Winegrowers must adapt their viticultural practices taking into consideration new environmental regulations and changing climate. Modelling the viticultural production system may provide guidance and supporting to winegrowers and decision makers. It will furthermore permit researchers and engineers to test prospective scenarios. Modelling must take into account environmental and human factors as well as the quality of products. The objective of the thesis is to propose and test a methodological approach. We designed a conceptual representation of the joint effects of soil, subsoil, landscape and cultural practices on vine functioning. Variables are of different nature and some of them uncertain. We chose appropriate mathematical methods to aggregate variables. These aggregations are based on expert knowledge and automatic learning from data. The joint effects were connected to 'grape/wine' product types that were previously characterized by winegrowers and advisor agents. Modelling and application of the conceptual representation use data from winegrower's plots. Our concept applied to studied plots, allowed showing that it was possible to characterize, in a simple and operational way, the global vine functioning and link it to 'grape/wine' product types. Results validated a two step methodology at the plot level: firstly, characterization of variables evaluating vegetative development and precocity; secondly, characterization of water supply conditions. Vine functioning variables can be predicted. Water supply conditions are predicted according to existing models. Vine development variables can be predicted. Our method highlighted interactions between environmental factors and cultural practices in the prediction of vine vigor. When all of the factors influencing vine development are considered, our results show that a simple model based on direct relationships, coming from expert knowledge and literature, is not efficient enough to predict vine functioning variables. Our modelling method takes into account all the interactions between the variables. Results of this study will aid winegrowers to optimize the choice of their cultural practices, depending on environmental factors and type of wine expected. This method can be applied to protected designation of origin, protected geographical indication and table wines areas.
