Demand forecasting in hospitality using smoothed demand curves

Demand forecasting is one of the fundamental components a successful revenue management system. This paper provides new model, which inspired by cubic smoothing splines, resulting in smooth demand curves per rate class over time until check-in date.This model makes trade-off between error and smoothness fit, therefore able to capture natural guest behavior. The tested on hospitality data. We also implemented an optimization module, computed expected improvement using our forecast optimal pricing policy. Using data four properties from major hotel chain, 2.9 10.2% more obtained than heuristic done hotels.