Pressure path is one of the most important parameters in hydrodynamic deep drawing process which affect on thickness distribution and bursting in the parts. In this study, a combination of finite element simulation and artificial intelligence was used to optimize the pressure path in hydrodynamic deep drawing process of cylindrical-conical parts to reach the minimum thickness reduction in critical region of product. First, a finite element simulation model was verified based on experimental results. Then, a neural network model was developed using the data generated from the verified finite element model to predict the thickness in critical region of product. The results indicated that the neural network model can be applied successfully for prediction of sheet thickness. In addition, the neural network model was used as a function in simulated annealing algorithm to maximize the thickness in the mentioned critical region. The final results showed that utilization of the optimized loading path yields good uniform thickness distribution of the part.