Modelling is a powerful and convenient tool for synthesizing knowledge about a complex system, exploring system functioning or simulate its evolution under different weather or technical scenarios. However, such models are few on perennial fruit crops and concern mainly well-studied temperate species but not tropical species. Our aim is to develop a crop model for mango that synthesizes the knowledge acquired on the processes involved in fruit yield and quality development, and biotic and abiotic factors that affect them. The ongoing modelling approach, developed for the cultivar Cogshall in Réunion Island, is presented. The crop model combines complementary phenological, architectural and ecophysiological viewpoints and relies on two sub-models. A structural sub-model accounts for architectural development and phenology of the mango tree, based on endogenous factors and temperature-controlled laws. It predicts the 3D vegetative and reproductive structure dynamics of the tree over successive growing cycles. The structural sub-model is described in Boudon et al. (presented in this symposium). A functional sub-model accounts for carbon- and water- related physiological processes occurring at the branch level and considers weather and source–sink factors, as affected by cultural practices. It predicts fruit growth and quality development. Currently, it is completed with fruit maturation processes and fruit sensitivity to fruit fly infestation; and the model extension at the tree scale is taken into consideration for C exchanges. Further challenging steps are: (i) to link the sub-models and validate the crop model; (ii) to further incorporate the effects of cultural practices (pruning, irrigation, and fruit thinning) on the temporal and quantitative components of tree development; and (iii) to link the crop model to pest models, particularly fruit fly and mango blossom gall midge. From an applied point of view, the global model will allow simulation-based design of mana