Neural Networks are powerful data modeling tools that are capable of capturing and representing complex input/output relationships. The motivation for the development of neural network technology stemmed from the desire to develop an artificial system that could perform "intelligent" tasks similar to those performed by the human brain. Using artificial neural network in approximating of complex data is one of easy way to save time and cost. Ultrasonic vibrations can be applied on the die during the extrusion process. Numerical and experimental analyses have been already performed in the literature on the application of ultrasonic vibrations on the wire drawing, deep drawing, upsetting and rolling processes. No attempts have been made to investigate on the effects of ultrasonic vibrations on the forward extrusion process, yet. A detailed analysis and understanding of the mechanism of improvement is not possible on the basis of conventional experimental observations because ultrasonic vibration processing phenomenon occur at high speeds. Hence, in order to progress the perceptive of the mechanism of ultrasonic vibration extrusion, the finite element analysis was performed by using the explicit analysis procedure. The proposed approach builds on a comprehensive Neuro-Finite Element simulation of the effects of ultrasonic vibration on the forward extrusion. Then use the resulting data to train a Multi-Layer Perceptron (MLP) Neural Network which would predict –accurately enough- those quantities throughout the speeds, vibration amplitudes and frequencies, friction factors and reductions body for any given input vector. The resulting neural simulator is intended to replace the computationally expensive cost-function evaluators that are traditionally used in numerical optimization algorithms. To demonstrate the applicability of the proposed approach, we examine data from FEM as from training with together ultrasonic vibration using the constructed neural simulator and present the results.