The increasing use of multimaterial additive manufacturing offers the opportunity to open further fields of application, such as the production of additively manufactured sensors [1]. However, alongside the opportunities of multimaterial additive manufacturing, this also poses challenges. One of these challenges is ensuring a sufficiently strong material bond at the joints [2, 3]. This paper describes the development of a method to automatically optimise the material distribution in joints of parts manufactured via the fused deposition modeling process according to the component requirements and the materials used. The method presented is created by coupling different operations: The main aspect of this approach is the adjustment of a Python-based FEM simulation [4]. The material model is based on the transversely isotropic material behaviour of components produced using the fused deposition modeling process, which has been proven in studies [5]. To accelerate the calculation of the individual components, a voxel-based approach is used, which has already been successfully applied to similar problems [6]. For the method, a three-dimensional arrangement of voxels is designed, to each of which one of the materials used can be assigned. In turn, regions are defined between the respective voxels, to which threshold values regarding component failure from experimental tests are assigned. In a series of simulation runs with increasing loads, regions are identified in which damage occurs at the micro level. This damage is modelled using the widely applied approach of local stiffness reduction [7]. If a macro failure is detected due to the formation of a cluster of damaged regions, the simulation runs are stopped. The load used in the penultimate simulation is then used as fitness value for optimising the material assignment to the voxels using a genetic AI. Initial tensile tests carried out to evaluate the method show a significant improvement of the strength of the optimised joints.