<![CDATA[The increasing demand for animal-based food products has intensified the need for efficient, data-driven livestock management practices. Artificial Intelligence (AI) has emerged as a key enabling technology within Precision Livestock Farming (PLF), supporting automated monitoring, prediction, and decision-making processes. This study presents a Systematic Literature Review (SLR) of AI applications in livestock farming, focusing on application domains, AI models, and evaluation metrics. Following the PRISMA 2020 guidelines, relevant studies published between 2013 and 2024 were systematically identified, screened, and assessed across major scholarly databases, resulting in 20 eligible articles for qualitative synthesis. The findings indicate that AI is primarily applied to animal identification, body weight estimation, disease detection, behavior analysis, and feed management. Deep learning models, particularly Convolutional Neural Networks, dominate image-based tasks, while traditional machine learning approaches remain effective for structured sensor and tabular data. Common evaluation metrics include accuracy, precision, recall, R², and Mean Absolute Error. Despite promising results, the review reveals substantial heterogeneity in datasets, evaluation protocols, and livestock sector coverage, which limits cross-study comparability. This review highlights methodological trends, identifies key research gaps, and provides insights to guide future AI-driven PLF research and implementation.]]>
Copyrights © 2025
