Modelling And Optimization Of The Single-Leg Revenue Management With Customer Behavior

Revenue management has its roots in the deregulation of the US airline industry in the late 1970s.The fierce competition in the civil aviation industry drove revenue management to develop rapidly and influenced other industries,and its theoretical model research has also been constantly enriched and improved.Revenue management is the process of determining the optimal capacity allocation or price strategy according to the market demand,and reasonably matching between the limited resource and the market segmentation demands,so as to maximize revenue.With fierce competition,economic slowdown and global epidemic,airlines are facing severe challenges.In addition to the traditional revenue management leverage,aviation revenue management based on customer behavior is the key to the survival and development of airlines.The application of revenue management in China’s civil aviation industry lagged behind that in North America for 20 years.Most airlines have introduced mature revenue management systems from Europe and America,but the application is not good.Based on the demand for revenue management optimization model driven by a domestic airline with independent intellectual property rights revenue management system,this paper studies the model and optimization of single-leg revenue management based on customer behavior.Firstly,the demand forecasting is made based on the historical data of a certain flight of the airline company,then using simulation to evaluate the accuracy of different forecasting methods,and the system benefits brought by the artificial control policy of the airline operator and the control policy of the traditional revenue management optimization model are also compared through simulation.Secondly,customer behavior including cancellation and no-show are further considered into the optimization model.This thesis explores how the airline hedge against the loss caused by cancellation and no-show through two mechanism,i.e.,set cancellation refund and delay purchase,then discusses the pros and cons of these two mechanisms.Finally,the optimization model with customer buy-up and choice behavior is studied.In view of the less-rational independent demand assumption of the traditional revenue management model and the customer behavior shown in the real data,customers are divided into two groups,namely product-oriented customer and price-oriented customer.We built a discrete-choice model with these two types of customers and further illustrate it through numerical study.Meanwhile,under this newly built optimization framework,from demand forecasting,parameter estimation,accuracy evaluation to comparation of airlines manager’s policy and optimal policy,we compare the result to that of traditional RM model and obtain some managerial insights.This thesis is driven by practical problems and supported by real world data.Through modeling optimization,theoretical analysis,numerical study and computer simulation,it studies the model and optimization of airline single-leg revenue management based on customer behavior and explains its important management insights,which has certain theoretical significance and important practical value.