A stochastic approach to hotel revenue optimization

Owing to similar business nature, it should be possible to directly migrate successful airline revenue management techniques to the hotel domain. However, one of the salient differences between airlines and hotels is rarely highlighted – the network structure of length of stay or the displacement effect. The hotel patrons go from a first stay-over night to a last stay-over night in consecutive night stays. The arrival demands for multi-night stays and the lengths of stay are stochastic in nature. In this paper, we propose a network optimization model for hotel revenue management under an uncertain environment. The network optimization is in a stochastic programming formulation so as to capture the randomness of the unknown demand (unknown number of arrivals and length of stays). A novel approach of robust optimization techniques for stochastic programming is applied to solve the problem. We also discuss the strategies for hotel management to take into account of risk trade-off; different pricing policies; cancellations and no-show; early check-outs; extended stay and over-booking are discussed. We showed that our proposed model can be modified to adopt these strategic considerations.