This study outlines the general performance of two different simulation-optimization (S/O) models developed to examine the capabilities of the new proposed management scenario to control seawater intrusion (SWI) in coastal aquifers. In the first S/O model, a finite element (FE) simulation model is directly linked with a multi-objective genetic model. In the second model a trained surrogate model is linked to the same optimization model. The simultaneous abstraction of saline water near the coast and artificial recharge of treated wastewater into the aquifer are the main principles used in the proposed management scenario. The recharge is implemented using a surface pond and therefore unsaturated flow theory is utilized in the simulation. The objective functions include minimization of the total economic cost of the management scenario and also the minimization of the total amount of salt in the aquifer. The results show that implementation of the surrogate model in the S/O framework results in a significant reduction in CPU time.