Using Fuzzy Logic-Based Mamdani to Predict Catfish Larval Rearing

Catfish (Clarias sp.) larvae are highly sensitive to environmental changes, particularly in water quality parameters such as temperature, pH, and dissolved oxygen (DO), which significantly affect their survival rate (SR). This study aims to design a real-time prediction system for catfish larval survival using Mamdani fuzzy logic to support more accurate and adaptive water quality management. The research was conducted from April to May 2025 at the Hatchery, IPB Vocational School. The methodology involves constructing a Mamdani fuzzy inference system in MATLAB based on secondary data (SNI 6484:3:2014 and previous studies) and field observations. Three main input parameters temperature, pH, and DO. Were categorized into fuzzy sets using triangular membership functions. A total of 84 fuzzy rules were developed to infer SR, which was also divided into three categories: low, moderate, and high. Simulation results using the Rule Viewer and Surface Viewer showed that DO had the strongest influence on SR followed by temperature, while pH had a relatively minor effect. Under low DO conditions (<3 mg/L), SR predictions were consistently low regardless of other variables. In conclusion, the Mamdani fuzzy logic system proves effective in predicting catfish larval SR and can be a valuable tool for optimizing aquaculture practices.