When working with x-ray systems, it is important to determine the dose output in order to get the organ dose, equivalent dose, etc. Up to now, different programs exist to simulate the dose output, but the calculations are only based on the filtration applied, the kVp, the ripple and the anode angle. As a consequence, the results of such programs are not characteristic of clinical systems but apply to all of them, hence the lack of precision. The goal of this master thesis is to provide a new program that will estimate the dose output of x-ray systems thanks to a few measurements. Using measurements will characterize the clinical system, and will thus increase the accuracy of the model. This program will work in two major steps: first obtaining a function for the dose output when no filtration is applied, and then for each filtration determining dose reduction factors that should be multiplied to the previous function to get the dose output when a specific filtration is applied. Different models of the dose reduction factor will be proposed, depending on the parameters chosen to describe it (kVp, thickness of copper, HVL or homogeneity coefficient). These models will be compared, and the optimal use will be determined.