Abnormal trend detection of sequence-disordered data using EWMA method [wafer fabrication]

Abstract

In this paper, we focus on the design issues of applying the exponentially weighted moving average (EWMA) chart to end-of-line electrical test data. Since the sequence of end-of-line test data is not the same as the sequence in each process step, an abnormal trend in any of the process steps is more difficult to detect based on end-of-line test data than based on single step process data (if available). Our approach uses EWMA chart because the moving average is able to smooth out the sequence-disordered effect and the weighting factor allows us to choose an effective moving average size. The correlation among weighting factor, detection speed, and sequence-disordered effect is studied. Fab data is used to verify the effectiveness of EWMA chart for detecting process shifts if we appropriately choose the weighting factor based on the derived correlation.