by

TTC Audit Committee: March 11, 2026

The TTC’s Audit & Risk Management Committee will meet in the Boardroom at TTC Headquarter, 1900 Yonge Street, on Wednesday, March 11 at 9:30am. Two items on the agenda are of considerable interest:

The review of customer performance, metrics and real-time information covers at length the many problems with passenger information from the TTC including its website and various outlets for notices including social media and apps. This is an unusually candid review and long overdue.

Also included are charts showing the status of various recommendations from TTC and City audits, as well as from outside reviews. However, these are only summaries and there is no link back to source documents to cross-reference specific items, their severity and status.

Two confidential attachments address “The Subway Tunnel Maintenance and Rehabilitation Audit” as well as some findings from the UITP review.

This article will be updated as additional material is posted, and after the committee meeting.

Audit Plan Status Update: Overtime

The main report primarily lists the status of various audits completed or in progress, but it adds a new one to the 2026 calendar: a review of overtime costs.

The report observes:

Our analysis identified that the TTC’s overtime expenditure grew by approximately 19% from 2019 to 2025, reaching $92.7 million in 2025 compared to $77.7 million in 2019. This increase is primarily driven by unionized employees, whose overtime costs rose by 24% during the same period, reaching $74.1 million in 2025 compared to $59.6 million in 2019.

It also notes that overtime as a percentage of total costs is lower than at agencies such as in New York, Boston and Chicago, but that direct comparisons can be tricky because of differences in organizations and reporting can skew the numbers.

The intent is to focus on areas where variations are substantial:

Despite the favourable benchmarking, the analysis highlights concentrated risk areas where overtime variances and unbudgeted expenditures suggest potential control gaps. These areas include operational and capital departments that have exceeded overtime budgets or experience significant increases in overtime. A targeted audit approach focusing on these high-variance areas is recommended to strengthen governance and optimize overtime practices without incurring the cost and complexity of a broader audit.

Overtime is a favourite target in budget reviews, and some politicians make a career out of demanding that it be reined in. However, in agencies like the TTC there are bona fide reasons for overtime, notably as a way of dealing with the fact that operator work assignments do not fit into convenient 8-hour chunks. The marginal cost of overtime does not, generally, inflate the cost of benefits and so the full dollar cost of an overtime hour is not as high, relative to base pay, as a percentage premium might suggest. However, high overtime costs can also reveal issues with short staffing (and by implication underbudgeting), or of excessive staff absences. The situation is likely more complex than simply saying “cut down on overtime” without understanding the underlying causes.

The Challenge of Meaningful Metrics

The audit report on customer key performance metrics, trip planning and real-time information covers a lot of ground. Several articles on this site have addressed problems with how service and system quality are reported to the Board and to the public. Broadly, there are two ways of looking at data:

  • Internal facing metrics intended to convey the quality and trends for management purposes, and
  • External facing metrics of service quality.

These will reveal different factors. Internal metrics tend to look at the organization as a whole, while external service quality should be measured at a level appropriate to riders’ trips.

As used by the TTC both of these suffer from problems with excessive consolidation of information into averages that hide more than they tell. Some of these deal with issues of vehicle and infrastructure reliability, and I will turn to them later in this article in the UITP [International Association of Public Transport] review.

Riders want current information that is accurate and timely, such as diversion and delay notices. Historic information is meaningless if it averages all routes and time periods together because problems with specific routes are submerged into system-wide averages. This hides the “bad actors” and allows management to claim that service is better than what riders actually experience. There is no way, short of the detailed analyses such as I publish, to see what is actually happening on individual routes.

These issues were driven home by a TTC survey in July-December 2025.

Survey insights informed the audit’s focus and helped distinguish between the types of information customers value for accountability versus those they rely on to manage trips in progress or during disruptions.

Credible information is essential in attracting and retaining riders. TTC’s public information leaves a great deal to be desired, and until recently there was little sense that they even acknowledged the problem.

“Information” is not just a question of posting schedules and hoping that service roughly matches them, but of letting rider know when things change.

This audit examined whether the TTC’s disruption notices, trip planning data, and open data feeds accurately reflect real‑time service conditions across the TTC‑owned channels and third‑party applications. When real‑time information is incomplete, delayed, or inconsistent, customers may encounter service conditions that differ from what is communicated, leading to frustration, poor decision‑making, and reduced confidence in the TTC’s information systems.

It is critical to develop integrated real‑time data ecosystems, where disruption notices, vehicle arrival predictions, and accessibility impacts are shared consistently across digital channels and open data platforms. Customers expect disruption information to be precise, location‑specific, and dynamically updated throughout their journey. Aligning the TTC’s real‑time data infrastructure, disruption processes, and open data feeds with these expectations is essential to meeting evolving customer needs and supporting reliable trip planning.

This is the sort of comment one would expect, possibly a tad less polished and more forceful, from a transit advocacy group or an aggrieved rider tired of waiting in the cold for a bus that will never arrive. Long, long overdue is the recognition by the TTC itself that change is needed.

