The profitability of chemical processes strongly depends on their control systems. The design of a control system involves selection of controlled and manipulated variables, known as control structure selection. Systematic generation and screening alternative control structures requires optimization. However, the size of such an optimization problem is much larger when candidate controllers and their parameters are included and it rapidly becomes intractable. This paper presents a novel optimization framework using the notion of perfect control, which disentangles the complexities of the controllers. This framework reduces the complexity of the problem while ensuring controllability. In addition, the optimization framework has a goal-driven multi-objective function and requires only a steady-state inverse process model. Since dynamic degrees of freedom do not appear in a steady-state analysis, engineering insights are employed for developing the inventory control systems. The proposed optimization framework was demonstrated in a case study of an industrial distillation train. (C) 2011 Elsevier Ltd. All rights reserved.