Forecasting occupancy rate using neural network at Hotel R

The hotel occupancy is an important factor to determine various policies in the hotel business. By hotel occupancy, hotel management creating targets in each department, determining hotel room prices, and stock management in F&B (Food and Beverage) and minor operating departments. In this study, occupancy rate forecasting was carried out for Hotel R using the neural network method. The dataset used in this study is room availability, room sold out, and available occupancy percentage at Hotel R from April 2018 to June 2023. The simulation was carried out by dividing the data into training data and testing data with a ratio of 60:40, 70:30, 80:20, and 90:10. Making neural network models is done using one hidden layer and two hidden layers. The forecasting results that have been carried out using a neural network with one hidden layer of optimal results were obtained for a 90%:10% data split with an RMSE of 0.009, while using a neural network with two hidden layers the optimal results were obtained for a 90%:10% data split with an RMSE of 0.016 (a difference of 0.007). The results of this study can be used by Hotel R in determining hotel policies in the future.