Improved revenues from various unconstraining methods in a passenger origin‐destination simulator (PODS) environment with semi‐restricted fares

In this article, we use the sophisticated passenger origin‐destination simulator (PODS) simulator to examine the revenue impact of four different methods of unconstraining – Expectation Maximization, Projection Detruncation, Booking Curve (BC) and Pickup. Because of the competitive nature of PODS (two airlines competing head‐to‐head for customers) and its allowance of customer choice, for the first time we are able to assess fully all the implications of switching to more sophisticated unconstraining methods, including the impact of spill, upgrades and recapture. We find that the optimization engine that is currently being used dictates quite different revenue results when changing unconstraining methods. In one case (under Displacement Adjusted Virtual Nesting optimization), we should switch away from BC unconstraining, and in the other case (under leg optimization), we should not. Finally, using realistic booking data from major global airlines to calibrate PODS, we show that upgrading the unconstraining process can lead to revenue gains of 2–15 per cent.