PREDIKSI PRODUKSI BIBIT IKAN NILA MENGGUNAKAN MULTIPLE LINEAR REGRESSION DI BALAI BESAR PERIKANAN BUDIDAYA AIR TAWAR (BBPBAT) SUKABUMI

The production of Nile tilapia fingerlings is a critical component for the sustainability of freshwater aquaculture. This study aims to predict the number of fingerlings produced based on environmental factors and feed input, with a case study conducted at the Center for Freshwater Aquaculture Development (BBPBAT) in Sukabumi, Indonesia. The Multiple Linear Regression (MLR) algorithm was used to model the relationship between water temperature, pH, dissolved oxygen (DO), ammonia concentration, and feed quantity with fingerling production output. The historical dataset used consisted of 147 records, which were analyzed and evaluated using several performance metrics: a coefficient of determination (R²) of 0.836, a Mean Absolute Error (MAE) of 35.664, a Mean Squared Error (MSE) of 2,014,982,858, and a Root Mean Squared Error (RMSE) of 44.852. The results indicate that the model achieved high accuracy and was able to explain a substantial portion of the variance in production data. To enhance field implementation, the model was deployed as a user-friendly web-based prediction system using the Streamlit framework, enabling interactive data exploration and real-time prediction. User
acceptance testing of the prediction system was conducted using a Likert scale and managing Nile tilapia fingerling production.. questionnaire distributed to 11 respondents, yielding an average score of 4.236 out of 5, indicating that the system is well accepted by users and considered easy to use. This study demonstrates that MLR can be effectively utilized as a decision-support tool in planning
Keywords: Production Prediction, Multiple Linear Regression, BBPBAT Sukabumi.