Methodology

What these numbers are, and what they are not.

Source

The statistics here come from the Bureau of Infrastructure and Transport Research Economics (BITRE), the Australian government's transport statistics agency. BITRE publishes monthly spreadsheets; this site stitches them into one consistent series and presents them by route.

Domestic city-pair passenger history begins in January 1984, international city-pair passenger history in January 1985, and punctuality history in November 2003. Most historical traffic and reliability analysis on this site comes from BITRE's published series; it is presented without forecasts.

Route pages also use a commercial schedule feed for enrichment that BITRE does not publish: flight numbers, observed aircraft types, scheduled operating days, and scheduled flight times or durations. Those fields describe the timetable snapshot dated beside them; they do not carry a punctuality or cancellation figure for any individual flight.

How the rates are calculated

Cancellation rate is cancelled flights divided by scheduled flights. On-time rate is on-time movements divided by flights actually flown, which is how BITRE defines it — a cancelled flight is not counted as late.

Twelve-month figures are weighted by flight volume, not an average of twelve monthly percentages. A month with 3,000 flights counts for more than a month with 300.

The national comparison (2.1% cancelled over the 12 months to Jul 2026) is BITRE's own all-airlines, all-ports total — not a sum of the routes shown on this site. Those routes account for roughly three quarters of monitored flights, so adding them up would overstate the national rate and make every individual route look better than it is.

Airlines are shown as BITRE reports them

Qantas and QantasLink are separate operators in BITRE's data and are shown separately here. They are usually booked as one airline, but on routes flown by both, the cancellation rates often differ by several times over — merging them would hide the most useful thing on the page. The same applies to Virgin Australia and Virgin Australia Regional Airlines.

How the lost-route record is defined

The lost-route record uses the comparable international city-pair passenger series from January 2003 onward. A retained city pair must have at least one positive passenger month in that series, and its last positive passenger month must fall strictly more than 12 months before the latest published positive passenger period. Freight or mail records do not advance that date.

A BITRE city pair is not necessarily a nonstop sector. Some services kept one flight number while making an intermediate stop, so this record does not describe every pair as a former nonstop route.

Peak annual passengers are observed calendar-year sums from 2003 onward. Partial years are not annualised, missing months are not interpolated, and the count of active months means positive recorded months rather than an uninterrupted operating history.

What this data cannot tell you

• Airfares by route. BITRE's fare index is national only, and samples the cheapest available fare on one day a month — not what passengers actually paid.
• Why a flight was cancelled. Weather, crew, engineering and commercial decisions are not distinguished.
• Anything about a specific future flight. These are monthly historical aggregates, published months in arrears.
• Delay length. BITRE records whether a movement was on time, not by how much it missed.

Corrections

BITRE revises its own figures; when it does, this site changes with it. If a number here looks wrong, it is worth checking the source — and telling us.

Data reused under CC BY 4.0. This site is not affiliated with BITRE, the Australian Government, or any airline.