Shannon's theory is applied for the first time to the simulation of naphtha pyrolysis for estimating selective coefficients of the primary reaction and predicting the feedstock component content. Compared with other methods, this approach can be extrapolated and provides more accurate results. Furthermore, it requires much less experimental data. Not only can exact constraints be used but also uncertainties can be specified by the user, resulting in a highly flexible model. The results demonstrate the effectiveness of Shannon's maximum entropy theory in obtaining the missing information for models of the naphtha pyrolysis process.