FORECASTING THE NUMBER OF VERTICAL REFERRALS FOR BPJS PARTICIPANTS AT HEALTH SERVICE CENTER IN UNIVERSITAS AIRLANGGA USING ARIMA MODEL

Autoregressive Integrated Moving Average (ARIMA) is a time series forecasting method to estimate future events using past data. It can perform short-term forecasting on data with all types of data patterns, both seasonal and non-seasonal data. This study aims to report the best ARIMA for vertical referral cases in BPJS PLK Universitas Airlangga participants. This method used to analyze the number of vertical references for BPJS participants. The research method used is non-intrusive using secondary data and the sample used was the entire population of vertical point reference data for BPJS PLK UNAIR in 2020. The purposive sampling used until 266 data are found. The results of the analysis show that the ARIMA model used to estimate the number of vertical referrals participants is AR [6] or also known as ARIMA [6,0,0]. As for PLK Campus B UNAIR is ARIMA [0,1,1]. In this model, all diagnostic tests have met the assumption requirements. The results of forecasting the number of vertical referrals in PLK Campus C shows the number of vertical referrals will tend to have a constant or horizontal trend with a smaller reference value than the data in 2020. Meanwhile, the number of vertical referrals of PLK Campus B UNAIR has decreased in number compare to 2020. This forecasting has a MAPE value below 10% so the forecasting model has a very good performance in forecasting examples of vertical referrals for BPJS participants in the future. Then the results are useful for making policies to deal with future cases.