Model of Dynamic Pricing for Two Parallels Flights with Multiple Fare Classes Based on Passenger Choice Behavior

Airline revenue management (ARM) is one of emerging topics in transportation logistics areas. This paper discusses a problem in ARM which is dynamic pricing for two parallel flights owned by the same airline. We extended the existing model on Joint Pricing Model for Parallel Flights under passenger choice behavior in the literature. We generalized the model to consider multiple full-fare class instead of only single full-fare class. Consequently, we have to define the seat allocation for each fare class beforehand. We have combined the joint pricing model and the model of nested Expected Marginal Seat Revenue (EMSR) model. To solve this hybrid model, we have developed a dynamic programming-based algorithm. We also have conducted numerical experiments to show the behavior of our model. Our experiment results have showed that the expected revenue of both flights significantly induced by the proportion of the time flexible passengers and the number of allocated seat in each full-fare class. As managerial insights, our model has proved that there is a closed relationship between demand management, which is represented by the price of each fare class, and total expected revenue considering the passenger choice behavior.