The TTC Board met at City Hall on Wednesday, February 27.

There was also a meeting of the Audit and Risk Management Committee at TTC Headquarters, 1900 Yonge Street, at 9:00 am on Tuesday, February 26 with many items that are also on the full Board’s agenda.

The City Auditor General’s report on Fare Evasion was on both agendas. Given its length and detailed content, I reviewed it in a separate article. An update on actions taken by the Board is included below.

Also in this article:

• CEO’s Report for February 2019
• Proposed Limitations on Public Speaking Time at Board Meetings
• Leasing of 640 Lansdowne Avenue (former Lansdowne Carhouse)

Auditor General’s Review of Fare Evasion (Update)

Toronto’s Auditor General, Beverly Romeo-Beehler, presented her report at both the Audit & Risk Management Committee and to the full Board. The video of the Board presentation and discussion is available on YouTube.

A thread in the presentation and the following Q&A was the question of “what is an acceptable average level of fare evasion”. TTC historically claimed 2% or below as the rate, while the actual average found in the AG’s study was over 5%. The surprising factor in the discussion was that nobody attempted to claim the TTC figures were gospel, or that the AG’s numbers must be out of whack. This is quite different from past Board discussions where defending old reports and methodology took precedence.

The AG hopes that TTC can get to somewhere between 5 and 2 percent evasion, maybe down by 1 percent in the short term. Her goal is that the TTC can lower the evasion rate as much as possible without affecting service.

One point missed in the discussion was that over half of fares are on passes where there is no incentive to evade. Therefore, the evasion rate on the remaining rider classes must be much higher.

Commissioner Shelley Carroll noted that is widely known Presto has problems, and worried that this leads to culture that the system is easy to beat. What does the TTC have to do to turn this around? Later in the meeting, Commissioner Alan Heisey remarked on his recent experience with students boarding buses and streaming through subway entrances without paying.

In phase 2 of her study, the Auditor General will look at equipment reliability in more detail to put anecdotal information about machines not working in context, and she aims to report in October 2019. She argued that a culture shift is needed so that TTC management has support from the Board to make changes, and hat support shhould come from Metrolinx in taking ownership of their problems. Issues such as the colour of passes need to come to the Board table for ownership and support, not simply be accepted as how things are.

The AG also observed that the TTC paid $56 million for fare gates, but if it is losing $40 million in revenue through evasion, some change in fare collection is needed.

The initial AG report claimed that Metrolinx did not want to issue unique cards by user class. However, the actual contract with the TTC includes paragraphs about this for every fare type. The text below is assembled from the relevant parts of the contract.

There was a sense through many comments that Presto was central to fare evasion problems, and this may well overstate the situation. There is a split responsibility in system design and controls as opposed to monitoring of actual usage and system management. What is not clear is the degree to which each aspect works against good fare control – bad design and management by Metrolinx as opposed to the actual level of inspection, enforcement and maintenance by TTC.

Commissioner Lalonde observed: [The full quote if at YouTube]

We all know that the Presto system is at the root cause of a lot of the issues we’re dealing with around fare collection right now. I come from a system where … systems like this require “five nines” performance at all times. We’re struggling to get “one nine” with Presto which is clearly unacceptable.

He went on to ask whether staff need better support from Board to ensure that Metrolinx takes the matter seriously.

I had already asked Metrolinx about the lack of a Service Level Agreement, and they responded:

We are still finalizing our SLA with the TTC to ensure it is in place once the rollout is complete, per our agreement. As a result, we do not have an SLA to share at this time. [Email from Fanny Sunshine, Metrolinx Media Relations, February 25, 2019]

In other words, there is no SLA, and one will not come until the Presto rollout is complete. There is nothing to measure system performance against until that time. This is like buying a house with a leaky roof that isn’t officially a problem because the landscaping is not yet completed.

Deputy CEO Kirsten Watson noted that there is a challenge in closing the reliability gap to 99% with TTC’s partners Presto [fare system] and S&B [Scheidt & Bachman].

We do not find ourself moving forward as quickly as we would like.

Commissioner De Laurentiis asked the Auditor General whether for phase 2 she could look at what technical standards should be expected of a payment provider/owner for fraud prevention. The collaborative approach used by the TTC with Metrolinx has not gotten them very far, according to the Commissioner. The AG replied that she will look at the contract, service levels and the governance process.

Metrolinx has handed a responsibility to the TTC (verification and enforcement) without giving them the tools to manage these issues. There was a strong suggestion that the relationship with Metrolinx has not been as rosy as presented in past when avoiding embarrassment to the provincial government was an overriding concern. Although there is a split responsibility between the TTC and Metrolinx, the AG felt that it is up to the provider to ensure that problems are handled to the degree possible when cards are issued (i.e. in the system design), not only after the fact as a matter of enforcement.

