Correspondence: Darius Walczak, PROS, 3100 Main Street, Suite 900, Houston TX 77002, USA. E-mail: dwalczak@prosrm.com

¹is a director of Optimization at PROS. Walczak holds a PhD degree in Business Administration from Sauder Business School, University of British Columbia (UBC) in Vancouver, BC along with Master's degrees in Mathematics from UBC and Applied Mathematics from Wroclaw University of Technology. Since joining PROS in 2001, Walczak has been involved in the design and deployment of many revenue management and pricing optimization applications across a variety of industries.

²is a scientist at PROS. Mardan joined the PROS Science and Research team in 2007. She earned her PhD in Industrial Engineering from the University of Minnesota at Minneapolis. She also holds a BS and an MS in Industrial Engineering. Before her PhD study, Mardan worked for Ghods_Niroo Consulting Engineers as a project control expert in Tehran.

³is a director of Science and Research at PROS. Kallesen joined PROS in 1999. He manages a team of scientists developing revenue management and pricing products. Before joining PROS, Kallesen was a member of the technical staff of Logicon, Inc., which is now part of Northrop Grumman. Kallesen earned his Master's of Science in Mathematics with an emphasis in Operations Research from the University of Nebraska at Omaha.

Received 3 September 2009; Revised 3 September 2009; Published online 30 October 2009.

Abstract

We describe marginal expected revenue data transformation, a mathematical technique that facilitates optimization with customer choice models. The main focus of this article is the analytical relationship between the transformation and Expected Marginal Seat Revenue (EMSR) as applied to customer demand with choice. We present an EMSR-based approach to single-leg revenue management with price-sensitive customers. This approach can be equivalently computed using the EMSR-b heuristic but with transformed problem data. We evaluate this approach by providing numerical comparisons to other control methods.