Chemical vapor deposition (CVD) is known as one of the most efficient methods of synthesizing carbon nanotubes. This Letter presents a computational fluid dynamics (CFD) model to predict the production rate of nanotubes via catalytic decomposition of xylene in a CVD reactor. In our model, two gas-phase reactions and four surface reactions were considered. The rate constants of the surface reactions were determined using the inverse technique based on the measured tail gas concentrations. The predicted production rate agreed well with the experimental data, validating our model.