Peramalan Permintaan Produk Dagang sebagai Strategi Mitigasi Kerugian Usaha melalui Pendekatan Data-Driven

Abstract

This research aims to improve the effectiveness of inventory management through a data-based approach by integrating forecasting methods and Economic Order Quantity (EOQ) in UD. Snack Award. The forecasting method used is classical decomposition with an additive model using POM-QM software version 5.2, which produces predictions over 14 periods. The results of the comparison between forecast data and actual demand in the January-April 2026 period show a relatively small difference, so the forecasting model is considered to have a good level of accuracy. The forecasting results are then used as a basis in the EOQ calculation to determine the optimal booking policy. Based on calculations, the number of economical orders was obtained of 7,799 boxes with an optimal order frequency of 3 times per year, significantly different from the company's previous policy of placing orders 24 times per year. In addition, a safety stock value of 746 boxes and a reorder point (ROP) of 1,237 boxes were obtained as an inventory control parameter. In terms of cost, the application of the EOQ method was able to reduce the total inventory cost from IDR 32,291,031 to IDR 8,189,310, resulting in savings of IDR 24,101,721 or around 74.6%. The results of this study show that the integration of forecasting methods and EOQ in the framework of data-driven systems is effective in improving planning accuracy, optimizing ordering policies, and minimizing inventory costs.