Key issues flagged by the audit include:

  • Real-time information varies in accuracy, completeness and consistency across various channels used by TTC to alert riders. There are multiple, fragmented, sources for information, some easier to find than others, and a rider may not receive what they need while travelling depending on where they look.
  • Key Performance Indicators [KPIs] for service are aggregated and bear no relationship to individual routes or periods of service. The audit recommends a customer-centric view, alignment with TTC service standards and review of practices at other large transit agencies.

Management intends to improve the quality of data available both through a rider dashboard and to trip planning apps. However, this is not simply a question of knitting together the various data sources on the website, but ensuring that the people responsible for publishing this information do so in a timely and consistent manner. Moreover, the multiplicity of disconnected sources on the TTC’s site can challenge even a hardened transit advocate, let alone a neophyte or occasional user.

There are 10 groups of observations (many more details in the report).

  • Inconsistencies in TTC data feeds affect third-party apps. Real-time subway info is not available to apps, and Line 6 information is hidden from view by a Metrolinx vendor. Future planned diversions are not shown. Some alerts are not structured in a way that can be consistently parsed by apps to determine how they should be handled. [Target completion 2Q 2026]
  • Detour information is not consistently created preventing riders from knowing what is actually happening on a route. [Target completion 2Q 2027]
  • There should be a strategy to engage external developers to ensure changes in available data are understood and incorporated in apps. [Completed]
  • Vehicle arrival times across modes are inconsistent and inaccurate for various reason including the use of a different algorithm to predict bus, streetcar and subway arrivals. Line 6 actual times are not available as noted above (this also applies to Line 5). [Target completion 4Q 2026]
  • Inconsistent display of disruption alerts on TTC screens. [Target completion 1Q 2027]
  • Accessibility alerts give information on specific devices, but not on the overall accessibility of stations. GTFS-RT feeds do not include devices as part of a trip path to identify when an alternate route should be advised. [Target completion 2-3Q 2026. Note that some station accessibility information was recently added to TTC home page.]
  • Fragmented and Incomplete Advisories on the TTC‑Owned Channels. With information on nine separate web pages, riders cannot access information in one place. Some disruptions are not published in the feed on X/Twitter. [Target 3Q 2027]
  • KPIs in the CEO’s report do not reflect rider experiences. TTC should create a customer dashboard with revised metrics such as those published by other major systems. [Target 1Q 2027]
  • KPI reporting masks peak-period variability and route-level gaps. The customer dashboard should have the ability to disaggregate KPIs by route and time of day to highlight individual locations and problem areas. [Target 1Q 2027]
  • Service standards are not reflected in KPIs undermining accountability. [Target 1Q 2027]

None of this will be news to riders or advocates who for years have called for improvements. The TTC’s website and related services have been a mess for a long time even though they were very proud of it when originally introduced. One cannot help asking how the people responsible were left in their jobs for so long, or why both management and the TTC Board seemed unconcerned with the situation.

The site bears the hallmarks of something that grew over the years with various functions grafted on for different user departments and no attempt at overall consistent navigation for users.

Other problems not mentioned explicitly by the audit include the posting of laughingly inaccurate information for service disruptions including locations and diversions that are physically impossible. That says something about the city knowledge of whoever posted the alerts.

Another issue is the discrepancy between posting updates to the electronic versions of schedules (the GTFS feed) and actual implementation dates. For example, the changes triggered by the Bay-College construction work will not show up online for two more week until the next official schedule change. It’s not as if the TTC didn’t know this was happening, but they don’t have the procedures in place to update various info sources outside of a standard schedule [aka “board”] period in the calendar.

TransitApp figures out from actual vehicle behaviour what is going on, but that’s one app, and this functionality is not part of the general offering. Moreover, without an update on the TTC’s end, the app has no way of knowing where the official stops have moved to.

With respect to the Service Standards, it is important that these are understood, including their shortcomings and data constraints. For example, the definition of “on time” is sufficiently generous that on frequent routes (also by definition the busiest), bunched service does not violate the “standard”. Another problem is that the crowding ranges used to show vehicle occupancy do not match the ranges specified in the standards, nor do they distinguish between peak and offpeak.

Overall, this is far more than a set of problems to be resolved by IT re-engineering. There is an organizational issue with information provision and with the toleration (or blissful ignorance) of shortcomings in existing procedures and services. Rarely does one see such an extensive list of issues with any functional area inside the TTC and it speaks to how a key part of transit service provision has been ignored for years.

Workcar Hydraulic Leaks

This item is not yet available online. However, it relates to subway infrastructure maintenance both for the specialized cars themselves, and to work that might not be completed promptly because the needed workcars are not available. The TTC has a plan to renew and expand its fleet, but this will not happen overnight and Toronto faces many slow orders to make up for deteriorating track and deferred maintenance.

Updates to follow.

