Relative permeability modifier (RPMs) is a water-shutoff technique that is implemented mostly in high water-producing wells when conventional mechanical isolation methods are not suitable. To optimize RPM injection treatment, numerical simulation is the most effective tool. In this study, laboratory experiments were conducted to investigate the rheology and characteristics properties of newly formulated polymer grafted bentonite (GB). Baseline GB dispersion rheological model was developed using the concept of the effective volume fraction of particles, which depends on hydrodynamic forces. The model considers the effects of the three most significant factors, namely concentration, shear rate, and temperature. GB was then tested in a coreflood to ensure effectiveness and applicability in real reservoir conditions. Coreflood results were matched using a numerical simulator and the rheology model was incorporated to reduce uncertainty and determine critical GB-related simulation parameters. Subsequently, a pilot-scale simulation, which uses the history-matched coreflood parameters, was built to predict and optimize 5-spot pattern performance. A sensitivity analysis was performed for the effect of GB volume injected, layer thickness, and permeability contrast. The results show that the viscosity equation resulting from the modeling allows appropriate correlation of the experimental data of shear stress versus shear rate with change in concentration and temperature. The simulation results reveal that the injection of GB has improved pilot-scale performance in both water-cut reduction and oil recovery. Sensitivity results indicate that the effective period of GB treatment is longer when treated in low water cut than in high water cut. However, more GB injection results in better water control. © 2023