Solid fuel conversion in a fixed-bed is a challenging modelling task due to different time and length scales and the importance of heat transfer mechanisms. The current study aims to propose a novel model that can capture all the main features of the conversion of fuel bed while maintaining a moderate computational cost. The model is based on the commonly applied porous media approach, which describes the solid phase using an Eulerian framework. A layered particle submodel with four types of solids, wet wood, dry wood, char, and ash, is implemented to account for different conversion stages. At each computational cell, a matrix is used to record the information on all the properties of the four types of solids, including the number and volume of particles. The model allows the exchange of particles between cells, thus capable of simulating the motion of the fuel bed during conversion, such as bed collapsing. In addition, the new model can efficiently calculate heat transfer between particles and particles and fluid in each computational cell. The proposed model is validated against a series of experiments on biomass conversion in a rectangular fixed-bed combustor operated in a counter-current mode with various air supply rates. Good agreement with experiments was found even at the limited combustion regime. With the overall low computational cost generated by the bed model, the proposed model framework has the potential to efficiently simulate a wide range of solid fuel conversion processes at a large scale, not only in fixed-beds but also in moving beds and rotary kilns.