Embedded Tutorial ET2: Volume Diagnosis for Yield Improvement

Abstract

Process variations in sub-nanometer technologies cause systematic defects in manufactured VLSI devices. Such defects may be process dependent as well as design dependent. This requires identification of root causes for systematic defects to aid device yield ramp up. Volume diagnosis or diagnosing a large volume of manufactured devices is necessary to identify systematic defects. Volume diagnosis requires highly efficient and effective software tools since physical failure analysis of a very large number of failing devices is not practical. Typically volume diagnosis uses two procedures. First, responses from failing devices are analyzed using defect diagnosis tools. Next the results of diagnoses are analyzed using statistical, data mining and machine learning techniques to effectively determine the underlying defect distribution for yield improvement. In this presentation, we will discuss diagnosis procedures and methods for analyzing diagnosis data in a typical software based volume diagnosis flow. We will also briefly discuss topics for future research in volume diagnosis.