<![CDATA[This study examines the accuracy of artificial intelligence-assisted forecasting in supporting hotel revenue management at Amarterra Villas Resort Bali Nusa Dua, a five-star luxury resort operating within Bali’s premium leisure tourism market. The research was motivated by the growing need for hotels to improve forecasting reliability amid fluctuating demand, dynamic pricing conditions, and limited dedicated revenue management manpower. A descriptive quantitative research design with comparative forecasting accuracy analysis was applied using secondary operational data consisting of AI-generated forecasts and actual hotel performance records for Room Nights, Average Daily Rate (ADR), and Revenue. Forecasting accuracy was evaluated using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), Mean Forecast Error (MFE), annual bias percentage, paired sample t-tests, and Pearson correlation analysis. The findings show that AI-assisted forecasting performed more accurately in predicting Room Nights than ADR and Revenue. Room Nights recorded relatively low forecast errors, with an MAE of 3.69, RMSE of 4.61, and MAPE of 11.65%, and the paired sample t-test indicated no statistically significant difference between forecasted and actual demand. In contrast, ADR and Revenue showed higher error values and statistically significant differences, indicating systematic underestimation in pricing and financial performance. Pearson correlation results further suggest that AI forecasts were useful in tracking general trends, although they remained less precise in predicting exact monetary outcomes.]]>
