Gain scheduled control techniques are widely used in the chemical and aerospace industries but suffer from the limitation to slowly changing scheduling variable (xi). A dynamic gain scheduling (DGS) algorithm is proposed to specifically address this constraint. The control synthesis is based on algebraic transformations of the composite nonlinear controller obtained using the input-output linearization (IOL) and internal model control (IMC) formalisms. The controller is reduced to linear form and implemented in a dynamic gain scheduling approach, scheduled in the two-dimensional xi and xi space. In this fashion, the time variation of the scheduling variable is explicitly accounted for. The algorithm is demonstrated on a simple isothermal continuous stirred tank reactor and a complex, highly nonlinear low-density polyethylene polymerization reactor. Simulation results for the polymerizer case study show that the DGS controller provides satisfactory control during polymer grade changes, outperforms IOL control for disturbance rejection, and is stable under noisy measurements. Performance under parametric uncertainty as well as uncertainty with respect to unmodeled dynamics is also evaluated. Structured singular value analysis for nonlinear and time-varying uncertainty facilitated the determination of theoretical stability of the DGS loop. Finally, extensions to multiple-input-multiple-output systems and systems with higher relative degrees are discussed.