PhD project by Jacinta Bus. Animal welfare is a dynamic process and differs between individuals housed in the same environment. Continuous monitoring of individual pig welfare can be achieved using novel Precision Livestock Farming technology, which combines sensors and intelligent algorithms to track behaviours, such as feeding behaviour.
The welfare of growing-finishing pigs is an important factor in the sustainability of pork production. Currently, pig welfare evaluation is limited to group level assessment at one or a few moments in time. Animal welfare, however, is a dynamic process and differs between individuals housed in the same environment. Continuous monitoring of individual pig welfare can be achieved using novel Precision Livestock Farming technology, which combines sensors and intelligent algorithms to track behaviours, such as feeding behaviour. Pig feeding behaviour follows a diurnal pattern, but deviations may occur during welfare issues, often before clinical signs emerge. Previous studies demonstrated that feeding patterns can be used to automatically detect health issues in pigs, but improvements are required before such a system can be applied in practice.
This study aims to expand our knowledge on individual pig feeding patterns and welfare, and to apply this knowledge to develop a system that records feeding patterns to continuously monitor the welfare of individual, but group-housed, growing-finishing pigs. First, the feeding patterns of commercial growing-finishing pigs are studied to enhance our understanding of the variation in feeding patterns between and within pigs, focusing on rhythmic patterns, temporal development and the influence of pig characteristics and environmental factors. Second, the relationships between individual feeding patterns and welfare (clinical and subclinical health issues, behavioural issues and positive welfare states) are identified. The obtained knowledge will be used to develop an algorithm that automatically processes feeding station data into welfare information and gives early warning signals when welfare issues occur. Finally, I will explore the applications of this algorithm to retrospectively create a representation of pig welfare that encompasses the entire growing-finishing phase. Ultimately, this study will contribute to improving pig welfare and communication of reliable welfare information to pork farmers and other stakeholders.