New ZealandTravel and aviation2023 – ongoingAuckland Airport
Machine-learning prediction of aircraft pushback readiness
Auckland Airport reported 100% data-entry compliance while ML pushback predictions matched manual accuracy
Problem
Staff manually entered expected aircraft pushback times at gates. Entries were incomplete, and their predictions were often inaccurate, complicating departure sequencing.
Approach
Auckland Airport combined CCTV observations of aircraft servicing with flight data. Machine learning predicted when aircraft would be ready to leave their stands. Its FY2023 disclosure reported complete automated data entry, with prediction accuracy still equivalent to the manual process.
Results
Takeaway
Complete data capture did not yet produce more accurate departure predictions. The airport reported that accuracy was improving as it added data.
Categories
- Industry
- Travel and aviation
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLComputer vision
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
The airport used machine learning to predict aircraft pushback times. It reported 100% data-entry compliance while prediction accuracy remained equivalent to manual estimates.
Full details are in the original sources.
