METODE PERAMALAN PERMINTAAN PADA USAHA UD. METE MUBARAQ LOMBE KOTA KENDARI

Abstrak

Tujuan dari penelitian ini adalah untuk menganalisis metode peramalan permintaan yang tepat untuk UD. Mete Mubaraq Lombe di Kendari, sebuah UMKM olahan mete, guna mendukung pengambilan keputusan produksi dan pengelolaan persediaan. Menggunakan pendekatan kuantitatif deskriptif, penelitian ini mengumpulkan data sekunder historis permintaan produk dari 2021-2023 dan melakukan wawancara. Menilai berdasarkan Mean Absolute Deviation (MAD), Mean Squared Error (MSE), dan Mean Absolute Percentage Error (MAPE), empat metode peramalan deret waktu adalah Moving Average, Weighted Moving Average, Simple Exponential Smoothing, dan Double Exponential Smoothing. Hasil penelitian membuktikan bahwaMoving Average menghasilkan MAPE tertinggi hingga 31.25% pada September. Weighted Moving Average juga menunjukkan fluktuasi MAPE, mencapai 31.25% pada September meskipun ada 0% pada Agustus. Simple Exponential Smoothing memiliki MAPE terendah 25% di Agustus dan tertinggi 231.25% di September. Metode Double Exponential Smoothing memberikan peramalan terbaik dengan MAPE terendah 2.27% pada Agustus, meskipun mencapai 28.91% pada September, menjadikannya rekomendasi utama untuk efisiensi operasional UD. Mete Mubaraq Lombe.

This research aims to analyze the appropriate demand forecasting method for UD. Mete Mubaraq Lombe in Kendari, an MSME processing cashews, to support production decision-making and inventory management. Using a quantitative descriptive approach, this study collected historical secondary data on product demand from 2021-2023 and conducted interviews. Four time series forecasting methods—Moving Average, Weighted Moving Average, Simple Exponential Smoothing, and Double Exponential Smoothing—were evaluated based on Mean Absolute Deviation (MAD), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE). The analysis results show that the Moving Average method yielded the highest MAPE of up to 31.25% in September. The Weighted Moving Average method also showed MAPE fluctuations, reaching 31.25% in September although there was 0% in August. Simple Exponential Smoothing had the lowest MAPE of 25% in August and the highest of 231.25% in September. The Double Exponential Smoothing method provided the best forecasting results with the lowest MAPE of 2.27% in August, although it reached 28.91% in September, making it the primary recommendation for UD. Mete Mubaraq Lombe's operational efficiency.