We analyze the effect of using adjusted parameters, corresponding to local and global optimums, in the NRTL thermodynamic model on the complete process synthesis (design, optimization and control) of homogeneous azeotropic distillation columns. The adjusted parameters that correspond to a global optimum were obtained with simulated annealing technique, while the adjusted parameters that correspond to a local optimum were taken from the Dechema Collection. Both sets of parameters were used to design a conventional sequence, a side-stream column, and a Petlyuk sequence. These designs were used as the initial solution to a multiobjective genetic algorithm with constraints handling, coupled to a processes simulator, where the number of stages and heat duty of each column were considered as objectives; as a result, a set of optimal designs, called the Pareto front, was obtained. Then, we chose some designs to analyze their theoretical control properties and the dynamic performance. Results show remarkable differences in structure, energy consumption, control properties, and dynamic performance of these schemes, depending on the use of adjusted parameters. The results show the importance of using the best adjusted parameters available, which in our case correspond to global optimums obtained with the simulated annealing technique.