Big data analytics is a powerful tool in smart manufacturing (SM). Semiconductor manufacturing data is complex, and also need big data methods to realize correlation analysis. Traditional regression, such as fitting data with certain formula, can't reveal the potential correlation between some semiconductor data. Here, “piecewise” regression (PWR) is proposed, in which raw data is divided into groups, to enhance the potential correlation. This method can be applied to different kinds of semiconductor, such as inline, WAT and CP, to highlight potential correlation and realize predictive analysis.
Discussion(0)
No comments yet. Be the first to comment.