FORCASTING DEMAND DENGAN METODE MULTIPLICATIVE DECOMPOSITION DAN TRACKING SIGNALS DI PT PERSADA NAWA KARTIKA

Abstract

This study chooses the right method to produce demand forecasts that have high accuracy so that it can be considered by PT Persada Nawa Kartika, Kertosomo, Nganjuk to develop future marketing plans. Researchers use exponential smoothing, moving average and multiplicative decomposition to make accurate forecasts. To measure the accuracy of forecasting, researchers used the mean square error (MSE), mean absolute deviation (MAD), mean absolute percentage error (MAPE). The research data used is serial data from the sale of bottled drinking water with the NUCless brand from August 2020 to July 2021. NuCless is a production from PT Persada Nawa Kartika, Kertosono, Nganjuk. Researchers in processing data analysis using mathematical analysis POM QM for windows version 5.2 to produce accurate forecasts. The results of this study recommend the value of forecasting demand for three months, namely August 2021 with 9113 boxes, September 2021 with requests for 13811 boxes and October 2021 with requests for 13811 boxes.