Neural network based fault reduction in wafer scanning system

Manufacturing of Integrated Circuits (ICs) plays a vital role in electronic industry. ICs are fabricated on a thin slice of crystalline silicon called wafers. Wafer Scanners are the equipments used for the manufacturing of Integrated Circuits (ICs). Modeling based approaches are being used to simulate the massive wafer scanner machines. Modeling consumes much time as well as expensive to build and conduct experiments with it. So, the actual implementation set up on wafer scanners with precise positioning measures becomes a difficult process. The IC manufacturing process using data-determined approach's research works are going on. In this paper, a novel approach is proposed to minimize the desired IC pattern's trajectory tracking error in a wafer scanner system. Data-determined approaches are more promising in wafer manufacturing than the model based approaches. A data-determined approach avoids the necessity of modeling and enables the direct tuning of the controller with the estimated data. A data determined approach, Iterative Feedback Tuning (IFT) is used for the optimization and Artificial Neural Network (ANN) is employed for minimizing the trajectory tracking error. The trajectory tracking accuracy is improved through reducing the tracking error of the motor using Artificial Neural Network.