PERBANDINGAN KEAKURATAN PERAMALAN PRODUKSI OBAT DENGAN METODE WINTER DAN METODE DEKOMPOSISI

Abstract

Forecasting is a way of predicting an event in the future based on past information. This research is a quantitative descriptive study with a statistical approach which aims to compare the results of forecasting the amount of OME drug production using the winter method and the decomposition method so that the best method is obtained. Differences in methods certainly provide differences in the accuracy of forecasting results. To see the differences between the two methods, this is done by looking at the comparison of the accuracy of the forecasting results. The research results show that forecasting demand for production quantities in 2022 uses Minitab.17 software with several different alpha, gamma and delta values so that the analysis results from each method can be seen. In the winter method, the smallest forecasting error value for each product is obtained from an alpha (level) value of 0.4, gamma (trend) 0.2 and Delta (seasonal) 0.2 with the accumulated value of OME medicinal products to obtain the MAPE value. amounted to 2.4649, MAD 3.4330 , MSD 17.6957. Meanwhile, for the decomposition method, the smallest forcasting error analysis results were obtained for OME medicinal products with MAPE values of 4.3437, MAD 5.2463, MSD 40.7027. From the results of forecasting error analysis using the OME Minitab.17 product software, it can be concluded that the method chosen to forecast demand for production quantities in 2023 is the winter method because the forecasting error value is smaller compared to the decomposition method.