<![CDATA[Rice is a strategic food commodity that provides more than 21% of global human caloric needs and up to 76% in Southeast Asia, including Indonesia. This study aims to analyze the dynamics of rice production and consumption on the island of Sumatra, predict annual rice demand, identify the dominant factors affecting production, and determine the leading rice-producing provinces in Sumatra. The method employed is a Neural Network integrated with the CRISP-DM research methodology to predict annual rice demand, identify key production factors, and determine the top rice-producing provinces. This study uses a dataset from the Central Bureau of Statistics (BPS) covering the years 1993–2020, consisting of six variables: province, year, production, harvested area, rainfall, humidity, and average temperature. The results show that harvested area is the most dominant factor influencing rice production across all provinces in Sumatra. The provinces with the highest rice production are Lampung, South Sumatra, and West Sumatra. The Neural Network model used has an architecture comprising six input nodes, five hidden layers, and one output layer. Model evaluation using Root Mean Square Error (RMSE) yielded a value of approximately ± 636.267 grams (0.636267 tons), indicating the predicted annual change in rice production per province. These findings are expected to assist the government and stakeholders in formulating strategies to stabilize rice production and distribution in Sumatra, thereby reducing price fluctuations and addressing supply imbalances that impact national food security]]>
Copyrights © 2025
