Penerapan Metode Peramalan Permintaan pada Produk Piece Pivot di PT. Trijaya Teknik Karawang

Abstract

Fluctuations in consumer demand cause companies to experience difficulties in production planning. PT. Trijaya Teknik Karawang is a manufacturing company engaged in fabrication, machining, and stamping parts. The production process is based on job orders. Piece Pivot is one of the parts with the highest number of requests and often fluctuates. Therefore, a strategy is needed, one of which is forecasting. This study aims to compare the four forecasting methods in order to determine the right method to be applied at PT. Trijaya Teknik Karawang. The comparison is done by using the 3-month moving average method, exponential smoothing α = 0.5 and α = 0.9, linear regression and naive method. Comparisons were made using the Mean Absolute Deviation (MAD) error accuracy, Mean Square Error (MSE), Root Mean Square Error (RMSE), and Absolute Percentage Error (MAPE). Based on data processing, the linear regression method produces the lowest error accuracy, namely MAD 2126,573, MSE 6091686.092, RMSE 2468.134 and MAPE 0.264. In tracking signal test, the movement of products meets the requirements, which are between -4 and 4. Then the Linear Regression method that can be applied to PT. Trijaya Teknik Karawang.