Sustainable agriculture in a context of global change should be based on diversified agro-ecosystems with the possibility for actors of community seed systems to exchange seeds. Although several studies have shown that seed flows among actors strongly impact the evolution of crop genetic diversity, there is a lack of knowledge on how seed circulation, farmers' practices and local selective pressures interact together to shape diversity pattern over time. Simulation-based approaches offer an attractive alternative to experimental studies, surveys, and analytical approaches. In the framework of the EU project DIVERSIFOOD, we developped a model called CropMetaPop that simulates the evolution of multilocus and multiallelic genomes. It allows to represent the evolution of different annual crop species by their mating system, along with the management practices of the populations by changing population sizes and selection pressures. It is particularly intented to represent the seed circulation using different network topologies. In order to better understand the functioning of the model, we performed a sensitivity analysis, which allowed us to determine the parameters that have the largest impact on the outputs of the model. We designed three different scenarios to test the behaviour of the model focusing: (i) on the impact of genetic drift, selection and mutation of isolated populations, (ii) on previous scenario plus colonisation and extinction, and (iii) on first scenario with migration. This allowed us to test the impact of 11 related parameters of these scenarios on several demographic and genetic output variables (e.g. survival rate, Nei's genetic diversity, genetic differentiation among sub-populations...). This allowed us to check that the outputs fit the theoretical expectations and to identify the parameters and their interactions that are the most influencial in the different responses. CropMetaPop can now be used as a support for companion modeling of genetic