Freshwater lobster farming, particularly Cherax quadricarinatus (redclaw crayfish), has significant economic potential due to high market demand. However, a key challenge is determining the optimal feed dosage to support growth while minimizing waste and environmental impact. This study proposes an IoT-based automatic feed control system using the Mamdani fuzzy model, incorporating input from temperature, water pH, and lobster weight, along with synthetic data for sensitivity analysis.
The results show that the fuzzy logic model provides dynamic feed recommendations based on input variations, demonstrating a nonlinear relationship between temperature, pH, and feed dosage. For example, in the 23°C–25°C range, a
2°C increase raises the feed dosage by 15.183%, while in the 29°C–31°C range, the same increase reduces it by 13.18%. The
centroid defuzzification method is used to generate precise feed dosage values.
This research contributes to the application of IoT-based fuzzy logic systems in freshwater lobster farming, improving
feeding efficiency, reducing waste, and promoting sustainable production. Future research is recommended to include additional parameters and integrate machine learning for better growth prediction and mortality risk mitigation, further enhancing the efficiency and sustainability of the aquaculture industry.