Production of biofuel pellets from hardwood, such as beech, is often very troublesome as observed in largescale pellet production plants, where reduction of capacity and frequent blocking may be experienced. The problems may be reduced through a laborious optimization procedure of the process conditions and the utilization of softwood and adhesive materials. To optimize and facilitate such a procedure, a deeper understanding of the fundamental physical-chemical mechanisms that control the pelletizing process is sought after by combining small-scale experiments and an advanced pellet production model. Mixtures of beech rich in corrosive alkali chloride salt, pine softwood, brewers spent grains (BSG), and inorganic additives are experimentally tested using a small-scale pellet mill. It was possible to pelletize a beech/pine mixture containing up to 40% (wt) beech. The addition of 15% (wt) BSG to the beech dust significantly facilitated the pelletizing process. The addition of necessary inorganic anti-slag and anti-corrosive compounds into pellets made of beech may enhance the problems in an unpredictable way. However, if the inorganic additives are added directly into BSG before mixing it with the beech dust, the material can easily be pelletized. A pellet production model is developed that describes the pelletizing pressure variation along the press channels of the matrix. Equations based on differential control volumes are set up to describe the forces acting on the pellet in the matrix. Important model parameters are the sliding friction coefficient, the ratio of compression, and the material-specific parameters, such as the elastic modules and Poisson's ratio. Model calculations show how the variation in the model parameters significantly changes the necessary pelletizing pressure. Using typical material parameters of beech and pine, it is illustrated why beech, in accordance with the experimental test results, is more difficult to pelletize than pine.