Research on multi-leg inventory control based on passenger choice

Abstract

To remedy the lack of mathematical programming and the Expected Marginal Seat Revenue (EMSR) model for multi-leg seat inventory control, this paper proposes a method based on passenger choice. Except for data about which seats passengers decide to opt for, there is no need to obtain distributions of passengers' demands or other "a priori" information. The proposed method can discover the real factors that affect passengers' choices, and then estimate the probabilities of seat choices and the revenue according to the weights of the factors. Simulated experiments and comparison with the shadow price method and the virtual "bucket" method confirm the feasibility and effectiveness of the proposed method in seat inventory control for multi-leg flights.