Optimizing control involves the continual adjustment of the plant operating point to coincide with the economic optimum, subject to prevailing constraints. While the economic optimum generally lies at the intersection of process constraints, the operating point would in practice need to be located a suitable distance into the feasible region to ensure feasibility in the face of unmeasured disturbances. This paper addresses the calculation of the optimal set points by considering the closed-loop dynamics, focusing in particular on rigorous handling of input saturation effects. Two alternative formulations are presented, and it is shown that failure to account for input saturation could result in overly conservative set points. The approach is demonstrated through two case studies, one of which is a fluid catalytic cracking unit (FCCU) example problem. (C) 2003 Elsevier Ltd. All rights reserved.