PENERAPAN METODE ARTIFICIAL NEURAL NETWORK DALAM MEMPREDIKSI HARGA BAWANG MERAH DI PROVINSI KEPULAUAN BANGKA BELITUNG

Abstract

Red onions are a horticultural commodity with high economic value and broad market potential, consumed by nearly all segments of society. This crop has become an irreplaceable staple food due to its continuously growing demand. Price fluctuations in the Bangka Belitung Islands Province pose a challenge for this commodity. This study aims to apply the Artificial Neural Network (ANN) method to predict red onion prices in the Bangka Belitung Islands Province. The data used consists of daily secondary data on red onion prices from January 2018 to June 2025, obtained from the National Strategic Food Price Information Center (PIHPS). The analysis was conducted using Matlab software to build a prediction model, with accuracy levels measured using Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The results of the study indicate that the ANN method with the best architecture is 261-94-1 yields an RMSE value of 0.0273 and a MAPE of 5.1746%.

Keywords: Forecasting, Artificial Neural Network, Red Onion Prices, Bangka Belitung.