Analisis Perbandingan Metode Regresi Linear dan Metode Exponential Smoothing Dalam Peramalan Penjualan Keripik Ubi Kayu dan Keripik Ubi Rambat

Abstract

Forecasting plays a crucial role in the decision-making process of any company or business organization. The aim of this study is to minimize potential losses through effective production planning. The study utilizes linear regression and exponential smoothing as forecasting methods. The results obtained from the study are specific to the cases in UD. Mustika Tip and it concludes that linear regression is the most effective prediction method. For cassava chips, the study reports a Mean Absolute Deviation (MAD) of 3,060.19 MAD, a method exponential smoothing (MSE) of 12,742,472.70, and a Mean Absolute Percentage Error (MAPE) of 0.52%. For yam chips, the MAD is 272.65, the MSE is 248,835.91, and the MAPE is 0.12%. Furthermore, the study recommends safety stock levels for both cassava chips and yam chips. The suggested safety stock for cassava chips is 211 kg, while for yam chips, it is 28 kg.