MIT/PODS Revenue Management Consortium

TThe PODS Consortium at the Massachusetts Institute of Technology (MIT) is funded by corporate members, who also provide funding for PODS programming and development by PODS Research LLC (PRL). In 2016, Consortium membership includes Air Canada, Amadeus, American Airlines, ATPCO, the Boeing Company, Delta, Emirates, Etihad, LATAM, Lufthansa/Swiss, PROS, Qatar Airways, SABRE and United. With guidance from the members, PODS is employed by MIT graduate students to test revenue impacts of existing and new models for demand forecasting and seat availability optimization in airline revenue management systems. New topics for PODS research emerge from regular meetings of the Consortium.

PODS research at MIT investigates the interactions between different RM optimization methods, forecasting, and estimation models. Moreover, PODS research has explored “willingness to pay” estimation and the revenue impacts of using sell-up expectations in combination with various forecasting and optimization tools. Enhanced capabilities of PODS now permit simulation of larger networks with different airline route and passenger flow characteristics.

The focus of current PODS Consortium research includes the modification of existing RM methods to accommodate changing airline fare structures, low-fare competition, ancillary revenues, and code-share alliances. New and/or revised RM methods including MNL passenger choice models, cancellation forecasting and joint cabin optimization are also being explored.

RESEARCH PROJECTS AND THESIS TOPICS AT MIT USING THE PODS SIMULATOR HAVE INCLUDED:

• Optimization of alliance network revenues through bid price exchange and modifications to partner seat inventory controls
• Demand forecasting and optimization for new “fare family” structures being introduced by airlines
• Estimation of passenger choice and willingness to pay for RM forecasting using statistical methods applied to RM booking history
• Joint optimization of multiple aircraft cabins to account for passenger choice and use of shared seat inventories
• Incorporating ancillary revenue potential into RM optimization models and distribution of seat availability
• Modeling the impacts of dynamic user influence on RM system performance
• Development of cancellation forecasting models and simulation of their benefits in different RM systems
• Simulation of potential effects under New Distribution Capability (NDC), including development of dynamic pricing and classless RM algorithms.