Creating resilience in pigs through artificial intelligence

As global meat consumption is projected to increase (FAO, 2021), there is a growing demand for pig meat. However, the intensive production of pig meat raises concerns about the health and welfare of the animals. Pigs that are resilient are better equipped to cope with environmental stressors, leading to improved health and welfare. Measuring resilience directly is challenging, but one potential indicator can be abnormalities in pig’s behavior, health, and the environment.

Sensor technologies can provide a powerful tool for monitoring pigs' status and detecting early signs of risks to their impaired resilience. As welfare problems in pigs, such as tail-biting or respiratory diseases, are often caused by multiple factors, integrating data from different sensors can improve risk prediction accuracy and reliability since various sensors capture different aspects of pig performance.

The objective of this PhD project is to develop machine learning algorithms that can effectively integrate and fuse data from multiple sensors to extract meaningful information on pigs' resilience. This aspect of the project is innovative and significant, as research on data fusion and multi-sensor systems in pig production is still limited. By leveraging the power of machine learning and sensor technologies, this project aims to predict welfare and health problems in an early stage so that remedial actions can be undertaken, thus, providing practical solutions to enhance pig production sustainability.