A software package was developed for the simulation and optimization of a multi-bed adiabatic reactor for the catalytic oxidation of SO 2 , using a heterogeneous plug flow model. The orthogonal collocation (OC) technique with up to eight collocation points was used for the solution of a nonlinear, two-point boundary value differential equation for the catalyst particle, and it was shown that the use of the OC technique with two collocation points can describe the system well. Because of the nonlinear behavior of the effectiveness factor along the bed, optimal catalyst distribution between the beds and corresponding inlet temperatures were determined by two methods, including: the use of (1) intrinsic or (2) actual rate of reaction in the optimization criteria. The results showed that for the second case, the minimum amount of the catalyst can be reached at lower temperatures, the amount of catalyst required is always less, and the number of beds is greater than or equal to that of the first case.