Abstract
In the hospitality industry, demand forecast accuracy is highly impacted by booking cancellations, which makes demand-management decisions difficult and risky. In attempting to minimize losses, hotels tend to implement restrictive cancellation policies and employ overbooking tactics, which, in turn, reduce the number of bookings and reduce revenue. To tackle the uncertainty arising from booking cancellations, we combined the data from eight hotels’ property management systems with data from several sources (weather, holidays, events, social reputation, and online prices/inventory) and machine learning interpretable algorithms to develop booking cancellation prediction models for the hotels. In a real production environment, improvement of the forecast accuracy due to the use of these models could enable hoteliers to decrease the number of cancellations, thus, increasing confidence in demand-management decisions. Moreover, this work shows that improvement of the demand forecast would allow hoteliers to better understand their net demand, that is, current demand minus predicted cancellations. Simultaneously, by focusing not only on forecast accuracy but also on its explicability, this work illustrates one other advantage of the application of these types of techniques in forecasting: the interpretation of the predictions of the model. By exposing cancellation drivers, models help hoteliers to better understand booking cancellation patterns and enable the adjustment of a hotel’s cancellation policies and overbooking tactics according to the characteristics of its bookings.
Highlights
Revenue management’s objective – increasing revenue – is achieved through demand management decisions, i.e., by estimating demand and its characteristics while implementing price and capacity control to “manage” the demand (Talluri & Van Ryzin, 2005, p. 2)
Despite the enormous potential of big data for the hotel industry, the results presented here show that significant performance improvements were only achieved by the addition of features that characterize a hotel’s specific operations (Model 4)
Estimation and forecasting are essential processes in revenue management, and machine learning can help managers improve their results by providing superior accuracy in a more timely way and above all in a more pragmatic way that is not highly dependent on personal estimations or speculations. 5.1 Limitations and future work As is true for most work involving machine learning, the new models’ product is a very complex prediction equation
Summary
Revenue management’s objective – increasing revenue – is achieved through demand management decisions, i.e., by estimating demand and its characteristics while implementing price and capacity control to “manage” the demand (Talluri & Van Ryzin, 2005, p. 2). It is believed that the use of industry-specific data sources such as hotels’ Property Management Systems (PMS), together with weather forecasts, events, and macroeconomic data, may improve forecast accuracy (Chiang, Chen, & Xu, 2007; Ivanov & Zhechev, 2012; McGuire, 2017; Pan & Yang, 2017b; Talluri & Van Ryzin, 2005). The present work will fill this gap by building machine learning models that can be used to predict hotel booking cancellations using large volumes of data from multiple sources This is aimed to answer the challenges mentioned by Antonio et al (2017a) and Pan & Yang (2017b) regarding possible performance improvement in demand forecasting, in the prediction of booking cancellation probability based on the use of big data. Rather than targeting only forecast accuracy as many big data forecasting studies have done (Hassani & Silva, 2015), we wish to use the algorithms’ interpretability features to explore other advantages of using big data and advanced prediction algorithms to understand whether the variables’ predictive power holds for all hotels and to identify the drivers behind the cancellation of bookings, an area that is in need of further research (Falk & Vieru, 2018; Morales & Wang, 2010)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
