Disruptions are sudden interruptions of plasma discharges in tokamaks. They are due to instabilities leading to the loss of thermal energy and magnetic energy of the plasma over periods of the order of a few tens of milliseconds. Disruptions can generate so-called relativistic runaway electron beams carrying a large part of the initial plasma energy, and likely to damage plasma-facing components. The proposed PhD focuses on the characterization and modelling of runaway electrons impact asymmetries on the wall. It is likely that runaway electrons will be generated during the lifetime of future machines, even though preventing their generation or suppressing them is highly desirable. Unfortunately the geometry and physical processes at work during impacts are still poorly understood. In particular, asymmetries in the toroidal direction have been observed on many tokamaks, concentrating the heat flux with reproducible patterns over time and despite varied experimental conditions. Few controlled experiments have been performed to study these phenomena. It is therefore proposed for this topic to start by building a statistical review of recent experimental impact data on JET and WEST tokamaks: deposition surface, peaking factors, heat flux, ejecta characterization. Simple heat propagation codes will be used. The characteristics of the decoupled electrons just before the impact should also be part of the study, using indirect measurements (hard X-ray spectra, post-mortem measurements) or interpretative codes. In a second step, runaway beam impact simulations will be carried out to test the two main hypotheses that explain the asymmetries: misalignment of the wall elements, or an intrinsically three-dimensional structure of the beam, potentially created by error fields. The 3D MHD code JOREK will be used, in particular for the second hypothesis. The goal will be to reproduce the experimental observations. Finally, once the correct hypothesis has been validated and the model developed, the simulations will be extended to ITER where the thermal loads and asymmetries of the beam impact will be calculated from potential values of misalignments and/or error fields.