Simulated moving bed (SMB) processes have been applied to many important separations in sugar, petrochemical, and pharmaceutical industries. However, systematic optimization of SMB is still a challenging problem. Two tailored approaches are proposed, full discretization and single discretization, where both the optimal operating condition and concentration profiles are obtained by a Newton-type solver. In a case study of fructose and glucose separation, it has been found that the full-discretization method implemented on AMPL with IPOPT is more efficient than single-discretization method on gPROMS with SRQPD. The reliability of the full-discretization method is also demonstrated with case studies of a bi-Langmuir isotherm and a PowerFeed optimization problem. (c) 2005 American Institute of Chemical Engineers