Comparative Analysis of Arima Model and Exponential Smoothing in Predicting Inventory in Automotive Companies

Abstract

Forecasting is an activity to predict future events by using and considering data from the past. Forecasting is an important tool in effective and efficient planning. So that demand forecasting can be predicted and the amount of inventory can be determined in order to anticipate the number of varied and fluctuating demand. In order to obtain good forecast results, a forecasting method is used that can predict seasonal data. This study aims to determine the best forecasting model using ARIMA and exponential smoothing methods and to compare the forecasting results with the two methods in order to obtain the best method. Data on the number of requests for cars PT. Suzuki Indomobil Motor 2017 – 2019 is data that contains seasonal patterns so that ARIMA and Holt-Winters exponential smoothing can be used. Data obtained by means of documentation with secondary data collection and literature study. The results show that PT Suzuki Indomobil Motor is more appropriate to use the Holt-Winters Additive exponential smoothing method because the resulting error rate is smaller.