Forecasting the Number of Patient Visits by Arima and Holwinters Method at the Public Health Center

Abstract

As the number of human populations increases and the economy becomes more advanced, people's awareness of health increases. This can increase the number of patient visits if the community will visit for treatment, therefore it is necessary to pay special attention from the health center to carry out readiness in the fulfillment of facilities and service support equipment, such as services in the outpatient registration place where registration documents must be adjusted to the number of existing patients, if the documents are lacking or have not been made, there can be long queues or accumulation of patients which leads to inadequate service. For this reason, the public health center must carry out careful planning activities, one of which is by conducting forecasting activities in order to overcome these problems.This study compares the best method among the 2 time series methods, then the forecasting results will be compared with the actual data to find which forecasting is the best.The final results showed the MAPE value of the arima method for Direct Patient Visits data was worth 22.55% while the Referral Patient Visits were valued at 47.40% with the Moderate/Feasible category, the Holwinters method for Direct Patient Visits data was worth 7.90% while the Referral Patient Visits were worth 11.90% with the excellent category.can be said that the smallest error value is Holtwinters from Direct Patient Visit data with MAPE 7.90% and from Referral Patient Visit data with MAPE 11.90%. Which is where it is said to be an excellent forecasting category