Real-time equipment health evaluation and dynamic preventive maintenance

With the great advancement of sensor and information technology, the semiconductor equipment data is now available to process engineers in real time. However the volume of data collected and quickly accumulated is often so large that the data is rarely analyzed and put to use effectively. In this paper, we propose an integrated approach that utilizes multivariate statistical techniques to combine various equipment data items into a single equipment health index. We then use this index to determine appropriate time points for equipment preventive maintenance (PM).