UITP Peer Review – Management Response Update

In 2025, the TTC considered a Peer Review by consultants from the International Association of Public Transport (UITP), an organization of which the TTC is a member. This was not a full system review but rather an examination of how the TTC tracks maintenance needs and manages its fleet and infrastructure. This caused the review to focus in part on Enterprise Asset Management, a large IT system now in implementation, and somewhat less on front line practices.

The current report cites four high-priority items:

  • Road traffic congestion
  • Operational Speed of Streetcars
  • Streetcar Overhead Linesperson Maintenance
  • Streetcar turnouts and double point switches

However, many other items are discussed in the appendices.

Road Traffic Congestion

Road congestion is seen mainly as an issue for the City to resolve through changes in road space usage, parking, active management with traffic wardens, and signalling. A City report is expected in 2Q 2026.

Operational Speed of Streetcars

The operational speed of streetcars (or lack of it) results from many factors, and the TTC appears to regard this as an external issue. However, it is directly linked to operating policies such as slow orders on all junctions, stop service times and a decline in typical operating speeds since the retirement of previous generation of streetcars (CLRVs).

Streetcar Overhead Maintenance

The UITP recommends that streetcar overhead line maintenance be performed under “power off” conditions. This has implications for minor, localized repairs where a power cut could affect a wider part of the system. Some major work is already performed with power off, but not all of it. In particular, this would make the ongoing conversion for pantograph operation more intrusive on operations than it already is. The TTC’s position is that line maintenance personnel are well-trained in working on live overhead and have appropriate equipment for this work (just as Hydro line workers often work on live cables to avoid power cuts).

Streetcar Track Switches and Controls

The issue of streetcar switching has received a lot of debate in various quarters, but with too much focus on the physical switches and less on the control systems. UITP recommends the use of automated double-point turnouts replacing the single-point versions common on the TTC and some older street railway systems.

The “stop-check-go” procedure at all facing point switches developed with the installation of a new switch control system to accommodate the introduction of the ALRV fleet. Until then, all streetcars were roughly the same length and an overhead contactor for switch controls was used. With two different lengths of cars (or more critically, different distances from the front of cars to the trolley shoe), a new system was needed. This was implemented using pavement loops and car-borne transmitters, but this scheme had enough reliability problems that switching became suspect. This was compounded by the declining quality of some intersections which triggered permanent slow orders on all junctions.

When the Flexitys arrived, slow operations at switches were already the norm, and had nothing to do with whether there was one or two blades in the switches. Double-blade switches would improve truck dynamics, but would not address issues with switch controllers or general track conditions at junctions. A program to replace old switch controllers has been underway since 2017, but still has several years to run based on spending plans in the Capital Budget. The specific recommendation is:

It is recommended that certain specific turnouts should be converted to motorized two-switch operations, activated by the destination or route transmitted from the vehicle. Each location should be linked to locking detection of the switch and associated confirmation by visual route indicator signal.

An obvious extension of this scheme would be integration with transit signal priority including turn protection.

The recommendation continues:

The initial choice of turnouts for the upgrade should be directed to those turnouts used for the standard routing of vehicles. Eventually, all turnouts should be equipped with this system. The investment required should be part of the Capital Plan.

A shortcoming of existing electric switch installations is that TSP is not provided at most locations, and is only present for standard “scheduled” moves. However, there are many common diversions and short turns that would benefit from TSP, but these are treated as second tier locations. Another issue is that if a switch is disabled (typically for a construction project), any associated TSP ceases to work because it is the route selection that calls out the “white bar” signal.

Improved switching and transit signalling will benefit streetcar operations, but they must be implemented as a suite of changes, not simply a project to change out single for double-blade switches.

Maintenance KPIs

TTC regularly reports the Mean Distance Before Failure (MDBF), but the UITP recommends also tracking Mean Time to Failure and Mean Time To Repair. The latter is particularly important if an equipment failure takes a vehicle out of service for an extended period either because of the complexity of repairs, or because of supply chain issues with part. This functionality will be incorporated in the new Asset Management System.

Something notable by its absence is any reference to tracking vehicle performance by group — manufacturer, age, duty cycle. This is important in comparing parts of the fleet to each other, and for understanding how buses operating on busy but slow routes might have different metric values than those on faster running lines with fewer stops.

There is no discussion of fleet availability and utilization, notably of how many vehicles are actually available for service compared with schedule needs. TTC has a history of operating with spare margins above industry targets (commonly 20%). This represents unused capital and gives the impression that more vehicles are available for service than is actually the case. Moreover, if the “lemons” never leave the garage, they never fail, and performance metrics are artificially inflated.

The City produces an annual report on its various fleets, but there is no comparable report for the thousands of TTC vehicles.

Non-Revenue Fleet

The fleet of non-revenue vehicles is an important part of maintenance activities, particularly on the subway. The persistence of some slow orders is a direct result of having insufficient equipment to handle all of the projects across the network. TTC has plans for new and replacement cars, but this is going slowly and it is not clear if the plans and funding keep up with current and future demand.

This is related to the workcar hydraulics report to come as part of the audit plan review (above). If cars are unreliable, then work schedules cannot be achieved.

Leave a comment