Resilience, when defined as the capacity of an animal to respond to short term environmental challenges and to return to the pre-challenge status, is a dynamic and complex trait. Resilient animals contribute to reinforce the capacity of the herd to cope with often fluctuating and unpredictable environmental conditions. Quantification of resilience at individual scale enables potentially including this trait into future breeding programs, for the animals with a better capacity to adapt to the changing environment, and therefore potentially reduce the risk for disease and improve the welfare of animals. The ability of modern technologies for longitudinal monitoring of multiple indicators of animal performance, at individual scale is a huge step forward to evaluate resilience of farm animals. Resilience is not directly measurable and the development of interpretive models is necessary to quantify indicators associated with resilience. In addition to the quantification of resilience capacity, another application of models described in this paper are detection of periods of perturbations as perceived by the animal, which helps the farmer to adapt an adequate management strategy to help the animal coping with the perturbation, and in this way reducing the use of medicines, and decrease the level of the pain of the animal. These applications do not require explicit knowledge of the origin of the perturbations, and are developed with case studies in real-time and post-treatment of data. The main objective of this paper is to illustrate with examples, different modeling approaches to get the most out of this new generation of data (i.e., with high frequency recording) to detect and quantify animals’ responses to perturbations. In addition, perspectives on the use of hybrid models for better understanding and predicting animal resilience are presented.

The R code of the examples are available at https://quantanimal.github.io/ .