Bus contract penalties in Ireland rest on three measures: how much scheduled service actually runs, how regular it is and how punctual it is. The National Transport Authority (NTA) applied deductions across its bus contracts in 2024, and its published report shows that the lost kilometre measure rests on the AVL system installed on each bus. For an operator, the practical lesson is to hold the same figures, calculated the same way, before the regulator does.
What did bus contract penalties in Ireland total for 2024?
On 20 April 2026, The Irish Times reported that more than €17 million in financial penalties had been imposed in 2024 on the operators of State-subsidised bus, rail and Luas services. The newspaper said the figures came from the NTA and the Department of Transport, which set them out in a letter to the Dáil Public Accounts Committee. For buses, it reported close to €5 million for Bus Éireann, €4.6 million for Dublin Bus and more than €1.3 million for Go-Ahead Ireland. Irish Rail (€3.7 million) and Luas (€2.1 million) are also named.
The NTA's own Public Service Obligation Contracts Annual Performance Report 2024 gives the bus detail in its Table 3 (page 29). The lines below are quoted from that table, in € million.
| Contract | 2024 deductions applied |
|---|---|
| Dublin Bus | 4.670 |
| Bus Éireann (Direct Award) | 4.986 |
| Bus Éireann Waterford City | 0.096 |
| Bus Éireann ECCC | 0.418 |
| Go-Ahead Ireland ODMA | 0.946 |
| Go-Ahead Ireland DCOM | 0.310 |
| Go-Ahead Ireland W4 W6 | 0.072 |
Adding the seven bus lines gives €11.498 million. That total is our own sum of the table, not a figure the NTA states. The Irish Times gave Dublin Bus as €4.6 million where the NTA table shows €4.670 million, so a careful reader should quote the NTA table when precision matters.
How does the NTA define reliability, regularity and punctuality?
The report uses three measures, and each one needs a different record from the operator.
Reliability is the lost kilometre rate. Lost kilometres are total scheduled kilometres minus total kilometres operated, and the rate is lost kilometres divided by scheduled kilometres, multiplied by 100 (page 18). The minimum performance standard for Dublin Bus and Bus Éireann is 2% per route per period from Period 1 of 2020. The report says operated kilometres are determined by the AVL system installed on each bus to record the route and distances travelled.
Regularity applies to high-frequency routes and is measured as excess wait time. It is the average actual waiting time minus the average planned waiting time (page 10).
Punctuality applies to low-frequency routes, defined as routes running fewer than five times per hour on weekdays outside peak periods. It is the share of actual departures that are on time, measured at every stop on each route. A departure counts as on time if it leaves no more than one minute early and no more than 5 minutes 59 seconds late.
That last definition is worth noting. The window is asymmetric: a bus that leaves two minutes early is a failure, while one that leaves five minutes late is still on time. Operators who think of punctuality only as lateness will misread their own performance under this contract.
Why do data caveats matter when penalties are applied?
The Irish Times quotes the Department of Transport as follows: "Where performance declines, penalties are applied rigorously unless the underlying cause is outside the operator's control." The NTA report makes a related statement on page 29, saying operators "were not unfairly penalised due to issues that were outside of their control".
The NTA report is open about the limits of its data. On complaints, the report states that "some complaints included in these figures may not be fully attributable to the operator", citing examples such as complaints about route frequency and about bus stops the operator does not maintain.
None of this means the regulator is careless. It means that a contract built on measured performance always has edge cases, and the party that holds a clean trip-level record is the party best placed to explain them. When a deduction is proposed, the useful question is not whether the headline rate is right, but whether the operator can reconstruct the individual trips behind it.
What does an AVM need to record to reproduce these measures?
Whatever the vendor, an operator that wants to reproduce a measurement regime like the NTA's needs five records for every scheduled trip.
- Scheduled and operated kilometres per trip, so that a lost kilometre rate can be rebuilt for any route and period.
