Prediction of NPK Fertilizer in Chili Plants Using SARIMA Model

Agriculture is one of the key sectors in the Indonesian economy, contributing approximately 14% to the national Gross Domestic Product. Chili plants are one of the main agricultural commodities in the country. Fertilization is crucial in chili plant cultivation to enhance yield and production quality. While proper fertilization can increase plant growth and productivity, excessive doses can lead to plant toxicity. Therefore, an accurate prediction model is needed to estimate the nutritional requirements of chili plants. This study aims to develop a prediction model for the nutritional needs of chili plants using a machine learning approach. The data used come from the Nutrient and Water Supply Monitoring System for chili plant growth and yield based on 10 T. The dataset includes chili plants treated with various doses of Nitrogen <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(N)$</tex>, Phosphorus <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(P)$</tex>, and Potassium <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(K)$</tex> fertilizers monitored by NPK sensors. The machine learning method applied is the Seasonal Autoregressive Integrated Moving Average (SARIMA), which can predict time series data. The model was evaluated using the metrics mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE). The SARIMA model (1,1,1)(1,1,0)[12J showed the best evaluation results, with an average MAE of 0.783, RMSE of 6.377, and MAPE of 0. 005. The SARIMA model can help farmers optimize NPK fertilizer practices to increase productivity and efficiency in the cultivation of chili plants.