Forecasting uncertain hotel room demand

Mihir Rajopadhye, Mounir Ben Ghaliay and Paul P. Wang Department of Electrical and Computer Engineering Duke University Durham, North Carolina 27708-0291 Timothy Baker and Craig V. Eister Bass Hotels and Resorts Atlanta, Georgia 30346-2149 Abstract Economic systems are characterized by increasing uncertainty in their dynamics. This increasing uncertainty is likely to incur bad decisions that can be costly in nancial terms. This makes forecasting of uncertain economic variables an instrumental activity in any organization. This paper takes the hotel industry as a practical application of forecasting using the Holt-Winters method. The problem here is to forecast the uncertain demand for rooms at a hotel for each arrival day. Forecasting is part of hotel revenue management system whose objective is to maximize the revenue by making decisions regarding when to make rooms available for customers and at what price. The forecast approach discussed in this paper is based on quantitative models and does not incorporate management expertise. Even though, forecast results are found to be satisfactory for certain days, this is not the case for other arrival days. It is believed that human judgment is important when dealing with external events that may a ect the variables being forecasted. Actual data from a hotel are used to illustrate the forecasting mechanism.