FORECASTING PERMINTAAN POMPA AIR DANGKAL SHIMIZU MENGGUNAKAN METODE TIME SERIES

Abstract

The amount of production of a shallow well PS 128 bit water pump indicates an imbalance with the number of requests, it indicates that the number of products that are in production is much more than the number of customer requests, causing high warehouse cost. Therefore, forecasting can be a solution to overcome the excess product in the warehouse and can help the company in determining the number of products that should be produced. PT Tirta Intimizu Nusantara, a division of Miyako Group, is one of the successful companies engaged in the home industry by producing electric water pump machines under the Shimizu brand. This research conducted the demand on PS 128 bit shallow water pump product by applying time series forecasting method namely Moving Average and Exponential Smoothing. The uses of both forecasting methods are to compare more accurately and find out the methods that can be used by the company in the next period using the POM QM Software for Windows. Based on the results of calculations, using POM QM Software for Windows that the most suitable method used in analyzing data with the smallest error rate is the Exponential Smoothing method with α = 0.1, which has a MAD value of 6421,412 and an MSE value of 59414,050. It concluded that the Exponential Smoothing method with α = 0.1 is an effective method used to predict the number of shallow water pump requests in the next period (month). The result of forecasting the number of shallow water pump requests in December 2019 using the Exponential Smoothing method with α = 0.1 is 44323.39. Thus, the researcher recommends that in the following year PT Tirta Intimizu Nusantara can use the Exponential Smoothing Method with α = 0.1 to perform demand forecasting.