In most areas of sub-humid tropics where the rainy period is too short to allow a system with a succession of crops, intercropping is an option to diversify culture and enhance agrosystems resiliency. However, interactions among associated crops (facilitation and/or competition) are complex, variable in time and will depend on the characteristics of each crop and the management of the whole system. Understanding and quantifying these complex interactions and their impacts on the agrosystem productivity require to consider temporal dimension. In fact, oneoff measures or entirely experimental approach cannot adequately answer these questions. However, crop modeling can complete experimental approaches by taking better account of changing interactions over time, allowing dynamic quantification of the flow of resources and their distribution. We will present an example of intercropping model with maize using STICS-CA model, adjusted calibrated and then evaluated for the Brazilian Cerrado system. We will then discuss on how to use a similar approach for millet-cowpea intercropping systems in the Senegalese peanut Basin.