Flight delays and cancellations are not just airline statistics. They affect people travelling for work, family, holidays and connecting flights. This data story uses Australian aviation on-time performance data to show how reliability varies across time, airlines, routes and airports.
This story fits The Conversation’s Blindsided topic because flight reliability is an everyday issue that can quietly affect thousands of passengers if it is ignored.
The original aviation dataset was cleaned before analysis. Column names were standardised, repeated header rows were removed, numeric fields were converted from text to numbers, and duplicate records were removed. Three new variables were created: departure delay rate, arrival delay rate and cancellation rate. These fields helped compare reliability across airlines, routes, months and airports.
The analysis focuses on flights from January 2023 onwards to keep the story timely and relevant.
| Total routes | Total airlines | Total departing airports | Total flights flown | Average on-time departure (%) | Average on-time arrival (%) | Average cancellation rate (%) |
|---|---|---|---|---|---|---|
| 129 | 11 | 39 | 3645456 | 73.4 | 73.5 | 2.2 |
The first chart shows that flight reliability is not fixed. On-time arrival and departure performance change across months, while cancellation rates add another layer of disruption. This matters because passengers cannot assume reliability is stable across the year; the month of travel can affect the chance of delay or cancellation.
This chart compares airlines using multiple reliability measures. Airlines with stronger arrival performance and lower cancellation rates create a more reliable experience for passengers. This means airline choice can shape the passenger experience, especially when on-time arrival performance and cancellation risk move in different directions.
Route-level analysis shows where passengers may be more exposed to disruption. To avoid very small routes distorting the story, only routes with at least 500 flown sectors were included. This shows that disruption is not evenly spread across the network; some routes expose passengers to a higher risk of arriving late.
Delays are disruptive, but cancellations can be even more serious because passengers may need to rebook or change plans completely. This chart shows that cancellation risk is not evenly distributed across airlines. This is important because cancellations create a deeper form of disruption than delays, often forcing passengers to rebook or change their plans entirely.
Airline and route choices matter, but the airport where a journey begins can also shape reliability. This chart compares departure reliability, cancellation risk and flight volume across departing airports. This suggests that reliability is also connected to where a journey begins, not only the airline or route selected.
Australian flight reliability is uneven. The data suggests passengers face different levels of disruption depending on the month, airline, route and departing airport. This makes flight reliability a public-interest issue, not just an operational airline metric.
Australian Government Department of Infrastructure, Transport, Regional Development, Communications and the Arts. (2026). Domestic airline on time performance. https://www.bitre.gov.au/statistics/aviation/otp_month The Conversation. (n.d.). The Conversation Australia. https://theconversation.com/au OpenAI. (2026). ChatGPT [Large language model]. https://chat.openai.com/
I used ChatGPT to support planning, R code troubleshooting, wording refinement and structure development. I reviewed, edited and made final decisions on the dataset selection, visualisation design, analysis interpretation and submitted content.