An extension to the standard Quadruple tank process is proposed and analyzed. It is shown that the extended setup provides an improved platform for experimental and educational studies of MIMO control systems. Based on dynamic models for the process a control configuration selection is performed and both decentralized and decoupling control schemes are designed and tested. Moreover sparse IMC controllers have been evaluated against a decentralized control scheme for improvement in the stochastic performance. It is shown how a performance assessment method based on the FCOR algorithm can be used to find the optimal tuning of the IMC controller. It can be concluded that the proposed extension to the quadruple tank process is useful for benchmark and educational purposes.