HotelSimu: a Parametric Simulator for Hotel Dynamic Pricing Andrea Mariello ∗ , Manuel Dalcastagnè, Mauro Brunato, Roberto Battiti Department of Information Engineering and Computer Science, University of Trento, Via Sommarive 9 I-38123, Povo (TN), Italy Abstract In Hotel Revenue Management, an optimal pricing policy is crucial for max- imizing the proft. However, the complexity of the involved processes makes the defnition ofefective and efcient models a challenging task. We propose an efcient hotel simulator, HotelSimu, which generates events by using Monte Carlo simulations. Diferently from previous works, cancellations and reserva- tions are interleaved, and they are generated according to parametric models that cover several scenarios. Seasonal averages can be set on a daily basis. The hotel registry changes after each single event, and the price of each reservation is set dynamically. The optimal pricing policy is retrieved by using black-box optimization. The applicability of the simulator is evaluated in a real scenario involving 10 structures in Trento, Italy. The adoption of optimized pricing poli- cies based on our simulator leads to an average revenue increase of ≈ 19% with respect to policies with fxed prices. Keywords: Revenue Management, Dynamic Pricing, Hotel Simulation, Simulation-based Optimization 1. Introduction Information Technology (IT) drastically changed how people plan and man- age travels. The Internet revolutionized the way travelers can get information about trips and hotels. Tools such as search engines, online travel agencies or price comparison websites are now used for travel planning by people of diferent 5 generations [1]. As a consequence, potential tourists can take decisions which are much more informed than before, and companies need to be competitive in order to survive. In fact, many organizations in tourism use IT to propose more convenient prices online, thus reducing commissions that would be given * Corresponding author Email addresses: andrea.mariello@unitn.it (Andrea Mariello), m.dalcastagne@unitn.it (Manuel Dalcastagnè), mauro.brunato@unitn.it (Mauro Brunato), roberto.battiti@unitn.it (Roberto Battiti) Preprint submitted to Simulation Modelling Practice and Theory October 13, 2018