De Laurentiis stated that the Metrolinx team has refused to or at least not responded to invitation to meet with TTC and that collaborative discussions “have not done the job”. She urged management to “go get ’em”.

Motions:

By Commissioner De Laurentiis:

• The TTC Board direct the TTC Chief Executive Officer to use all tools and means available to him to ensure Metrolinx addresses the operational and policy issues identified by the TTC and the Auditor General, and resolved to the satisfaction of the TTC in the interests of its customers.
• That the service level agreement with Metrolinx be assessed against a standard expected of a supplier and owner of a payment card system, such as a debit or credit payment system.

By Commissioner Lalonde:

• That the TTC Chair request that the CHair of Metrolinx agree to co-sponsor an initiative to raise Presto performance standards to contracted levels.
• That the President & Chief Executive Officer of Metrolinx and the Chief Executive Officer of the Toronto Transit Commission report back to the TTC Board on this initiative no later than Q3 2019.

By Chair Robinson:

• That the TTC Board direct the Chief Executive Officer, Toronto Transit Commission, to conduct a cost-benefit analysis of the TTC’s major fare categories, utilizing new Presto data, as part of the Fare Pricing Strategy Report expected in Q4 2019.
• That the TTC Board direct the Chief Executive Officer, Toronto Transit Commission, to initiate a review of the Child Presto card fare category data, including the following, and report back to the TTC Board in Q3 2019:
  • Overall Presto card usage by children under the age of twelve.
  • Station by station usage of the Child Presto card.
  • A specific, costed action plan to reduce Child Presto card fraud and evasion.
• That the TTC Board Chair, Toronto Transit Commission, write a letter to the Chair, Metrolinx reaffirming the Board’s commitment to ensuring that Metrolinx equipment meets the contractually-required service levels and establishing a final Service Level Agreement to ensure future service standards.
• That the TTC Board direct the Chief Executive Officer, Toronto Transit Commission, to advise the President & Chief Executive Officer, Metrolinx that the TTC intends to be involved in meaningful consultations as a key customer of Presto equipment and report back to the TTC Board in Q2 2019.
• That the TTC Board direct TTC staff to review the status of all service level agreements for Presto equipment and report back to the TTC Board in Q2 2019.
• That the TTC Board direct TTC staff to work with Metrolinx to implement controls for all concession cardssimilar to the controls recommended by the Auditor General with respect to child cards.

By Commissioner Bradford:

• That the TTC Board request a response from Metrolinx on the timeline for re-introducing debit/credit payment on Presto fare and transfer machines.
• That the response be brought to the April 11, 2019 TTC Board meeting.

CEO’s Report

The CEO’s Report is now in a standardized format with similar page and chart layouts for each section. This may improve the look of the report, but another change is more important. The CEO no longer provides financial updates in his monthly report. These have been moved to the quarterly reports from the CFO of which the first appeared at the January 24, 2019, Board meeting. This covered the first three quarters of 2018.

This change means that although ridership numbers are reported monthly for the period about 1.5 months back, revenue and expense numbers show up with a longer lag. Given the controversy about whether “ridership” as reported by management accurately reflects the health of the system, this is a backward move.

[Apologies for the resolution of charts below. This is a limitation of the pdf in which they were published.]

For all of 2018, ridership numbers tracked below budgeted levels, and there was a pronounced dip in December. Although figures for early 2019 have not yet been released, the extremely cold weather and storms are likely to depress counts over what might have been achieved in a milder winter.

Ridership for 2018 totalled 521.4 million, 18m (3.3%) below budget and 12.3m (2.3%) below 2017 results. These numbers are suspect both because of the way Metropass sales are translated to “rides” and because of the shift to Presto plus the two-hour transfer.

With the availability of monthly passes on Presto, the TTC had budgeted for strong growth in Presto usage in late 2018. This did not materialize as many Metropass users held on to the bitter end with their old, but simple and reliable technology. In early 2019, with no other option, former passholders migrated wither to Presto monthly and annual passes, or reverted to pay-as-you-go fares, and 77% of fares were paid with Presto in January 2019. A side effect of the lower-than-expected takeup of Presto in 2018 is a saving in fees paid for that system’s operation in the 2018 budget year. This is a non-recurring “saving” which will not be available in 2019 and following budgets.

Service Reliability: Rapid Transit

Performance metrics for the subway system are now reported based on two factors:

• “On time” performance at terminals, and
• Proportion of scheduled capacity delivered during peak periods.

