How models contribute to livestock farming system research ? Overview of recent advances
Résumé
The objective of this paper is to bring insights into the contribution and advance of modelling for livestock farming systems (LFS). First, we rely on a few examples to illustrate how models can be powerful tools to increase our understanding of livestock production. Second, we argue that: (i) the choice of model type depends on the goal that modeller wants to achieve; (ii) any model is a compromise among generality, precision or realism; and (iii) any model requires hypotheses, data and careful evaluation against reality. Third, we develop two examples showing that new challenges faced by LFS demand to incorporate new scales of analysis. The dairy goat model shows that it is possible to simply formalize basic biological processes at female level. This representation is crucial to increase our understanding on the role of the variability of female's lifetime performance at herd level. The co-viability model shows that accounting for grazing impact on the conservation of biodiversity requires enlarging spatial and temporal scale of LFS analysis. It thus implies a simplification of grazing process formalization. Both examples open new fields of collaborations between animal production sciences and other disciplines such as animal nutrition and population ecology.