July 10th, 2007
BENEFITS OF OPTIMIZING PRICES TO MANAGE DEMAND IN HOTEL REVENUE MANAGEMENT SYSTEMS*, THE
We investigate the revenue impact of a new Price Setting Method (PSM) and compare it with the industry standard Bid Price Method (BPM). This comparison is performed via a simulation that was validated by a major hotel chain. In 27 out of the 32 cases, the PSM outperformed the BPM based on statistically significant tests. The PSM produces an average revenue increase of 34%, which can be thought of as an upper bound on the realistic revenue increase.
A Revenue Management System (RMS) consists of dynamic methods to forecast demand, allocate perishable assets across rate classes, and decide when to overbook and by how much, and what price to charge different rate classes. A perishable asset might be a hotel room, an airline seat, or broadcast advertising space. The asset (service capacity) is perishable because its revenue-generating capabilities immediately drop to zero at a point in time-right after the sales period ends. For example, once a night has expired, the lost revenue from an empty hotel room is lost forever.
Although there have been nonpublicized failures in applying RMS (see Cross 1997), most of the applications of RMS increase revenues while maintaining or improving service to customers. For example, Marriott’s RMS improved its revenues by $25-$35 million in 1991 (Lieberman 1992). Royal Caribbean Cruise Lines obtained a revenue increase in excess of $20 million for 1 year (Lieberman 1992). American Airlines increased its revenue by an estimated $1.4 billion over the 1989-91 time frame (Smith, Leimkuhler, and Darrow 1992). Revenue management saved National Car Rental from liquidation in 1994. During the first year of RMS implementation, revenues at National Car Rental increased by $56 million (Geraghty and Johnson 1997). The application of revenue management concepts and techniques has been proposed in other industries as well, such as golf courses (Kimes 2000) and internet service providers (Nair and Bapna 2001).
The focus of RMS is primarily on revenue maximization due to the cost structure and economics of perishable assets. For example, the fixed cost of running a hotel is high, and once the hotel occupancy rate is above the break-even occupancy rate, the contribution to profit and overhead per incremental customer is high.
RMS is viewed from strategic and tactical perspectives. The strategic view involves designing rate classes that cause customers to choose one service provider over the other. The rate class is a clearly defined set of tangible (goods-content) and intangible (service-content) features that benefit the customer and cause the customer to recognize, pay for, use, and experience the service (Collier 1994).
RMS uses “fences” to encourage business customers willing to pay a higher price from taking a lower price because of advanced purchase restrictions. Advance purchase restrictions may include items such as (1) minimum time in advance of arrival that booking must be made, and (2) Saturday night stayover requirements. Refund policies for reserved rooms, kids-stay-free policies, senior discounts, room upgrading policies, and penalties for checking out early or checking out late are other examples of RMS rate class characteristics that are related to the “service-content” or “intangible-content” of the hotel business. In summary, a rate class is a unique combination of room rate, advance purchase restrictions, physical hotel amenities, and other hotel policies.
Tactical revenue management makes rooms available for sale at certain points in time, at specific prices, and for various lengths-of-stay to maximize revenues over a typical 1-year planning horizon. It revolves around opening and closing hotel room availability sometimes on an hourly basis. For example, during peak demand periods, the higher priced rate classes are left open and represent a higher percentage of total room capacity, while the lowest priced rate classes are closed (i.e., unavailable to customers). Weatherford and Bodily (1992) define a taxonomy of tactical revenue management problems and solutions. Definitions of hotel yield management terms are presented in Appendix A. See Hanks, Cross, and Noland (1992) for a detailed discussion of strategic revenue management.
We compare the Price Setting Method (PSM) with an industry current and best practice Bid Price Method (BPM) approach (Talluri and Van Ryzin 1998). We define both methods in detail later in this article. We use the BPM as the competing algorithm to PSM because BPM is the most used hotel allocation approach developed to date. For example, Sabre Decision Technologies and Manugistics are the dominant consultancies for installing hotel revenue management systems, and both use BPM (Furniss 2002; Stewart 1999).
The article first gives a synopsis of revenue management literature. Then, a detailed description of the PSM and BPM methods is provided. We show here that the PSM is a concatenation of Gallego and Van Ryzin’s (1997) undifferentiated network approach and the popular bid price method. The simulation model for method comparison is then described, with details left in the Appendices. We then present the research design, describing the broad range of cases used to evaluate the PSM and BPM. The Results section provides insights into why the PSM dominates the BPM. Finally, we discuss avenues of further study in the analysis and development of explicit price setting methods.