Correspondence: Guillermo Gallego, Department of Industrial Engineering and Operations Research, Columbia University, New York, New York 10027, USA. E-mail: gmg2@columbia.edu

¹joined Columbia University in 1988 after graduating from Cornell University. Professor Gallego has published extensively in the areas of production management, supply chain management and revenue management. He has been the recipient of numerous grants from industry and from the National Science Foundation. In addition, he has held editorial positions in the flagship journals of his field.

²is a graduate student in the IEOR Department at Columbia University and recently joined IBM's Watson Research Center. The material presented in this paper is part of her PhD research work.

³is a senior research scientist in the Sabre Holdings research group. He leads advanced R&D efforts in airline pricing and revenue management including collaboration with universities.

Received 2 September 2008; Revised 2 September 2008.

Abstract

Revenue management (RM) based on customer choice models has gained interest in the 21st century. Changes in airline distribution strategies and improvements in online search engines have resulted in fierce competition, fare transparency and greater usage of restriction-free pricing. Existing solutions to RM with restriction-free pricing have proven ineffective and use either ad hoc adjustments to traditional capacity allocation or computationally intensive dynamic programming. We present new generalised EMSR formulations for the single-leg, nested, multiple fare RM problem that work in both restriction-free and traditional airfare conditions. In our framework, demand for different fare classes is estimated by a customer choice model. We show how a multinomial logit demand model can provide upsell estimates for handling dependent demands and also account for competitive effects. We develop efficient and nearly optimal static heuristics for RM optimisation that are more general and provide better performance than the widely used EMSR-b algorithm for independent demands. Variations of the algorithm for both low-to-high and mixed arrival order cases are provided.