As part of Transit Committee's motion requesting a plan to achieve reliable service by the end of 2027, staff presented a report on improving bus service reliability at Transit Committee on 11 September.
The relevant document is the report, linked in PDF form.
OC Transpo defines reliability with three headline statistics, which bundles the entire network into one number. The trips delivery metric measures the percentage of trips which were not cancelled and ran at all; regularity measures the gap between buses on frequent routes [it is not clear if this is applied to frequent routes outside frequent operating periods], with a wide band of 40% allowance; and punctuality, which is on-time performance for less frequent routes, with an allowance of one minute early/five minutes late.
Both Transit Committee and OC Transpo love to talk about their commitment to reliable service for Ottawa residents. But OC Transpo has a poor track record of presenting good information to Transit Committee that would enable action; Transit Committee has a poor track record of taking action to improve reliability. To improve bus service reliability, Council must begin to understand that there is no free lunch, and make the difficult choices to improve service.
Discussion
[At] Transit Committee in April 2025, OC Transpo outlined the following accomplishments that were achieved in 2024:
• Hired 222 bus operators and 12 Para Transpo operators
• Increased O-Train Line 1 service delivery from 97.1 per cent to 98.8 per cent
• Increased bus service delivery from 97.8 per cent to 98 per cent
• Added new zero-emission buses to the fleet with more arriving in 2025
When OC Transpo brags that reliability has increased, it must demonstrate that reliability has increased. To do so, they must provide useful data to analyze.
OC Transpo does not provide useful data. OC Transpo's statistics for "increased bus service delivery" only says "in 2024," but does not provide any useful information on which time period it is referring to. A more useful comparison may be the YTD service delivery; from January-August 2024, the service average was 98.3%; in the same period in 2025, that number was 97.4% (it was 97.5% in 2023). After a few years of improvement from the nadir in 2021 to 2024, cancellations are sliding again.
The report is structured around three primary requirements to provide reliable bus service for customers:
• Sufficient buses and staff available to deliver service every day
• Buses ready and able to start every trip on time
• Conditions to allow consistent travel time over the length of every bus route
"Sufficient buses and staff available to deliver service every day"
For Fall 2025, OC Transpo has 106 regular bus routes plus 73 special school routes, providing 7,543 separate bus-trips per weekday, serving over 480 square kilometres of the city of Ottawa and central parts of Gatineau. For each trip there must be a schedule for the bus to follow, a set amount of work hours for each operator and then, on these routes, there must be a safe and reliable bus available every morning, and a series of bus operators to drive the bus and assist customers through the full day.
Currently, the OC Transpo conventional bus fleet is significantly aged, with just over half of the fleet past its expected useful life ... The oldest half of the OC Transpo bus fleet is in the process of being replaced. In all, 350 new buses are on their way, with delivery schedules between 2025 and 2027 ... If the new buses achieve the same reliability as the newest diesel 40-foot buses from 2019-2021, bus availability for service will improve, service delivery will increase, and bus service reliability will improve as a result.
That's a big if. According to reports from elsewhere, electric buses have limited range, therefore requiring more buses to operate the same amount of service.
As reported at Transit Committee, the Nova LFS's are being driven significantly more than anticipated, and will likely need replacing ahead of schedule.
Staff have built the long-term fleet plan, coordinated with the Long Range Financial Plan for Transit, to purchase replacement buses progressively and continuously, rather than in large numbers as was done for articulated buses in 2010 and is currently being done with the battery-electric buses that are replacing those 15-year-old buses. With a nominal bus fleet size of 750 and a planned replacement of buses after 15 years, there would ideally be 50 replacement buses purchased each year.
This would be a major improvement from the current procurement system, and prevent fleet meltdowns such as what is currently ongoing. However, it requires council to fund the purchase of ~50 buses every year, is cannot be counted on.
It requires City Council to understand that $50 million of annual purchases will ensure reliable fleet availability, and that the tradeoff of fleet deferral is semi-regular fleet meltdowns and a $500 million purchase once or twice a decade.
As the new battery-electric buses arrive, a series of critical software upgrades are in progress to manage the new characteristics of these vehicles ... Upgrades to the computer-aided dispatch system ... A new yard management system ... A new energy management system ... [and] An update to the current scheduling and work management software.
