Abstract
For the airline business, airliners are allowed to sell the seats of a same cabin for different prices. Normally, they allocate the seats of a cabin class into a number of the discrete fare classes based on the demand forecast of customers with different acceptable prices. A customer with demand in a higher fare class is willing to pay more. Each time when a customer with a fare class arrives, an airliner must decide whether to fulfill or reject the request immediately. In order to increase the revenue, the airliner may want to reject certain lower fare class customer requests and reserve the seats for future higher fare class customers. The problem this study focuses on is rationing decision for a single-leg flight with multiple fare classes and the customer arrival process of an individual class is non-homogeneous Poisson. The objective is to maximize the revenue of a flight. This study develop two real-time decision procedures, dynamic seat rationing (DSR) and expected revenue gap (ERG) to help the airliner make a decision when a customer arrives. The simulation results show that ERG approach perform best among all test approaches and is very robust under various problem conditions. Also, it takes a very short computation time to execute on computer.