Improving the accuracy of defect diagnosis by considering reduced diagnostic information

Abstract

It was noted earlier that the accuracy of defect diagnosis may be improved if certain tests are removed from consideration by the defect diagnosis procedure. This paper observes that the effects, which support the removal of tests, also support the removal of observable outputs from consideration during defect diagnosis. Specifically, a test may create an output response that a defect diagnosis procedure will not be able to interpret correctly. This may affect some observable outputs more strongly than others. There fore, the removal of observable outputs from con sideration can improve the accuracy of diagnosis. This paper describes a generalized augmented defect diag nosis procedure that removes tests and observable out puts from consideration. It presents experimental results to demonstrate the effects of removing observ able outputs on the accuracy of diagnosis.

Online Scan Diagnosis - A Novel Approach to Volume Diagnosis

“…Yield analysis (YA) is performed to (a) accelerate yield ramp, (b) debug excursions, and (c) improve mature yield. Scan diagnosis is a useful tool in the YA framework [1] [2] [3] [4] [5], and currently it can be summarized into two categories: hardware/tester based methods [6] [7] [8] [9] [10] and software based methods [11] [12] [13] [14]. However, selecting representative units, which highlight either existing or new defect modes, for failure analysis (FA) is difficult.…”