The ideal bed occupancy rate (BOR) not only represents or reflects the efficiency of facility utilization, but also directly affects the quality of patient care, distribution of medical resources, and financial stability of healthcare institutions. The Abepura Specialized Hospital (RSK), which functions as a referral hospital in the field of mental health in the Papua region, faces its own challenges in managing bed capacity and the flow of inpatient facility usage. Unlike general hospitals, which have high patient turnover rates, RSK Abepura shows a pattern of much longer hospital stays with low basic occupancy rates and more planned patient visits. To address issues related to increasing hospital BOR values, researchers proposed a BOR prediction model using the time series forecasting method, the Seasonal Autoregressive Integrated Moving Average (SARIMA) method for predicting seasonal patterns tailored to RSK characteristics, comparing the performance of the model built with baseline methods, and validating the model's accuracy. This study successfully developed a SARIMA model for BOR prediction at RSK Abepura with unique RSJ characteristics. The SARIMA (0, 1, 2)(1, 1, 2)7 model with a weekly seasonal period proved to provide the best prediction performance with a WAPE of 52.26%, MAE of 5.54%, and RMSE of 6.81%, outperforming the monthly SARIMA model (s = 30).
Copyrights © 2026