Forecasting seat sales in passenger airlines: introducing the round-trip model

The innovative idea of this research is changing the focus from valuation based customer segmentation into segmenting customers in base of their travel plans, and using price information of flights that are likely to be considered by passengers as pairs, in forecasting models. This improves forecasting models precision, supports better management of the demand among a set of potential related flights, and results in revenue increase. Finite regression mixture models were tested to identify linked traffic in the two directions and potentially connected flights on a route in reverse directions; Adaptive Neuro-Fuzzy Inference System (ANFIS) is applied to develop comparative models for studying sales effect between one-way versus round-trip models. The RTM demonstrated 5% error reduction, comparing with one-way model by entering only one linked flight. This empirical study is performed on a database with 22,900 flights which was obtained from a major North American passenger airline.