The knowledge of accurate solid-liquid phase equilibria serves in general as a strong basis for the design of any crystallization process. The available database is often limited and the experimental determination of necessary and reliable thermodynamic data is a tedious and time-consuming task. In this study, the predictive power of a multi-component g(E) model fed with limited experimental data was evaluated. Hereby, the impact of uncertainties in the underlying experimental data and the obtained model parameters was of particular interest. Statistical methods were used in order to extend the knowledge about the reliability of the model correlation and to evaluate its predictive characteristics. Therefore, binary and ternary phase equilibria for the system methionine/water were taken from the literature, and analyzed.