Robust revenue opportunity modeling with quadratic programming

Abstract

This practice-oriented paper describes a new, network-level revenue opportunity model based on a novel formulation, the sales-based quadratic program. This optimizes revenue while providing market-level allocations that are more stable and robust over time than traditional solutions based on linear programming. Because airline origin–destination networks foster passenger connections, they comprise many more markets served than flights operated; such structures provide additional degrees of freedom for revenue management (RM) controls and often lead to alternate optimal (or near optimal) solutions. These alternate control solutions cause manageability issues for airline RM analysts in practice. Our proposed approach provides better stability in revenue opportunity model (ROM) controls over time, aiding RM analysts in setting effective default allocations and monitoring outliers. It is also a consideration when holding RM analysts accountable to market-level ROM performance metrics. Methods for reducing ROM control variation have not been addressed in prior literature.