The identification of the correct, stable solution to a phase equilibrium problem, given a particular thermodynamic model, is essential for the design of separation processes.It is also important in the selection of an appropriate model to represent experimental data.The need for a completely reliable method to test for phase stability is particularly pressing when the number of phases likely to be present is not intuitive to the user, as is frequently the case with high-pressure systems. Previously, we have a presented a completely reliable computational technique, based on interval analysis, to correctly identify phase equilibrium and test for phase stability in binary solvent-solute systems, that include the possibility of a solid phase, using any of a variety of cubic equations of state as the thermodynamic model.Here we extend the methodology to include multicomponent solvent-solute-cosolvent systems where the likelihood of additional phase formation is even greater than in the binary case. Gaseous or liquid cosolvents are frequently used in supercritical fluid extraction processes, and are integral in processes such as the gas anti-solvent process (GAS) to precipitate uniform solid particles. Using several examples from the literature, we demonstrate how the new computational technique can be used to identify experimental data that may have been misinterpreted and to identify models that do not predict what the modeler intended.