Perbandingan Metode Double Eksponensial Smoothing dan Autoregressive Integrated Moving Average dalam Peramalan Omset Penjualan (Studi Kasus UKM Funan Mart)

HABU, Maria Strambi Dwifortiani (2025) Perbandingan Metode Double Eksponensial Smoothing dan Autoregressive Integrated Moving Average dalam Peramalan Omset Penjualan (Studi Kasus UKM Funan Mart). Undergraduate thesis, Universitas Katolik Widya Mandira Kupang.

	Text ABSTRAK.pdf Download (281kB)
	Text BAB I.pdf Download (127kB)
	Text BAB II.pdf Restricted to Repository staff only Download (174kB)
	Text BAB III.pdf Restricted to Repository staff only Download (134kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (408kB)
	Text BAB V.pdf Download (87kB)
	Text DAFTAR PUSTAKA.pdf Download (125kB)

Abstract

Uncertainty in predicting sales turnover is a major challenge for small and medium-sized enterprises (SMEs), especially in making decisions related to inventory management, financial planning, and sales strategies. This study was conducted to compare two forecasting methods, namely Double Exponential Smoothing (DES) and Autoregressive Integrated Moving Average (ARIMA), in predicting sales revenue at Funan Mart SMEs operating in Belu Regency, East Nusa Tenggara. The data used were weekly sales revenue data from November 2022 to September 2023, including total sales and sales by product category. This study employs a quantitative approach using time series analysis, and model accuracy evaluation is conducted using three primary indicators: Symmetric Mean Absolute Percentage Error (SMAPE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The results of the study indicate that the ARIMA method performs better than the DES method. The ARIMA method yields an SMAPE value of 8.42%, an RMSE of 59.525, and an MAE of 47.486, while the DES method produced an SMAPE of 25.85%, an RMSE of 182,501.45, and an MAE of 189.015,26. This indicates that predictions using the ARIMA method are closer to actual values and have a lower error rate. Additionally, ARIMA's ability to capture autocorrelation patterns and moving average components makes it more stable in analyzing complex time series data such as SME sales revenue. Therefore, the ARIMA method is recommended as the primary approach for forecasting sales revenue at Funan Mart SMEs due to its advantages in terms of accuracy, ability to capture complex temporal patterns, and more stable prediction results.

Actions (login required)

View Item