I would be skeptical.
Maintaining buses also requires there to be enough skilled licensed truck and coach technicians to carry out all the required certification, preventative and repair work. As staff have reported to the Transit Committee, industry best practices would require the City to employ approximately 188 certified truck and coach mechanics, based on current labour hour requirements. OC Transpo currently has 138 on staff and has been conducting a targeted recruitment campaign seeking to quickly address the shortage of licensed technicians.
This shortage was not, as far as I am aware, made public until the Auditor General's Office released its investigation last year. It should not take the Auditor General to get staff to be more transparent about issues building up at OC Transpo; City Council must shoulder much of the blame for being unwilling to confront the hard choices that building a reliable service involves.
Transit is very expensive, and they do not want to pay, but also do not want to hear of service reductions. This is how OC Transpo has gotten to its current state.
Finally, launching service every morning and keeping it available all day requires that there be enough bus operators available to work each day. During 2022 and 2023, staff reported to the Transit Commission that there were too few bus operators employed by the City and that a more intensive recruitment campaign and an expanded training program were being put in place. As staff reported in October 2024, that recruitment plan was successful in hiring 828 Operators ... After accounting for factors such as vacation, sick leave, acting assignments, and other absences, the number of available bus operators was estimated at approximately 84 per cent of the total budgeted positions in the first half of 2025.
OC Transpo's own presentations to Transit Committee show that operator availability is a large component of cancellations. An operator who is not scheduled for service does our reliability no good. To schedule operators requires that budget room be made for them.
Because delays propagate, this affects on-time performance and bunching far beyond the small proportion of trips cancelled due to lack of operators.
The following factors can reduce the availability of buses and operators or can delay improvement in the availability of buses and operators. These are factors which plans must account for, or which can address how successful Transit Services can be in achieving set targets:
• Changes in the labour market and the attractiveness of transit as a career choice.
• Delays in the supply of new buses by the manufacturers.
• Unexpected defects in the bus fleet ... New characteristics of having battery-electric buses in the fleet
• The moment-to-moment availability of the right type of bus for each trip can be affected by the timing of ongoing maintenance activities.
• [The] industry shortage of licensed 310T technicians could impact OC Transpo’s ability to meet daily bus service availability requirements.
The job market is awful. This should improve OC Transpo's ability to attract and retain mechanics, not harm it; if they cannot do retain mechanics at 7% unemployment, they certainly will not be able to do so in good economic conditions.
"Buses ready and able to start every trip on time"
Factors that can delay the ability of a bus to start its next scheduled trip on time include buses operating in mixed traffic where the schedule may be susceptible to traffic congestion, collisions, construction activity, poor weather, and detours.
For short-term delays, the Transit Operations Control Centre (TOCC) provides real-time solutions, such as assigning a standby bus to cover a trip that is susceptible to being cancelled, asking operators who are finishing their trip to extend their work and accept overtime to cover a trip, or respacing buses on frequent routes to avoid long gaps in service.
For long-term predictable but temporary delays, such as routes operating through longstanding construction zones, additional resources are required as the route will have to travel longer distances through a detour or spend more time operating through the delays of the construction.
This is correct. However, these fixes require resources, which means increasing funding and improving operations, or reducing nominal service and reallocating the service hours.
Preventative improvements are those that would reduce the incidence of buses being delayed to the extent that they are not available to depart on time. Improving the mechanical reliability of buses, as outlined in greater length under the previous requirement, will reduce the likelihood that a bus suffers a defect that prevents it from completing its previous trip. Scheduling more recovery time at terminals, at a cost, can give more buffer time to absorb variation in the time that buses arrive there. Scheduling more buses and operators to stand by and be ready to fulfil trips which are at risk of 14 being delayed or cancelled will also help, again at a cost.
Here it is: reliable service, whether that's consistent bus procurement, extra runtime, or transit priority, requires making tradeoffs and sacrificing something else. Instead, Transit Committee is too stupid (or malicious) to understand this. If Council will not increase the service budget and hold staff accountable for using the budget well, then the tradeoff must be understood and acted upon. Council has yet to even internalize it.
