<p>This study proposed a new approach for optimizing a reinforcement technique using geogrids with wraparound ends for a strip footing resting on a cohesionless sand bed. The proposed method is based on the Multivariate Adaptive Regression Splines (MARS) model and the Multi-Objective Genetic Algorithm (MOGA), which were used to identify the optimal geogrid parameters. MARS was employed to obtain the correlation functions between input parameters related to geogrid installation and output parameters, including ultimate bearing capacity and settlement ratio. The input parameters for geogids included axial elastic stiffness, normalized width of reinforcement, normalized depth of the first layer, normalized vertical spacing between successive layers, normalized vertical length of the wrapping ends, and normalized lap length of the wrapping ends. MOGA was implemented to satisfy optimization conditions and constraints. MOGA results showed that the geogrid parameters proposed by this optimization method effectively maximized bearing capacity, minimized settlement, and minimized reinforcement cost.</p>