A cellular automata approach for modeling swelling-controlled drug release is presented. In the model, a drug release device is divided into a square grid space. Each cell in the grid contains information about the material, drug, polymer or solvent in that domain. Cells are allowed to change their state according to statistical rules designed to mimic physical phenomena. Diffusion and swelling are modeled by a random walk of mobile cells, and kinetics of chemical or physical processes by probabilities of conversion from one state to another. The model is applied to drug release from a swelling binary polymer/drug device. The effect of simulation parameters on the drug release profiles and the locations of erosion and diffusion fronts are considered. The model was able to produce realistic simulations and is proposed as a new tool for the design of controlled release devices.