- Actual departure time at each stop against the scheduled time, so that the on-time window (one minute early, 5 minutes 59 seconds late) can be applied exactly as the contract states it.
- Actual headways at stops on high-frequency routes, so that excess wait time can be recomputed from the same events the regulator uses.
- A reason for every cancelled or shortened trip, so that causes outside the operator's control, such as roadworks or an incident, can be separated from the rest.
- A timestamp and an audit trail for manual changes, so that a corrected record can be explained afterwards.
The fifth item is easy to forget. A record that is edited by hand without trace is less useful in a dispute than a slightly imperfect record with a clear history.
How can an operator prepare its own performance view?
Waiting for the regulator's annual report is too slow. The minimum standard is set per route per period, so an operator that compares its own AVM output with the contract definitions once a period, route by route, finds disagreements while the cause is still fresh in the depot.
A practical routine has three steps. First, compute the three measures from the AVM record using the contract's own definitions, not a generic punctuality figure. Second, list the routes closest to the minimum standard, for example those near the 2% lost kilometre limit, and review them trip by trip. Third, attach a documented cause to every trip that was missed or cut for a reason outside the operator's control, while the dispatcher who handled it still remembers.
The Pysae platform page lists, for Operations, a live map of delays, advances and regularity per line calculated automatically, plus alerts and incidents for service breaks and critical delays with immediate escalation. Pysae Fleet is listed as generating punctuality, mileage and journey time reports automatically from field data. Both can feed this kind of routine, but the contract's definitions decide how the numbers are read, so check that your reporting reproduces them before the first deduction notice rather than after. For the platform overview, see Pysae platform, and for the modules involved, Pysae Operations and Pysae Fleet.
What is the link between performance records and public real-time feeds?
They are related but not the same thing. The performance report ties operated kilometres to the AVL system installed on each bus, while passengers and app developers see the public real-time feed. For an operator, the point is consistency. If the two tell different stories about the same trip, someone will notice. Our guide to GTFS-RT in practice explains how feeds go wrong, and our piece on tracking subcontractor performance covers the same discipline when part of the network is run by someone else.
What should operators in other markets take from the Irish model?
The Irish regime is useful as a reference because the definitions are public, the formulas are written down and the amounts are reported by contract. Any authority or operator elsewhere can copy the structure: one reliability measure, one measure for regular services, one for timetabled services, each with a clear data source and a minimum standard.
The weak point in any such regime is the gap between the definition on paper and the record in the system. The NTA's own caveat about complaints shows where that gap can open. The operators best prepared for a measured contract are those that treat the AVM record as an asset to be maintained, with the same care as vehicles or driver rosters.
Frequently asked questions
What is the lost kilometre rate?
It is the share of scheduled kilometres that were not operated. The NTA calculates it as total scheduled kilometres minus total kilometres operated, divided by total scheduled kilometres, multiplied by 100. For Dublin Bus and Bus Éireann the minimum performance standard is 2% per route per period from Period 1 of 2020.
How does the NTA define an on-time departure on low-frequency routes?
A departure is on time if the bus leaves no more than one minute early and no more than 5 minutes 59 seconds late compared with the scheduled departure time. Low-frequency routes are those running fewer than five times per hour on weekdays outside peak periods.
What is excess wait time?
Excess wait time measures regularity on high-frequency routes. It is the average actual waiting time minus the average planned waiting time.
Are penalties applied when the cause is outside the operator's control?
According to the penalty policy quoted by The Irish Times, penalties are applied rigorously unless the underlying cause is outside the operator's control. The NTA report adds that it monitored operating conditions in 2024 so that operators were not unfairly penalised for issues outside their control.
Do the 2024 figures cover rail as well as buses?
The Irish Times figure of more than €17 million covers bus, rail and Luas operators. The NTA table used in this article covers bus contracts only.
Ready to see your own performance data?
If you want to see how your network's delays, advances and punctuality can be tracked in one platform, use the "Request a demo" button on this page and tell us which contract definitions you work to.