Two new model based control schemes have been developed in the present study and implemented on a moderate purity two product pilot scale distillation column. The first method modifies the well-known dynamic matrix control (DMC) algorithm by making it adaptive. The second one uses a nonlinear model predictive controller based on a semirigorous analytical model of the process. A multilayer control scheme has been developed to implement these techniques. The experimental results showed that both the approaches are superior to the conventional PI control and have the potential to meet the industrial process control requirements.