To decarbonise aviation, airports need to plan early which energy carriers they will have to supply in the future and in what quantities. Whether kerosene, sustainable aviation fuels, hydrogen or electricity: the basis for this is a reliable forecast of traffic volumes at each individual airport. Existing forecasts are mostly based on classical statistical methods and often only consider the national level. Artificial intelligence offers new opportunities to capture airport-specific developments and their driving factors.
The aim of this thesis is to develop AI-based forecasting models for demand development at major German hub airports. Possible directions are:
- Comparison of classical statistical methods with machine learning methods regarding their forecasting accuracy
- Use of pre-trained AI models for time series in the presence of short data series and structural breaks such as the COVID-19 pandemic
- ...
The specific direction will be agreed upon before the start of the thesis based on a short exposé.
Note: Programming experience in Python is recommended.
If you are interested, please contact Alexander Barke.