An Empirical Study Of Choice-Based Airline Revenue Management

Along with Chinese economy booming persistently,the quantity of airline passenger traffic is increasing year by year. The revenue management is an effective management technology which has been testified widely by airlines at home and abroad. Nowadays,the revenue management based on customer choice model has been the leading issue which shows the idea of passenger-oriented in the field of revenue management. This paper brings forward analytical groping study on the problem of airline passenger classes choice behavior,which is based on the revenue management theory and bands together the real marketing data and the theory of passenger choice behavior. The main content of this paper can be included as following aspects.Firstly, the emergence, development of revenue management was summarized systematically in this paper. Then the latest progress of domestic and foreign revenue management was summarized and the future research was pointed out. At the same time, the research progress of air passenger choice behavior was summarized and analyzed.Secondly, use the latest real airline tickets marketing data to describe and analyze, which shows as the price measure of dispersion is becoming wider, the significance of the scheduled flight vicarism and irreplaceable vicarism go stronger and customers'choices preference character comes lucider, under the background of Chinese ticket booking mechanism and international fierce competition.Then gets the main factors that have an influence on passenger choice from the view of consumer choice theory, and analyzes passenger preferences of all classes by the use of utility theory, and then expatiate on the research meanings in both theory and practice.The most important part of this paper is that we developed a statistical estimation algorithm based on maximum likelihood estimation that uses a variation of the Expectation-Maximization (EM) method to account for unobservable data. The procedure was applied to data for a set of test markets from Shenzhen to Beijing.We computed the parameters of every attributes value, and then get the probability of passenger seat choice and the revenue of the flight. Comparing with the amount of revenue based on original capacity control policies with EMSR assisted in. Overall our study suggests that choice-based revenue management is both feasible to execute and economically significant in China real-world airline environments. The models can offer reference and a new angle for further revenue management in airlines.