather than pushing units out the door to achieve a target for service delivery.

Garage technical staff will continue to focus on quality repairs and analysis of repeaters as well as heating and cooling systems. Maintenance facilities staff have also completed the
State of Good Repair program on 383 buses since starting in early July 2016 and performed post-repair circle checks to improve bus pull-out and minimize disruptions to service. [p. 46]

Service Reliability

Peak service delivered on the rapid transit lines continues to run below the target (scheduled) level in most cases. These are monthly averages, and so some days will achieve the schedule while others will be well below the target value. Also, of course, the actual experience at various locations and directions will vary because a delay at one point may not affect service over an entire route.

Line 3 SRT has good stats in part because it is scheduled with fewer trains (5) and at a lower speed than its service design in past  years. The result is a schedule of 12 trains/hour (5′ headway) compared to 17 trains/hour (3’30” headway) with 6 trains. The service target is set at what the elderly technology and small fleet can achieve.

Of particular concern is the consistent inability to reach the scheduled service level on Line 1 YUS. This begs the question of which factors contribute to the shortfall, and which of these will be (or can be) addressed through the move to automatic operation that will roll out starting later in 2017 through 2019 when it will cover the entire line.

Delays continue to plague the system although they come from a variety of sources as the CBC recently reported. Although the TTC does have breakdowns of the cause of delay, these are not presented in the CEO’s report even though the details are often requested at Board meetings. The most common delay, passenger illness, is, in theory, “uncontrollable” in that these events are not of the TTC’s doing. However, some “illnesses” are brought on by overcrowded trains, delays and problems with ventillation such as air conditioning failures.

There are two aspects to the delay statistics: how many delays occurred, and how quickly were they dealt with. Unfortunately, these data are not broken down by type in the CEO’s report and so a long delay for a fire or suicide will be mixed together with a short delay for a jammed door.

Although it is now April 2017, the charts do not yet reflect data for early 2017 as these are summarized quarterly.

On Line 4 Sheppard, there was a considerable increase in delays due to the change to One Person Train Operation (OPTO) on October 9, 2016. This is supposed to improve as time goes on, but it is a cautionary tale for OPTO implementation on the much busier Line 1 YUS planned for 2019-20 following completion of the ATC project.

The majority of the delay incidents were related to the introduction of One-Person Train Operation. It must be noted that this increase in delay incidents was not unexpected without compromising safety standards. The bulk of those delay incidents were in the first ten weeks of the program launch and delay incidents have already begun to return to normal levels. This measure is anticipated to decrease in 2017. [p. 35]

On the surface routes, the TTC continues to fail at achieving its targets for service reliability. “On time performance” is measured at terminals, and a vehicle is “on time” if it leaves no more than one minute early or up to five minutes late. That is a rather generous metric especially for routes with frequent service where bunching can occur right at the terminal and still be within the “on time” window.

In future articles, I will be reviewing data for several bus routes, and a preliminary review shows that erratic service from terminals is a pervasive problem throughout the TTC on a wide variety of routes and under various conditions. It is part of “TTC culture” to be only vaguely on time, and this causes service as experienced by riders to be considerably less reliable than the advertised schedule. Bunching is commonplace.

The TTC is attempting to drive down the number of short turns, and on some routes has done this by adding running time to the schedules to ensure that vehicles will complete their trip without a short turn even under the worst conditions. The problem with this is that under “normal” conditions, it is possible for vehicles to be chronically early. Operators react by driving quite slowly, or by taking longer than scheduled layovers knowing they can make up time. The situation reaches an extreme when extra time is provided for a diversion but this does not actually take place as on the current 501 Queen operation.

Update: Subway and Surface Route Performance

Additional details were presented at the meeting by management.

For rapid transit modes, the base period is the first quarter of 2014.

Compared to 2014, there has been a considerable reduction in the number and length of delays with incidents falling by 17.6% and minutes of delay by 37.7%. Broken down by route, the largest drops came on Line 1 YUS and Line 3 SRT, while Line 4 Sheppard showed an increase in incidents.

• Line 1: 29.6% fewer incidents, 41.9% fewer minutes
• Line 2: 0.1% fewer incidents, 9.8% fewer minutes
• Line 3: 54.8% fewer incidents, 79.2% fewer minutes
• Line 4: 22.1% more incidents, 9.2% fewer minutes

These results can be linked to route-specific situations:

• Line 1 now has a new more reliable fleet, the Toronto Rockets.
• Line 3 has the same old fleet, but fewer trains are in service, and a considerable amount has been invested to improve the vehicles and infrastructure on this route.
• Line 4 converted to One Person Train Operation triggering some delays from reluctance by crews to co-operate with this scheme.

However, the big change year-over-year is that the beginning of 2014 was a particularly bad winter which began with the ice storm of December 23, 2013. Because the charts show only 2014 and 2017, not the intervening years, it is unclear whether there was a gradual improvement (for example, as TR trains displaced older equipment on Line 1), or if the weather-related drop in the second quarter of 2014 achieved the lion’s share of the improvement. Lines 1 and 3 have more outdoor operation than Line 2, and Line 4 has none, as one example of the variation in environments which each line must face.

Management are quite keen that there have been great strides in reducing certain types of delay where this is “controllable”. That actually refers to two type of incident:

• Issues arising directly from quality of maintenance and operations such as vehicle reliability
• Counter-measures to passenger-caused issues such as better cleaning of tunnel debris and provision of paramedics during key periods to provide faster response to medical emergencies.

