| This dissertation studies the current practice of yield management in the airline industry, identifies areas with potential for improvement, and recommends alternative strategies that would make a positive impact on an airline's earnings.; Airlines operating a hub and spoke flight system realize that it would be beneficial to them if they had a seat inventory control system that could efficiently handle the yield management problem for a bank. Bank Seat Inventory Control would be a significant step towards market class control for an airline. In this direction, this dissertation focuses on developing dynamic bid pricing models for seat inventory control at the bank level. This in itself is a major task, since a typical bank could have about 3200 different itineraries with some significant demand flowing through it. Effectively managing the seat inventory between these 3200 itineraries could possibly mean a much better yield management system than a simple virtual nesting of fare classes technique. We show that the proposed Probabilistic Network Bid Price Algorithm when compared against the existing Virtual Nesting System, using the Network Simulation Model that was developed as part of the research effort, on a Bank, can result in revenue improvements of as much as 0.6%.; The other drawback identified in the current system was in the use of market share of the airline as the sole indicator of the percentage of denied bookings in a market that could be recaptured by an airline. It makes intuitive sense that the percentage of denied bookings that are actually recaptured will be inversely proportional to the load factors on the flights that an airline operates in a market. Using the Network Simulation Model developed as part of this research effort, we propose an adjustment to the current model, that incorporates the effect of the load factors on flights in the market and show that the adjustment does in fact impact the initial decision making (accept or reject a booking) that an airline's yield management system executes, and will result in revenue improvements of the order of 0.2%. |
