The test process generates the most data in the enterprise, but the lack of tools to handle the data properly has limited its utilization. In particular, the data generated in the wafer test process reflects the quality level of the fab, but engineers were only utilizing a portion of it. In order to increase the utilization of such test data and contribute to yield/quality improvement, we developed a user-friendly CTQ Define Add-in that derives CTQ of wafer testing by incorporating ML algorithms supported by JMP. In this presentation, we will introduce a series of analysis processes based on the add-in to define CTQ for each target and derive the optimal Coefficient conditions for field application for CTQ Items.