The acronyms in the chart on the right mean:

• SUDP: Disorderly patron
• SUUT: Unauthorized person at track level
• MUIR: Ill customer, medical aid refused
• MUI: Ill customer, transported
• MUPR1: Train in contact with a customer
• SUAP: Assault, patron involved
• SUBT: Bomb Threat

Passenger-related delays have not changed much, and there has been a jump in unauthorized persons at track level  as well as in assaults on patrons. Management noted that passenger illness incidents are most common during the AM peak southbound on Line 1 from Lawrence southward due to crowding. This is an example of a problem which is lumped into “passenger” delays, but which has a cause related to the failure to expand transit capacity, as much a City as a Management responsibility.

On the surface routes, stats for improvement in the two metrics used by TTC Management to rate service show mixed levels of improvement. These measures are:

• On time performance, defined as the proportion of departures from a terminal within a window of one minute early to five minutes late.
• Number of short turns

The values for streetcar routes have improved substantially because running times have been extended on most routes to account for actual operating conditions. One might argue that this has really gone too far in that schedules are now so generous that vehicles must dawdle along a route and still arrive at terminals early enough to cause on-street backlogs of cars waiting to enter loops. This is a case of overkill, and it also creates an expectation of a lengthy terminal break whether time for one is available (e.g. during periods when service is badly disrupted) or not. This also reinforces the impression of streetcars as being an unduly slow mode of transport.

Buses show less improvement on this score both because the problem was not as pervasive to begin with, but also because fewer routes have had their schedules adjusted. In coming weeks, I will review several major suburban bus routes and the degree to which they achieve reliable service.

The one key measure the TTC does not use relates to bunching. They assume that if service leaves on time, it will stay properly spaced. However, both the metric and the TTC’s solution to on time problems mask or contribute to bunching:

• A window of six minutes relative to the scheduled departure allows two or more vehicles to leave together “on time” because the scheduled headway is short, and the variation through bunching does not trigger an “off schedule” report.
• Excessive running time allows operators to take extended breaks and leave late on the assumption that they will make up the time enroute. This may not be “on time” to the target, but the bus or streetcar does not have to be short turned.

The number of short turns has dropped considerably, although it is clear that by late 2016, the TTC had squeezed much of the potential improvement out of the system. This begs the question of whether there is an inherent lower bound to short turns, and the trade-offs between schedule padding and rider convenience (which includes the contrasting aspects of slower travel, but more reliable service to terminals).

Scarborough RT Vehicle Repairs

TTC management recommends that the Board issue a contract to Bombardier in the amount of $6 million for major repairs to 26 of the SRT vehicles. The condition of these cars, now over 30 years old and expected to last another decade until the Scarborough Subway Extension opens in mid 2026, is dire and there is a danger of structural failure.

The repair work is urgently required to address corrosion holes found in the vehicle frames. These holes were found during a vehicle condition assessment to determine the feasibility of operating existing SRT vehicles to 2026. If the problem areas are not repaired as soon as possible, the ongoing stress of service operations may result in fatigue cracks that will jeopardize the ability to safely continue operating Line 3. The fleet does not have spare vehicles that can be permanently removed from service. Waiting for cracks to appear before executing repairs is not an option, as it would result in vehicle shortages and drastically increased repair costs (the frames would then need to be replaced not repaired). [p. 1]

And further:

As a result of the decision to extend the life of Line 3, including rolling stock, the TTC requested Bombardier, the original equipment manufacturer, to perform a review and assessment of the fleet condition in July 2015. While undergoing an assessment of the vehicle structures, perforations and sizable corrosion holes (in the critical high-stress joints of the door post and the car-body frames) were discovered.

A finite element analysis on the frames further determined that the corrosion holes, if not addressed, could lead to fatigue cracks that would compromise the structural integrity of the vehicle’s frame. The existing condition could potentially lead to catastrophic vehicle failure and put the service plan of operating the system until 2026 at risk. The finite element analysis showed that the existing stress levels around the corrosion holes exceed the original intended stress levels. Failure to repair this condition as soon as possible may lead to fatigue cracks, and the need to replace entire frames as opposed to perform a less expensive repair. In order to mitigate and eliminate this risk, the frames must be restored as soon as possible. [p. 3]

Disposition of Toronto Coach Terminal

The terminal at 610 Bay Street and 130 Elizabeth Street is the last remaining asset of what was once Gray Coach Lines Limited, a subsidiary of the TTC, now known as Toronto Coach Terminal Incorporated (TCTI). For many years it was operated by TCTI with tenant bus companies using the space, but more recently the site has been operated under lease by Greyhound. This transferred the ongoing operating cost to the lessee and has resulted in some reduction in the long-standing loan from TTC to TCTI that covered its past losses.

The lease arrangement was from 2012-17 with two two-year extensions. Greyhound expects to decamp to the new Metrolinx terminal near Union Station midway through the second extension period, 2019-21. The TTC will declare the property surplus when Greyhound leaves, and it will be transferred to Build Toronto as per the standard City policy for property matters. Proceeds from any development would be used to clear off the $7.7 million debt from TCTI to TTC, and the remainder would be available to fund City projects.

The TTC would finally be out of the interurban bus business whose history reaches back into the days of “radial” carlines predating the TTC’s creation.