Outpatient forecasting model in spine hospital using ARIMA and SARIMA methods

Background: Examining the matter of how to appropriately allocate the limited supply of medical resources is a crucial issue in terms of the management of a medical institution. Based on the time-series data on all outpatients visiting N hospitals in Gangnam-gu, Seoul from January 2, 2017 to December 31, 2017. Methods: This study utilized Auto Regressive Integrated Moving-Average (ARIMA) and Seasonal Auto Regressive Integrated Moving Average (SARIMA) models to build an outpatient prediction model. And we determined to be ARIMA (3,0,2) and SARIMA (2,0,1) (1,0,0) 6 . Further, the accuracy of the SARIMA model was confirmed by comparing and analyzing the ARIMA model, which was built using the SARIMA model, and its predictability, which is mainly used in the existing forecasting field. Currently, the use of the SARIMA model is extremely rare in areas that predict the number of outpatients in hospitals. Results: Comparing the predicted accuracy of outpatient visits, the SARIMA model was found to be relatively more accurate than the ARIMA model. Conclusions: The study was conducted by applying the time unit at the “daily” level to predict the suspension rather than the quarterly and monthly data used to predict the existing time series. It is thought that this study will serve as basis for hospital-to-house management and policymaking by using the SARIMA model to predict the number of patients visiting hospitals.