<![CDATA[Abstrak Penelitian ini bertujuan membandingkan akurasi metode Moving Average dan Exponential Smoothing dalam peramalan bahan baku, produksi, dan ekspor fillet ikan di PT Beejay Seafood. Data historis dianalisis menggunakan Moving Average 3, 5, dan 7 bulan serta Exponential Smoothing dengan α=0,2; 0,5; dan 0,8. Tingkat akurasi diukur melalui Mean Absolute Deviation (MAD), Mean Squared Error (MSE), dan Mean Absolute Percentage Error (MAPE). Hasil penelitian menunjukkan bahwa Exponential Smoothing memberikan nilai MAPE lebih rendah dibandingkan Moving Average, khususnya pada variabel ekspor (MAPE 19,53% vs 27,11%) dan produksi (MAPE 22,46% vs 29,08%). Sementara itu, Moving Average masih cukup efektif untuk variabel bahan baku dengan perbedaan kesalahan yang relatif kecil antar metode. Secara keseluruhan, Exponential Smoothing dinilai lebih tepat digunakan dalam kondisi data yang bersifat fluktuatif, karena mampu menyesuaikan bobot data terbaru secara lebih adaptif. Temuan ini menegaskan bahwa pemilihan metode peramalan berpengaruh terhadap hasil analisis, serta dapat menjadi dasar bagi perusahaan dalam meningkatkan akurasi perencanaan operasional dan strategi bisnis.Kata kunci: Exponential Smoothing, Moving Average, Fillet Ikan, Ekspor, Perbandingan MetodeAbstractThis study aims to compare the accuracy of the Moving Average and Exponential Smoothing methods in forecasting raw materials, production, and fish fillet exports at PT Beejay Seafood Tbk. Historical data were analyzed using 3-, 5-, and 7-month Moving Average as well as Exponential Smoothing with α=0.2, 0.5, and 0.8. Forecasting accuracy was evaluated using Mean Absolute Deviation (MAD), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE). The results indicate that Exponential Smoothing achieved lower MAPE values compared to Moving Average, particularly for export (19.53% vs. 27.11%) and production variables (22.46% vs. 29.08%). Meanwhile, Moving Average remained relatively effective for raw materials, where the error differences between methods were smaller. Overall, Exponential Smoothing is considered more suitable for fluctuating data patterns, as it adapts better to recent changes in the time series. These findings highlight the importance of selecting appropriate forecasting methods to improve accuracy and provide a stronger basis for operational planning and business strategy in the fishery industry.Keywords: Exponential Smoothing, Moving Average, Fish Fillet, Esportation, Methodological Comparison]]>
Copyrights © 2026
