The full text of this article

 

Modelling choice of flight and booking class - a study using Stated Preference and Revealed Preference data
by Staffan Algers, Muriel Beser
International Journal of Services Technology and Management (IJSTM), Vol. 2, No. 1/2, 2001

 

Abstract: In 1994, Scandinavian Airlines System (SAS) took the initiative in a collaboration with the Royal Institute of Technology (KTH) concerning a project on the estimation of Deviation, Recapture and Buy up in the context of yield management. The KTH part of the project was to design and analyse Stated Preference (SP) or conjoint experiments and to support the field interviews for this. Passenger choice of flight and booking class also was modelled using loggings of reservations (and other data) supplied by SAS. This technique is called Revealed Preferences (RP) since the passenger ''reveals'' his preferences when actually choosing a product in the market. The logging is a record of an actual outcome in a choice process; each flown coupon represents a choice of flight and booking class. Therefore, methods of analysing discrete choice should be applicable. In the booking situation, the focus is on the availability of alternatives, which varies over individuals and thus makes it desirable to base the mathematical analysis on disaggregate data. The logit model is widely used for analysing discrete choice; it was applied in the project. Booking process data define choices, alternative class availability, and the characteristics of the alternatives in terms of departure times, prices, booking restrictions, service levels etc. The tariff structure is such that these factors are highly correlated. It will therefore be difficult to obtain reliable parameter estimates. An advantage of the data from the booking process is that they do reflect market preferences. SP interviews were conducted in order to permit analyses of the impact of individual factors. Air passengers were asked to state choices between hypothetical alternatives that were presented in pairs. This permits analysis of the data by means of a logit choice model, in which each choice between two alternatives is an individual observation. RP and SP data each have their strengths and weaknesses; but there is an analysis technique that takes advantage of the strengths in both data types. The utility function derived from the SP experiments is then ''scaled'' using the RP data. The paper describes the SP experiments that were carried out, and the results in terms of monetary values. Factors that were analysed in the SP experiments include price, service level, departure time and booking restrictions. The paper further describes the RP data from the booking process, and the resulting models. Separate models have been estimated for different trip legs and for international and domestic flights respectively.

Online publication date: Fri, 04-Jul-2003

 

is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

 
Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

 
Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Services Technology and Management (IJSTM):
Login with your Inderscience username and password:

 

    Username:        Password:         

Forgotten your password?


 
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

 
If you still need assistance, please email subs@inderscience.com