The development of new breeding strategies to find more sustainable and productive systems is a major challenge to cope with ceaseless increasing demands for palm oil. Optimizing plant architecture to increase radiation interception efficiency could be an option for enhancing potential crop production. Most oil palms grown in the world are Deli dura x AVROS pisifera crosses displaying large phenotypic variations due to genetic segregation. The objective of this study was to analyse and model oil palm architecture taking into account genetic variability. Data were collected in Sumatra, Indonesia, on 5 progenies (total of 60 palms), to describe the aerial architecture from the leaflet to crown scales. Allometric relationships (using linear and non-linear functions) were applied to model these traits according to ontogenetic gradients and leaf position within the canopy. Results pointed out contrasting architectural traits among genotypes, especially for the leaf insertion angle, leaf length and leaflet shape. Mean values of the simulated architectural traits were compared with field measurements to evaluate the quality of the modelling approach. The allometric relationships, parameterized for different genotypes, were then implemented in the AMAPstudio software to generate 3D mock-ups to estimate integrative traits (such as crown dimensions) that were not directly modelled in the allometric-based model. Forthcoming works will be dedicated to the analysis of the impact of the different architectural characteristics on plant light interception using sensitivity analysis methods in order to identify some key architectural traits that could be considered in breeding programs.