Font Size: a A A

New approaches to yield management: Comprehensive overbooking/allocation heuristics for the hotel industry

Posted on:1995-10-07Degree:Ph.DType:Dissertation
University:The Ohio State UniversityCandidate:Baker, Timothy KevinFull Text:PDF
GTID:1479390014491407Subject:Business Administration
Abstract/Summary:
Yield management is the dynamic pricing, overbooking, and allocation of perishable assets across market segments to maximize short-term revenues for the firm. Numerous overbooking and allocation algorithms exist for the airline industry. In this industry, the perishable asset is the airline seat; when an airline departs, no revenue can be gained from the sale of the empty seats. The literature on overbooking/allocation algorithms for the hotel industry is comparatively sparse. In this industry, the perishable asset is the hotel room; once a night has passed, no revenue can be gained from the renting of unoccupied rooms. No research has compared overbooking/allocation algorithms in a realistic hotel industry environment, and nobody has produced hotel overbooking/allocation algorithms that (1) integrate overbooking with allocation; (2) model the phenomenon of hotel patrons extending or contracting their stay on a moment's notice--the "stayover extension/contraction" phenomenon.;This research develops two new hotel overbooking/allocation algorithms that both integrate overbooking with the allocation decisions. The first new heuristic is the Hotel Nested Network heuristic. The second new heuristic, the Hotel Bidprice method, extends the Hotel Nested Network method by modeling stayover extensions/contractions. Both of these heuristics attempt to maximize revenues over a string of consecutive nights. These two heuristics are tested against (1) the best performing airline overbooking/allocation heuristic as determined by a previous simulation study--the Nested By Deterministic Shadow Prices (NDSP) method; (2) the Threshold Curve methods, which are commonly used in the hotel industry. The testing is performed in a realistic simulation environment.;The simulation has been conceptually validated by a major hotel chain, and the data has been obtained largely through interviews with personnel at another hotel.;The Hotel Bidprice method generates greater revenues than all others by typically at least 3% under these two operating environments: (1) high noshow levels and medium-to-high (120% to 150% capacity utilization) demand levels; (2) large gaps in room revenues between market segments and high demand levels. The Threshold methods perform as well as any other method in resort operating environments and in low (90% capacity utilization) demand conditions.
Keywords/Search Tags:Hotel, Allocation, Overbooking, Heuristic, New, Method, Revenues
Related items