“On time” for the subway is measured relative to the planned headway (time between trains), not to the location and time set out in the operating schedule. This gives a measure that reflects what riders see. There is, however, a major shortfall in the “on time” stats in that they are reported on an all day basis for all weekdays in a month. This eliminates any effects based on time of day (peak vs off peak, am vs pm) as well as day-to-day variations. To put it another way, hitting a 90% average is meaningless of most of the “good” trips occur off peak and the “bad” ones are concentrated in a few hours per day, or a few days through the month.

There is a point beyond which uneven headways affect the riding experience quite severely, but the charts in the CEO’s report give no way to see how often or when this happens.

Line 4 has an almost perfect score because there are rarely delays, and the schedule has generous provisions for terminal recovery time.

The measure of actual vs scheduled capacity is constrained to 7 hours/day (6-9 am and 3-7 pm), but has the same problem with monthly totals hiding the degree of fluctuation. Moreover, the metric is calculated from data at many points on each route. For Line 1 this is:

Total number of trains that travelled through 12 key sampling points during a.m. and p.m. peak as a percentage of trains scheduled.

The number of sampling points for other lines is 10 on Line 2 and 2 each on the short Lines 3 and 4.

Service reliability is all about providing the expected service all of the time, not just “on average”, and especially at key congested locations.

Service Reliability: Surface Routes

The metrics for surface modes have not changed and are still reported relative to scheduled service.

• “On time” departures from terminals, and
• Number of short turns.

“On time” for streetcars and buses is defined as leaving a terminal no more than one minute early or five minutes late. This metric allows bunching on routes to be reported as “on time” especially when the scheduled headway is less than the 6 minute “on time window”. When this bunching occurs at terminals, the situation typically gets even worse along a route as more vehicles catch up to the parade where the leader is carrying a wide gap. Most riders see the gap, not the string of vehicles which are “on time” by TTC calculations.

For example, on a three minute scheduled headway, four buses running 1 minute apart could be considered to be “on time”:

1. Scheduled at 12:00, actual 12:05 : 5 minutes late
2. Scheduled at 12:03, actual 12:06 : 3 minutes late
3. Scheduled at 12:06, actual 12:07 : 1 minute late
4. Scheduled at 12:09, actual 12:08 : 1 minute early
5. Scheduled at 12:12, actual 12:17 : 5 minutes late

In this example, a parade of four buses followed by a nine minute gap, are all “on time”. For wider scheduled headways, a parade cannot be as long and still all be “on time”, but the same principle applies. On a six minute headway, two buses could run together (one five minutes late and its follower one minute early) with both “on time”. The resulting service could be a pair of buses every 12 minutes, far from the advertised service quality.

The metric only serves to overstate the quality of service riders actually experience. A related problem is that bunched vehicles will vary greatly in crowding levels, and most riders will be packed on (or trying to board) the full bus.

As with subway metrics, the use of monthly, all-day values hides the degree of variation by time of day, and for various days through the month. Consolidating stats for the comparatively quiet periods of evenings and weekends with peak periods yields pretty charts, but hides a lot of important detail.

Figures for the two surface modes cannot be directly compared because the streetcar network operates in congested areas where there is a lot of construction underway, while many parts of the bus network run in comparatively benign conditions.

Streetcar On Time Performance:

Bus On Time Performance:

Short turns are reported as a count of events, not as a percentage of total trips operated. Obviously if the count goes down, this is better for riders, but raw counts do not reveal the extent of the problem relative to overall service. Again, as with previous charts, averaging data on an all-day basis hides both the effect on riders during various times of day, and the degree to which service reliably gets to the end of the line day-to-day.

A further problem with this metric is that the goal is to “reduce short turns”. This results in schedules that are padded with extra running time so that it is very difficult to run late, and operators dawdle along routes for fear of running “early”. In a worst case, short turns that are needed to restore service do not happen and large gaps are left to sort themselves out. The short turn counts look good, but service to riders is not.

Streetcar Short-Turns:

Bus short-turns:

Vehicle Reliability: Rapid Transit

For rapid transit lines, a vehicle “failure” is one that causes a delay of five minutes or more. In the charts below, mileage is counted as car-km, not as train-km.

The T1 fleet serves Line 2 Bloor-Danforth. There is a substantial pool of spare equipment because original plans called for some of the T1 fleet to remain on Lines 1 and 4. When this changed, the surplus trains went to Line 2 where they triggered a storage capacity problem at Greenwood, and required the re-opening of Keele Yard. These trains were delivered between 1995 and 2001, and were originally planned for replacement starting about 2026. However, the TTC’s recent 15 year Capital Plan now shows this fleet as being rebuilt, at least in part, for an extended life. Reliability through 2018 has been gradually rising over 2017 thanks to an overhaul of door pocket guides that were responsible for many problems.

