Abstract: To evaluate a semiconductor-fabrication process, a semiconductor wafer is obtained that includes die grouped into modulation sets. Each modulation set is fabricated using distinct process parameters. The wafer is optically inspected to identify defects. A nuisance filter is trained to classify the defects as DOI or nuisance defects. Based on results of the training, a first, preliminary process window for the wafer is determined and die structures having DOI are identified in a first group of modulation sets bordering the first process window. The trained nuisance filter is applied to the identified defects to determine a second, revised process window for the wafer. A third, further revised process window for the wafer is determined based on SEM images of specified care areas in one or more modulation sets within the second, revised process window. A report is generated that specifies the third process window.

Abstract: To evaluate a semiconductor-fabrication process, a semiconductor wafer is obtained that includes die grouped into modulation sets. Each modulation set is fabricated using distinct process parameters. The wafer is optically inspected to identify defects. A nuisance filter is trained to classify the defects as DOI or nuisance defects. Based on results of the training, a first, preliminary process window for the wafer is determined and die structures having DOI are identified in a first group of modulation sets bordering the first process window. The trained nuisance filter is applied to the identified defects to determine a second, revised process window for the wafer. A third, further revised process window for the wafer is determined based on SEM images of specified care areas in one or more modulation sets within the second, revised process window. A report is generated that specifies the third process window.

Abstract: Hybrid methods for classifying defects in semiconductor manufacturing are provided. The methods include applying a flexible sequence of rules for defects to inspection data. The sequence of rules includes deterministic rules, statistical rules, hybrid rules, or some combination thereof. The rules included in the sequence may be selected by a user using a graphical interface. The method also includes classifying the defects based on results of applying the sequence of rules to the inspection data.

Abstract: Hybrid methods for classifying defects in semiconductor manufacturing are provided. The methods include applying a flexible sequence of rules for defects to inspection data. The sequence of rules includes deterministic rules, statistical rules, hybrid rules, or some combination thereof. The rules included in the sequence may be selected by a user using a graphical interface The method also includes classifying the defects based on results of applying the sequence of rules to the inspection data.

Abstract: An extensible data analysis system for analyzing integrated circuit fabrication data produced during integrated circuit fabrication, including an application tier that selectively runs analysis nodes. The application tier has an architecture for optionally including and excluding a desired selection of the analysis nodes. The application tier architecture allows the selection of the analysis nodes to be dynamically added by a user. A data access tier selectively runs data reader nodes. The data access tier has an architecture for optionally including and excluding a desired selection of the data reader nodes. The data reader nodes interpret a desired variety of data source files containing the integrated circuit fabrication data having different formats for access by the application tier. The data access tier architecture allows the selection of the data reader nodes to be dynamically added by the user.

Abstract: A system for determining an assigned classification for a set of physical events on a substrate. Sensors sense the physical events on the substrate and produce event data. A plug in rule module manager receives and manages any number of plug in rule modules. Each plug in rule module has an input, a local filter, an analyzer, and an output. The input receives the event data and confidence values from preceding plug in rule modules. The local filter analyzes the received confidence values from the preceding plug in rule modules and selectively by passes the plug in rule module based at least in part upon the received confidence values from the preceding plug in rule modules. The analyzer analyzes the event data in view of a given classification associated with the plug in rule module, and assigns a confidence value based at least in part upon how well the event data fits the given classification. The output provides the confidence value to subsequent plug in rule modules.