Improvements to ensure that buses are available to start trips on time can be made under two major categories – preventive and reactive.
Preventative improvements are those that would reduce the incidence of buses being delayed to the extent that they are not available to depart on time. Improving the mechanical reliability of buses ... scheduling more recovery time at terminals ... scheduling more buses and operators to stand by.
Reactive improvements are those that enhance the ability for staff to deal with a delay that has occurred and to get the next trip away from its starting point on time. Increasing the number of standby buses at key locations has a budgetary requirement ... Taking buses away from scheduled service to increase standby buses improves reliability but reduces the resources available for planned and scheduled service.
The improvements listed are correct, but the categorization of preventative/reactive is not a useful distinction. The stated measures are the same and should be treated as such.
Staff are also monitoring advances in software that may soon be able to predict delays based on machine learning about traffic patterns and other influences and to provide advice on options that controllers might select. If these innovations prove reliable, they will enhance monitoring of service quality and speed decision-making in the TOCC.
This is a basic requirement of transit service. It's called a "line chart," and requires zero advances in software.
Service quality and decision making will not improve if buses are underscheduled, they propagate delays randomly, or there are no extras, even if they could be predicted with 100% accuracy (and that is not possible). This is a cop-out, not a solution.
An end to random interlining would be the cheapest, fastest, and most reliable way to "predict delays" on transit routes. This would ensure that delays do not spread across the network, and can be managed with various tools (line management, cancelling trips) without spreading those effects across the entire bus network.
The following factors can impact OC Transpo’s ability to ensure that buses are available to start trips on time:
• Legislative changes can affect how trips need to be scheduled.
• All the mechanical reliability risks
• Scheduling more recovery time and scheduling more standby buses or operators both require additional capital funding ... and operating funding
• The moment-to-moment availability of the right type of bus for each trip can be affected by the timing of ongoing maintenance activities.
• [The] industry shortage of licensed 310T technicians could impact OC Transpo’s ability to meet daily bus service availability requirements.
Some of these risks are predictable, such as legislative requirements. The reasons for the consequent increase in budget should be clearly and accurately communicated to Council; let City Council take responsibility for their actions. Some risks are not predictable, but given the lag in reporting issues like the diesel operator shortage (reported after Line 2 opened) or the mechanic shortage (revealed through the Auditor General), OC Transpo clearly unable or unwilling to discuss problems in a manner that allows them to be tackled proactively.
This may be a decision from management to save their skins, or it may be a Council decision to prevent budgetary increases (which also occurred in Toronto under Mayor John Tory and TTC CEO Rick Leary). In either case, the responsible party must be held accountable.
"Conditions to allow consistent travel time over the length of every bus route"
This is only true to the extent that better scheduling cannot. More consistent travel time reduces the need for long layovers, which reduces costs; but a variable travel time should not affect low-traffic routes, nor should it cause the erratic service seen on many OC Transpo routes.
The on-time performance for buses along every route is subject to variation caused by auto traffic congestion, the configuration of roads, construction and detours, transit ridership levels, weather, on-board incidents, and more.
Both the TMP and the annual capital budgets reflect the support that Council has given for transit priority measures such as bus stop relocation, queue-jump signals, and bus only lanes.
In 2013, Council committed to transit priority across the City with their then-new Transportation Master Plan. Very few of those projects ever materialized.
Staff are unable to say "this is your fault" without losing their jobs, but the hypocrisy displayed by councillors like Troster, Menard (Bank; who said that staff's slow speed on transit priority was "unacceptable"), and Brockington (Merivale, Carling) is shocking.
The largest and most significant initiative that Council has taken to reduce transit travel time and reduce its variability has been the rapid transit projects that have been built 17 consecutively since the early 1980s. The construction of the Transitway was an industry-leading innovation, removing buses on principal corridors from city streets subject to congestion and delay and putting them on an exclusive, grade-separated roadway ... The Transitway accommodated the growth of the city and the growth in transit travel to the point that it was approaching its physical limit to accept more customers, prompting the construction of O-Train Line 1.
This is not only some remarkable revisionism, but also, the Transitway does not reduce delays if Transitway routes are interlined, because unreliable surface operations bleed into Transitway routes.