The TR fleet serves Line 1 Yonge-University-Spadina and Line 4 Sheppard. As on Line 2, there is a substantial pool of spare trains because of purchases for future growth following conversion to automatic train control (ATC). Although there is a recent growth in reliability numbers, they are still below the high points through 2017. Note that there are wide swings in the MDBF values month-to-month because the number of incidents is small.

Vehicle Reliability: Surface

The new Flexity low floor streetcars continue to perform much better than the cars they replace, but still not near the target set for them by contract with Bombardier. As with the subway, a “failure” is defined as an event which causes a service delay of five minutes or more. As for the older CLRVs and ALRVs, their reliability continues to be poor, and there are only a few ALRVs still  in service. Recent stats for the old fleets are as good as they are only because the least reliable cars have been removed from service. Note that the vertical scale in the three charts below is not the same.

Flexity Reliability:

CLRV Reliability:

ALRV Reliability:

Current plans call for all of the “legacy” fleet to be retired in 2019 although this depends on Bombardier completing its Flexity shipments this year.

A measure of overall reliability is the number of times a vehicle requires a road call or change off (replacement) while in service. The target is 1.5% of the peak cars in service or 2.4 per day on average based on peak service of 160 cars. The actual numbers are much higher as the chart below shows, although there is some improvement over 2017 thanks to retirement of the worst of the old fleet.

Streetcar Road Calls and Change Offs (RCCO) (entire fleet):

For the bus fleet, the situation is much rosier thanks to the large-scale replacement of the oldest buses and the new life-cycle target of 12 rather than 18 years. The fleet has been running consistently at an MDBF greater than 20,000 km since February 2018.

Bus Reliability (entire fleet):

Bus RCCO (entire fleet):

The number of road calls and change offs reflects the much greater reliability of the new, comparatively young bus fleet, and is running close to the 1.5% of peak service target.

Proposed Limitations on Public Speaking Time at Board Meetings

This item was held over to the April 11 Board meeting.

Although it was not discussed, an important issue here is that the TTC is, for all practical purposes, very much like a committee of Council and has for decades entertained deputations. This dates back to August 1972 when a citizens’ group who wanted a bus on Jones Avenue climbed up six flights of stairs after they were been barred from elevators to reach the Board Room at 1900 Yonge Street. There they argued that under the Municipal Act the TTC Board meetings were open to the public. Thus began the era in which at least some TTC business was carried out in view of more than the odd member of the Press Gallery who showed up.

The TTC would do well to consider making its procedures match those in the City of Toronto Municipal Code Article 4 (beginning at p15 of the linked document). This has more than adequate provision for limiting speaking time without building in a hard cap that could hamstring attempts at participation.

Leasing of 640 Lansdowne Avenue (Former Lansdowne Carhouse)

Lansdowne Carhouse was a Toronto Railway Company facility a few blocks north of Bloor Street that opened in 1911, comparatively late in that company’s 1891-1921 franchise. When the TTC took over municipal operations in 1921, the site passed to them and, unlike many older TRC buildings, Lansdowne Carhouse was in good shape. It remained in service housing streetcars, trolley coaches and diesel buses over the years, and closed in 1996. The building survived until 2004, and since then the site has been used for storage including, recently, by the TTC’s Streetcar Maintenance and Infrastructure Department for pre-assembly of track panels for intersection replacement work.

Underground pollution from a nearby General Electric site limited the use of this land, particularly the western portion.

The Wade Avenue Yard, originally a streetcar and later trolley coach facility, was a block south of the carhouse. This land has been occupied by housing for a few decades. Now, the eastern part of the carhouse property will also be converted for affordable housing and long term care.

The Streetcar Infrastructure operations currently at Lansdowne will move to a new, permanent location at 301 Rockcliffe Road in the Black Creek area not far from the Streetcar Overhead Wiring Section at 391 Alliance Avenue.

The western part of the site will be leased to Metrolinx as a staging area for their work on the Davenport Diamond grade separation project (which is nearby to the north) from June 2019 to December 2023.

Here is the TTC’s official response on the status of this property:

Remedial action is required when there is a change of land use and you need to file a Record of Site Condition with the MECP. Sensitive land uses (ie residential) have a higher standard to meet than non- sensitive uses.  As part of the redevelopment the clean-up and  mitigation measures will be completed to the satisfaction of the Ministry and a Record of Site condition will be filed for the residential portion.  No cleanup works have been completed related to the GE dispute, nor are anticipated for the remainder of the site, which will continue to be regularly monitored. Soils impacted by historical diesel fueling activities in the south west corner of the site, unrelated to the GE dispute,  were excavated and removed in 2001-2002, to control off-site impacts to the south and west of the property. [Email from Susan Sperling, TTC Communications, February 26, 2019]