| dc.description.abstract | Livestock management in Sri Lanka remains largely dependent on manual practices and
farmer experience, resulting in challenges such as delayed disease detection, inefficient
feeding strategies, and limited access to veterinary expertise. Although artificial
intelligence (AI)–based livestock management solutions have been widely explored in
prior studies, most existing approaches are either function-specific, technologically
complex, or insufficiently localized for smallholder farmers in developing contexts. This
study aims to analyze and synthesize existing literature to identify how AI-driven health
monitoring, disease prediction, and nutrition planning techniques can be integrated into
a unified, farmer-friendly mobile platform suitable for Sri Lanka. A structured literature
review and secondary data analysis were conducted to examine prior research on
computer vision-based health monitoring, machine learning based disease prediction,
and AI-assisted nutrition planning in livestock systems. The analysis indicates that
machine learning models show strong potential for early disease detection, while
sensor and image-based monitoring techniques effectively identify abnormal animal
behaviour. Additionally, rule-based, and data-driven nutrition planning approaches
have been reported to improve feed efficiency and productivity. However, challenges
related to data availability, infrastructure limitations, model localization, and farmer
accessibility remain significant barriers to practical implementation. Based on these
findings, this study proposes a conceptual AI-powered mobile application framework
without system implementation, providing design insights and research directions for
future development. Finally, the study contributes to the field by identifying research
gaps and offering a structured foundation for developing context-aware AI solutions to
enhance livestock productivity, animal welfare, and decision-making in Sri Lanka. | en_US |
