Predicting Dry Matter Intake in Lactating Crossbred Cows Under Semi-Arid Conditions Using an Error-Based Evolving Takagi-Sugeno Fuzzy Model

Accurately predicting dry matter intake (DMI) in lactating crossbred cows under semi-arid conditions is a critical challenge in dairy farming. Precise DMI predictions are essential for optimizing nutrient utilization and enhancing animal performance, particularly in semi-arid environments, where complex and dynamic conditions prevail. This paper introduces a novel Error-Based Evolving Takagi-Sugeno (EBETS) Fuzzy Model designed to address these challenges by dynamically adapting to real-time data and leveraging both current and historical information to enhance prediction accuracy. The EBETS model employs fuzzy set theory and multivariate Gaussian functions to model complex, nonlinear relationships between variables, offering a more robust and flexible approach to DMI prediction. In experiments involving data from 114 crossbred cows under semi-arid conditions, the EBETS model was compared to traditional models and an artificial neural network. It achieved a mean squared error of 0.25, which is more than 90% lower than that of all the other models, while demonstrating a higher correlation coefficient with the expected values, indicating superior predictive performance and reliability. These results underscore the EBETS model’s substantial superiority over existing methods, highlighting its potential to significantly improve dairy farming productivity by providing precise and reliable DMI predictions in challenging environmental conditions. The model’s ability to continuously learn and adapt to changing conditions makes it a promising tool for various applications in agricultural science.