Optimal Plus Ltd.

Ansys, Inc.

Dan Sebban

Shaul Teplinsky

Michael Schuldenfrei

Bruce Alan Phillips

Arie Peltz

Leonid Gurov

Gal Peled

Ashok Alagappan

Craig Damon Hillman

Craig Hillman

A method for analyzing device test data includes accessing a core analytics rule that is based on manufacturing data of a plurality of devices. Each of the plurality of devices are produced in one of a plurality of manufacturing facilities and are of a same type as a first device being tested on a tester. The method also includes receiving initial test results of a plurality of other devices of a same type tested at a testing facility, generating, based on the initial test results, an edge analytics rule, modifying the core analytics rule based on the edge analytics rule, wherein the modified core analytics rule including modified binning limits, applying the modified core analytics rule to testing data obtained by testing the first device, and determining, based on applying the modified core analytics rule, that the first device is an outlier with respect to the modified binning limits.

Method and system for real time outlier detection and product re-binning

A method for generating a reliability performance model includes developing a reliability prediction machine learning model for predicting reliability performance of a product based on data obtained from manufacturing and testing of the product, and obtaining feature names for the reliability prediction machine learning model and their predictive power values. The feature names may correspond to features from the data obtained from manufacturing and testing of the product. The method may further include extracting a set of feature names corresponding to features having highest predictive power values from the feature names, and generating a reliability performance model using one or more model parameters derived from the set of feature names.

Augmented reliability models for design and manufacturing

Methods and systems include receiving a product attribute that identifies a product. A first component attribute and a second component attribute are received. The first component attribute identifies a first component included in the product and the second component attribute identifies a second component included in the product. The methods and systems further includes receiving first manufacturing data associated with the first component and second manufacturing data associated with the second component, applying a set of compatibility rules to the first manufacturing data and the second manufacturing data, determining pairing data from the application of the set of compatibility rules to the first manufacturing data and the second manufacturing data, applying a set of pairing rules to the pairing data, determining one or more actions from the application of the set of pairing rules to the pairing data, and performing the one or more actions.

Methods of smart pairing

A method of identifying defects in an electronic assembly, comprising, by a processing unit, obtaining a grid of nodes representative of a location of electronic units of an electronic assembly, wherein each node is neighboured by at most eight other nodes, wherein a first plurality of nodes represents failed electronic units according to at least one test criterion, and a second plurality of nodes represents passing electronic units according to the least one first test criterion, based on the grid, determining at least one first and second straight lines, and attempting to connect the first and second straight lines into a new line, wherein if at least one node from the new line belongs to the second plurality of nodes, concluding that an electronic unit represented by the node on the grid is a failed electronic unit, thereby facilitating identification of a failed electronic unit on the substrate.

Methods and systems for detecting defects on an electronic assembly

A method of augmenting a reliability model of products including a particular product in use by an end user. The particular product includes a first component and a second component each having component attributes. The method also includes receiving first data of the first component and second data of the second component generated during use of the particular product by the end user and comparing the first data to the second data. The method also includes identifying a measure of compatibility between the components, determining pairing data based on the measure of compatibility, and applying a set of pairing rules to the pairing data. The method further includes modifying the reliability model based at least in part on the application of the pairing rules to the pairing data to generate an augmented reliability model and initiating one or more actions in association with the augmented reliability model.

Augmenting reliability models for manufactured products

A method for generating an augmented reliability performance model for a product includes obtaining a reliability performance model for the product, developing a reliability prediction machine learning model for predicting reliability performance of the product based on data obtained from manufacturing and testing of the product, and obtaining, from development of the machine learning model, feature names for the machine learning model and their predictive power values. The feature names may correspond to features from the data obtained from manufacturing and testing of the product. The method may further include extracting a set of feature names corresponding to features having highest predictive power values from the feature names, and generating the augmented reliability performance model for the product by modifying the reliability performance model to incorporate model parameters derived from the set of feature names.

A method for identifying outlier devices during testing, includes: establishing binning limits for a device being tested based on one or more rules generated from external test results data of tests involving similar devices; receiving test results data in real time for the device being tested while the device is on a device tester; applying the one or more rules to the test results data for the device in real time; determining in real time, based on results of applying the one or more rules to the test results data, whether the device is an outlier with respect to the binning limits; and in response to determining that the device is an outlier, binning the outlier device separately from tested devices having test results data falling within the binning limits.

A method includes receiving system test data for a plurality of electronic systems. Each of the electronic systems includes a plurality of electronic components. The method also includes determining a relationship between a set of electronic components and the electronic systems upon which the electronic components of the set of electronic components are assembled and receiving manufacturing attributes including spatial data for the set of electronic components. The method further includes selecting a data subset from the system test data corresponding to a subgroup of the set of electronic components. The subgroup includes components within an area defined on a substrate according to a spatial pattern and that is fewer than all of the set of electronic components on the substrate. Additionally, the method includes identifying an outlier relative to the data subset and communicating information about the outlier to at least one of a system or a component manufacturer.

Method and system for data collection and analysis for semiconductor manufacturing

A method of augmenting a reliability model for a manufactured product includes receiving a product attribute identifying a product, receiving a first component attribute and a second component attribute, and receiving first manufacturing data and second manufacturing data, the first manufacturing data comprising manufacturing data associated with the first component and the second manufacturing data comprising manufacturing data associated with the second component. The method can include applying a set of compatibility rules to the first manufacturing data and the second manufacturing data, determining pairing data from the application of the set of compatibility rules to the first manufacturing data and the second manufacturing data, obtaining a reliability model of products including the product, augmenting the reliability model based on the pairing data, and performing one or more actions with the augmented reliability model.

Methods and systems include receiving a product attribute that identifies a product. A first component attribute and a second component attribute are received The first component attribute identifies a first component included in the product and the second component attribute identifies a second component included in the product. The methods and systems further includes receiving first manufacturing data associated with the first component and second manufacturing data associated with the second component, applying a set of compatibility rules to the first manufacturing data and the second manufacturing data, determining pairing data from the application of the set of compatibility rules to the first manufacturing data and the second manufacturing data, applying a set of pairing rules to the pairing data, determining one or more actions from the application of the set of pairing rules to the pairing data, and performing the one or more actions.

Disclosed are methods, systems and computer program products where an item may or may not be determined as counterfeit based on result(s) of a comparison between test data for the item, and test data for items that are associated with manufacturing data in the cluster that is most likely to include manufacturing data that is associated with attribute data obtained for the item. In some embodiments, such methods, systems and computer program products may allow automated, universal non-destructive, and/or non-invasive detection of counterfeit electronic items. In some embodiments, counterfeit detection may be integrated into existing supply chains, including high volume manufacturing supply chains, and may be performed for a large variety of items without a need for a major adjustment to manufacturing. However, the counterfeit detection in some embodiments may not necessarily be integrated into manufacturing and may occur at any time, even when an item is in use.

Growing community of inventors

Rishon Lezion, Israel

Dan Sebban