Know more

Precision Feeding in gestating sows

Precision feeding aims to provide each animal or group of animals with the right ration, at the right time, to meet their physiological needs. This system is particularly suitable for pregnant sows reared in groups because it allows to respond to the large variability of nutritional needs within the group and over gestation. Indeed, the model for determining nutritional needs takes into account several individual characteristics of sows (age, weight, thickness of back fat) and of the farm (reproductive performance) in order to best estimate the daily nutritional needs of each sow during gestation (Dourmad et al., 2008; Gaillard et al., 2020). The intake and composition of the rations to be distributed are then calculated by considering the feed available on the farm.

Influence of welfare and health on nutritional needs

The welfare and health state of animals can be associated with behavioural changes: decrease in physical activity, variation in the number of visits and time spent in the automatic feeder, changes in social interactions, etc. These changes lead to altered nutritional requirements. For example, the energy requirements are doubled when a sow is standing compared to a lying position (Noblet et al., 1993). Also, environmental conditions, including temperature and humidity, can cause behavioural variations and thermoregulatory needs, influencing nutritional requirements. Feed intake must be increased to compensate for heat loss when ambient temperature decreases (Noblet et al., 1989; Wegner et al., 2014).

Identification of welfare and health indicators

The first step of this thesis is to perform different experiments to evaluate the influence of situations of “stress” encountered on-farm (establishment of the social hierarchy, competition for feed, noise stress, environmental impoverishment, heat or cold stress) in order to complete a precision feeding database. The sows and the gestation room are equipped with cameras and sensors to record all events: watering, feeding behaviour, physical activity, social interactions, zootechnical performance, noise level, environmental conditions.

The second step will consist, by machine learning, in extracting from the database indicators of welfare and health, in order to quantify the effects of changes in behaviour on the nutritional needs of sows. These indicators will then be integrated into the nutritional models.

The last step is the development of a flexible and robust decision support tool to better adjust feed supplies and to detect and/or anticipate the outbreak of health or welfare problems.

Maëva Durand is working on this subject since October 2020 for 3 years. She is supervised by Jean-Yves Dourmad and Charlotte Gaillard in the Swine Systems team and Christine Largouet in Irisa.

References

• Jean-Yves Dourmad, Michel Etienne, Alain Valancogne, Serge Dubois, Jaap van Milgen, et al. InraPorc: a model and decision support tool for the nutrition of sows. Animal Feed Science and Technology, Elsevier Masson, 2008, 143 (1-4), pp.372-386. ⟨10.1016/j.anifeedsci.2007.05.019⟩. ⟨hal-02664875⟩
• C. Gaillard, Nathalie Quiniou, Raphaël Gauthier, Laetitia Cloutier, Jean-Yves Dourmad. Evaluation of a decision support system for precision feeding of gestating sows. Journal of Animal Science, American Society of Animal Science, 2020, 98 (9), ⟨10.1093/jas/skaa255⟩. ⟨hal-02917057⟩
• Jean Noblet, Jean-Yves Dourmad, Jean Le Dividich, Serge Dubois. Effect of ambient temperature and addition of straw or alfalfa in the diet on energy metabolism in pregnant sows. Livestock Production Science, Elsevier, 1989, 21, pp.309-324. ⟨hal-02717962⟩
• Jean Noblet, X.S. Shi. Energy cost of standing activity in sows. Livestock Production Science, Elsevier, 1993, 34, pp.127-136. ⟨hal-02705223⟩
• Wegner, K. et al. Climatic conditions in sow barns in Northern Germany. Züchtungskunde, Eugen Ulmer Gmbh Co, 2014, 86 (3), pp.200–211.

Contact

• Maëva Durand: maeva.durand[at]inrae.fr (PhD student)
• Jean-Yves Dourmad : jean-yves.dourmad[at]inrae.fr (thesis director)
• Charlotte Gaillard: Charlotte.Gaillard[at]inrae.fr (thesis co-supervisor)
• Christine Largouet: christine.largouet[at]irisa.fr (thesis co-supervisor)