The construction of production performance prediction system for semiconductor manufacturing with artificial neural networks

The major performance measurements for wafer fabrication system comprise WI P level, throughput and cycle time. These measurements are influenced by va rious factors, including machine breakdown, operator absence, poor dispatch ing rules, emergency order and material shortage. Generally, production man agers use the WIP level profile of each stage to identify an abnormal situa tion, and then make corrective actions. However, such a measurement is reac tive, not proactive. Proactive actions must effectively predict the future performance, analyze the abnormal situation, and then generate corrective a ctions to prevent performance from degrading. This work systematically cons tructs artificial neural network models to predict production performances for a semiconductor manufacturing factory. An application for a local DRAM wafer fabrication has demonstrated the accuracy of neural network models in predicting production performances.