CanadaTravel and aviation2024 – ongoingAir Canada
In-house On-Time Performance Schedule Optimizer for network planning
Air Canada built a tool that predicts the on-time performance of a proposed flight schedule before it is flown
Problem
Air Canada's Network Planning team decides where the airline flies and builds its schedule. Weak points in a schedule were hard to see until operations showed them.
Approach
A team from IT, Commercial and Operations built the On-Time Performance Schedule Optimizer in about two years, starting from scratch. It combines simulations with machine learning to predict on-time performance of a proposed schedule at high and low levels. It flags stress points so planners can change the schedule months ahead, change an aircraft rotation or change how operations run. Network Planning started using it in October 2024. Air Canada has built AI tools in-house since November 2018 and has more than 75 data staff.
Results
No published figures
Takeaway
Air Canada put schedule-quality prediction in front of planners before flights are sold and flown. The company gave no performance results at launch.
Categories
- Size
- Large enterprise
- Industry
- Travel and aviation
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLOptimization
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Air Canada's media site describes its in-house AI work and the On-Time Performance Schedule Optimizer launched the month before. It explains how the tool is used by Network Planning and who built it. It gives no results for the tool.
Full details are in the original sources.