When preparing trip schedules for bus routes, staff set the end-to-end travel time based on a database of actual travel times so that 85 per cent of the buses will arrive within the set time. Recovery time at the terminal is then allotted so that 95 per cent of the buses will start their next trip on schedule.
This measure keeps being failed, since reliability is at about 70% according to official statistics (and may be fudged, in my experience with unofficial results). Some routes see 0% on-time arrivals in certain time periods, and many busy routes see 20%, 30%, or more delayed trips from the origin.
Continuous improvement can include:
• Making bus routes shorter and more self-contained, so that the effect of an incident on one part of a route or on another linked route will cause fewer delays for buses and customers on the rest of the route.
The report is very close to understanding that interlining is what causes delays to spread. A shorter bus route does not help if the trip is delayed from a different route; "self-contained" in this context is meaningless without an end to interlining.
• Making bus routes straighter, with fewer diversions off the main travelled path, reducing the number of turns required and the exposure to sources of congestion.
This is correct, fewer turns improves both schedule consistency and speed. It also requires councillors with too much power understand that their pet diversions cost money to operate.
• Having route terminus points at Transitway and O-Train stations, where more spare buses and operators – whether scheduled or agreeing to work overtime – are available to cover trips that are delayed.
This only improves reliability if there are adequate extras. If OC Transpo does not schedule (or have the resources to schedule) extras, this is totally useless.
• Regularly collecting and analyzing data from thousands of bus trips to periodically adjust not only the scheduled end-to-end travel time but also the allocation of travel time between stops along the route, improving the ability for customers to know when the bus is expected and increasing their confidence that it will arrive at that time.
This requires no fancy software, expensive capital investments, or specific expertise that cannot be built in-house. If OC Transpo is not internally collecting detailed data to store away for scheduling purposes, then management should be fired and replaced with people who will meet the basic requirements of operating public transportation.
Variables
• Adverse weather creates a significant risk to on-time performance ... In 2025, two major snowfalls in February reduced punctuality to 72 per cent.
In January, punctuality was 75%. This metric also fell below 72% after May; this is overstating the effect of weather (in the long, cold winter, OC Transpo should be scheduling extra time anyways) and understating the effect of everything else.
• Increased congestion resulting from accidents, road closures on parallel corridors, construction detours, and events.
Increased congestion across the city adds travel time to every route; congestion on certain roads adds travel time to certain routes. With the exception of broad ranging closures (eg. the weekend closures of the Kichi Zībī Mīkan), this only propagates through the network thanks to interlining.
• Implementation of transit priority measures, while ideal for transit customers and transit operations, is always balanced against other road users for other means of travel.
Transit priority is certainly great to have, but transit service operates reliably in cities across the world, regardless of transit priority. Certainly, priority measures are not infinite. This is a tradeoff that Council has largely accepted, against transit. Council should end the farce and use straight words to describe their positions on transit priority.
• Currently, staff do not have policy direction from Council on whether to favour reliability over service levels, nor is there a mechanism to ensure that budgeted funding is increased to maintain or improve service reliability.
This is a damning condemnation of City Council's waffling. For years, Transit Commission/Committee have talked about improving bus service reliability, but councillors are not willing to step up and accept the tradeoffs that must be made to achieve such a thing.
There are mitigating factors and possible savings, but they cannot be investigated in any detail while Council blindfolds themselves to the tradeoffs.
The current state of OC Transpo service reflects the total failure of Councils since the Larry O'Brien era to understand that transit service is not free. They are consistent only in their lack of comprehension of the tradeoffs, and in their indecisiveness, have allowed an undisciplined service to deteriorate to its current state.
Council should ...
A mayor and council interested in understanding the tradeoffs to be had between reliable and more notional service would:
- Direct staff to study the results of an end to random interlining;
- order staff to collect and publish granular, detailed reliability data;
- direct that a reasonably reliable service be prioritized over increasing notional service, with a focus on reducing deadheads but axing the worst performing service if necessary, and;
- ask staff to publicly present estimates of increased resources for transit reliability, on an annual or bi-annual basis.
Until next time.
No comments:
Post a Comment