Electronic Theses and Dissertation

Industri kosmetik Indonesia, khususnya produk perawatan kulit seperti Body Butter,
berkembang pesat, namun dihadapkan pada tantangan manajemen persediaan
akibat ketidakakuratan peramalan penjualan. PT. XYZ Natural Indonesia
mengalami masalah overstock (kelebihan stok) dan understock (kekurangan stok)
berdasarkan rekapan data persediaan dan permintaan tahun 2022 hingga 2024 yang
berdampak pada peningkatan biaya penyimpanan, risiko kedaluwarsa, serta
penurunan kepuasan pelanggan. Penelitian ini menganalisis peramalan penjualan
produk Body Butter menggunakan metode Exponential Smoothing (ES) dan
Autoregressive Integrated Moving Average (ARIMA) dengan data historis tahun
2022–2024. Metode ES dipilih untuk responsivitasnya terhadap perubahan jangka
pendek, sementara ARIMA digunakan untuk mengidentifikasi pola musiman dan
tren jangka panjang. Hasil analisis menunjukkan bahwa Exponential Smoothing
menghasilkan akurasi lebih tinggi dengan nilai MAPE 36,26%, dibandingkan
ARIMA dengan MAPE 41,75%. Simulasi penerapan ES mengurangi risiko
overstock, sedangkan ARIMA lebih efektif pada pola musiman meski memerlukan
optimasi parameter yang kompleks. Berdasarkan karakteristik data dan kebutuhan
operasional, Exponential Smoothing direkomendasikan sebagai metode optimal
untuk PT. XYZ guna meningkatkan akurasi prediksi, efisiensi persediaan, dan
pengambilan keputusan berbasis data. Penelitian ini memberikan kontribusi praktis
berupa rekomendasi kuantitatif pengelolaan stok dan kontribusi teoritis dalam
aplikasi metode peramalan di industri kosmetik, serta menjadi referensi bagi sektor
sejenis.
Kata Kunci : Peramalan Penjualan, Exponential Smoothing, ARIMA, Persediaan,
Industri Kosmetik.

The Indonesian cosmetics industry, particularly skincare products like Body Butter, is growing rapidly but faces inventory management challenges due to inaccurate sales forecasting. PT. XYZ experiences overstock (excess inventory) and understock (shortages), leading to increased storage costs, expiration risks, and reduced customer satisfaction. This study analyzes sales forecasting for Body Butter using Exponential Smoothing (ES) and Autoregressive Integrated Moving Average (ARIMA) methods with historical data from 2022–2024. ES was chosen for its responsiveness to short-term fluctuations, while ARIMA was applied to identify seasonal patterns and long-term trends. The analysis revealed that Exponential Smoothing achieved higher accuracy with a MAPE of 36,26%, outperforming ARIMA’s MAPE of 41,75%. Simulation results showed ES reduced overstock risks, while ARIMA proved more effective for seasonal patterns despite requiring complex parameter optimization. Based on data characteristics and operational needs, Exponential Smoothing is recommended as the optimal method for PT. XYZ to enhance forecast accuracy, inventory efficiency, and data-driven decision-making. This study provides practical contributions through quantitative stock management recommendations and theoretical contributions by advancing forecasting applications in the cosmetics industry, serving as a reference for similar sectors. Keywords : Sales Forecasting, Exponential Smoothing, ARIMA, Inventory, Cosmetics Industry.