Abstract A large number of engineering applications involve granular material or a particulate phase in combination with a gaseous or liquid phase. Predominant applications are as diverse as pharmaceutical industry e.g. drug production, agriculture food and processing industry, mining, construction and agricultural machinery, metals manufacturing, energy production and systems biology. Common to all these application is that they cover a large spectrum of length scales ranging from inner particle length scales to global dimensions of the reactor. In order to describe the processes and their interaction accurately, tailored algorithms are required for prediction and analysis. The current numerical approach of the Extended Discrete Element Method (XDEM) is based on an Eulerian-Lagrange coupling. For this purpose the solid phase consisting of individual particles is treated by the Lagrange method that describes both the dynamic state i.e. position and orientation of each particle in space and time and its thermodynamic state e.g. internal temperature and species distribution. The flow of gas in the void space between the particles is predicted by traditional and well-proven Computational Fluid Dynamics (CFD) taking into account heat and mass transfer between the particles and the surrounding gas phase. Hence, the entire process represented by the sum of all particle processes in conjunction with fluid dynamics. The afore-mentioned numerical concept was applied to predict pyrolysis of a packed bed of wood particles in a cylindrical reactor. A comparison of predicted results with experimental data show good agreement. Hence, the numerical concept is able to resolve a large range of length scales for solid reaction engineering. An analysis of detailed results helps to uncover the underlying physics of the process, and thus, allows for an improved design and operation conditions.