Three-stage decision approach of network dynamic pricing and seat inventory control

The dynamic joint pricing and seat inventory control is more practical but complicated in both formulation and solving. This paper presents a three-stage decision approach (TSDA) to attack this problem. In the first stage, the relationship between dynamic prices and their pre-sale periods is built. The game process between passengers and the airline based on maximisation of both passenger utility and airline's network revenue is applied. Passenger's booking and cancellation processes are simulated according to respective distributions. In the second stage, the seat allocation model for different itineraries is built based on presented unified prices. It greatly decreases computation complexity since prices, itinerary legs and pre-sale time periods are excluded from combination. In the third stage, itinerary-based revenue management model is built with embedding the nested control strategy inside. Meanwhile, some practical factors such as cancellation and no-show are considered. The experimental computation results show TSDA is effective.