Hotel Intelligent System using Machine Learning and Cognitive Tools (Hotelnet)
This project aims to propose a new approach for daily hotel demand forecasting by using clusters of stay dates generated from historical booking data.
Main Applicant:
Heo, C., EHL Hospitality Business School
Co-applicant(s):
Partners:
HotelNet
Start and End Date:
01/01/2020 → 31/12/2020
External Funding:
Private Funding
Project Description
The aim of this project was to present a novel approach to daily hotel demand forecasting by forming clusters of stay dates from historical booking data. This method is distinct from the traditional forecasting techniques for hotels that assume the booking shapes and trends remain consistent in the trailing period. A machine learning algorithm is used to group past booking curves and the additive pickup model is implemented. The effectiveness of this new forecasting approach is evaluated using the real hotel booking data of three hotels, and the results reveal that hotel demand forecasts are more precise when they are generated on a cluster-level for all forecasting horizons.