<![CDATA[The production of crude palm oil (CPO) in Indonesia experiences fluctuations influenced by various factors such as rainfall, number of rainy days, and the quantity of fresh fruit bunches (FFB). This study aims to develop a predictive model for estimating crude palm oil production using the Extreme Learning Machine (ELM) method, applied to production data from PT. Bakrie Pasaman Plantations in West Sumatra. ELM was chosen due to its fast learning capability and high accuracy in non-linear regression tasks. The study utilizes historical production data from the past two years. The research process involves data normalization, model training, testing, and performance evaluation using the Mean Absolute Percentage Error (MAPE). The results show that the developed model achieves a good level of accuracy with a MAPE value of 12.07%, which is considered reliable. The predictive model is also implemented as a web-based application that displays forecast results and comparative graphs between actual and predicted data. It is expected that this system can support more effective and efficient production planning.]]>
Copyrights © 2025
