The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here: https://www.microsoft.com/en-us/microsoft-365/windows/end-of-ie-support).

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

No-show Forecast Using Passenger Booking Data

På svenska

Author

  • David Zenkert

Summary, in English

Amadeus IT Group provide revenue management systems for the airline industry. The concept of overbooking has been known and applied within the industry since the middle of the 20th century, thus playing a large role in a revenue management prospect. The passengers booking data is something that could improve the forecasting of the rate at which passengers don’t show up or cancel their respective flights, henceforth referred to as cancellation/no-show rate. This thesis will only address the no-show part but both the concept of cancellations and no-show together are important when overbooking flights optimally. Overbooking too little will result in lost revenues and overbooking too much will result in fees for compensating possibly upset passengers and of course the issue of having to deny boarding to them as well. Therefore, the investigation around how to optimally overbook flights is of importance for Amadeus.
In this thesis, machine learning algorithms are tested with the objective to improve the no-show rates. The revenue management part of this project will not be discussed in great detail

Publishing year

2017

Language

English

Document type

Student publication for Master's degree (two years)

Topic

  • Mathematics and Statistics

Supervisor

  • Umberto Picchini