Nowadays, with increasing the population growth, we can see increasing needs for water in all sectors including industrial, domestic, and agricultural uses. By increasing water demands and on the other hand, decreasing surface water due to drought and climate changes, the extraction of groundwater increased in the last decades. Since groundwater recharge and remediation is very time consuming, overdraft of groundwater is a major challenge for decision-makers. One of the practical and effective solutions which in recent decades have been considered is conjunctive use of surface water and groundwater resources. Conjunctive use is generally defined as the allocation of surface and groundwater resources to one or several uses on a quantitative and/or qualitative basis while certain constraints are also taken into consideration. The advantages of the conjunctive use of surface and ground water resources are not limited to lifting water shortages but it is also capable of enhancing water use efficiency and improving the environmental conditions of irrigated areas. In this study, Najafabad sub-basin located in Zayandeh Rud Basin has considered as case study. This sub-basin due to having the large gardens and agricultural production compared to the other sub-basins undoubtedly is one of the most important sub-basins in the Zayandeh Rud Basin. Negative balance, drawdown in groundwater levels, existing Zayandehrud as a permanent river and Nekooabad modern irrigation systems, and finally high agricultural needs in this area undoubtedly have become this sub-basin to a very complex hydrosystem in terms of water resource management. In this study, due to the need for an integrated management model for allocation of surface water and groundwater, we used a linked simulation-optimization model. In this simulation/optimization model, genetic programming method is intended as a simulation model. This model considered as a surrogate for the numerical simulation model. For training and testing, we have used 24 years hydrological data. In this model, GP used for determine the average groundwater level in response to management strategies suggested by the optimization model. For optimization, we considered a multi-objective management model. Two-objective genetic algorithm with non-dominated sorting method is used as optimization model. For considering all hydrologic conditions, we have used S/O model for left and right Nekooabad regions for wet, dry and normal years. Two objectives of optimization model are: minimizing lake of supply for agriculture needs and also determining optimal crop patterns in order to maximization the net benefit. In the part of simulation model, we have observed very good performance of genetic programming for simulation and prediction stages with R 2 value is about 0.99 and the error less than 0.05. Simulation/optimization model as well as having very good performance has been able to both left and right areas of Nekooabad networks in which planning period, pprovide a model ship, while satisfying all conditions and maximize the net benefit. Keywords: Simulation, Optimization, Water Resource Management, GP, NSGA2, optimum Cropping Pattern, Groundwater, Surface Water