Advanced Micro Devices Corporation

Globalfoundries Inc.

Other

Gregory A Cherry

Matthew A Purdy

Daniel Kadosh

Eric Omar Green

Jin Wang

Elfido Coss, Jr

Christopher Allen Bode

Richard J Markle

Robert J Chong

Brian K Cusson

Ernest Dean Adams, Iii

Alexander James Pasadyn

Thomas J Sonderman

Michael Lee Miller

Richard P Good

Naomi M Jenkins

Howard Ernest Castle

Rajesh Vijayaraghavan

Kevin Andrew Chamness

Patrick M Cowan

Luis De La Fuente

Rick Good

Carl I Bowen

Jason Williams

A method includes receiving a first set of parameters associated with a plurality of die. A first die performance metric associated with a selected die is determined based on the first set of parameters. At least one neighborhood die performance metric associated with a set comprised of a plurality of die that neighbor the selected die is determined based on the first set of parameters. A second die performance metric is determined for the selected die based on the first die performance metric and the neighborhood die performance metric.

Determining die performance by incorporating neighboring die performance metrics

A method and an apparatus for dynamically adjusting a sampling rate relating to wafer examination. A process step is performed upon a plurality of workpieces associated with a lot. A sample rate for acquiring metrology data relating to at least one of the processed workpiece is determined. A dynamic sampling rate adjustment process is performed to adaptively modify the sample rate. The dynamic sampling rate adjustment process includes comparing a predicted process outcome and an actual process outcome and modifying the sampling rate based upon the comparison.

Dynamic adaptive sampling rate for model prediction

A method includes providing a process controller for controlling a process tool. The process tool is controlled in accordance with a process parameter. Measurements associated with the processing parameter for a plurality of runs of the process tool are accessed. A performance measure for the process controller is generated based on the process parameter and the measurements. A system includes a process tool, a process controller, and a performance monitor. The process controller is configured to control the process tool in accordance with a process parameter. The performance monitor is configured to retrieve measurements associated with the processing parameter for a plurality of runs of the process tool and generate a performance measure for the process controller based on the measured processing parameters.

Method and apparatus for assessing controller performance

The present invention provides a method and apparatus for multivariate fault identification and classification. The method includes accessing data indicative of a plurality of physical parameters associated with a plurality of processed semiconductor wafers and providing at least one summary report including information indicative of at least one univariate representation of the accessed data and at least one multivariate representation of the accessed data.

Method and apparatus for multivariate fault detection and classification

A method for identifying faults in a semiconductor fabrication process includes storing measurements for a plurality of parameters of a wafer in the semiconductor fabrication process. A first subset of the parameters is selected. The subset is associated with a feature formed on the wafer. A principal component analysis model is applied to the first subset to generate a performance metric. A fault condition with the wafer is identified based on the performance metric. A system includes a data store and a fault monitor. The data store is adapted to store measurements for a plurality of parameters of a wafer in a semiconductor fabrication process. The fault monitor is adapted to select a first subset of the parameters, the subset being associated with a feature formed on the wafer, apply a principal component analysis model to the first subset to generate a performance metric, and identify a fault condition with the wafer based on the performance metric.

Method and apparatus for detecting faults using principal component analysis parameter groupings

A method includes receiving a current residual vector. The current residual vector is compared to a plurality of historical residual vectors. Each historical residual vector has an associated fault classification code. At least one of the historical residual vectors is selected responsive to determining that the current residual vector matches at least one of the historical residual vectors. A fault condition is classified based on the fault classification code associated with the selected historical residual vector.

Method and apparatus for fault classification based on residual vectors

A method and an apparatus are provided for adjusting a sampling protocol in an adaptive control process. The method comprises determining a performance value based on a measurement associated with at least one or more previously processed workpieces, adjusting a sampling protocol for one or more processed workpieces based on the determined performance value, and measuring the one or more processed workpieces according to the sampling protocol to provide one or more measurements. The method further comprises adjusting at least one of a process model and a control parameter based on at least a portion of the one or more measurements.

Adjusting a sampling protocol in an adaptive control process

A method and apparatus is provided for fault detection and classification based on calculating distances between data points. The method comprises receiving a data sample representative of measurements of one or more variables associated with a process operation, determining a distance between the data sample and one or more data points of a history data of the process operation and detecting a fault associated with the process operation based on the distance between the data sample and the one or more data points of the history data.

Fault detection and classification based on calculating distances between data points

A method and apparatus is provided for fault detection spanning multiple processes. The method comprises receiving operational data associated with a first process, receiving operational data associated with a second process, which is downstream to the first process and performing fault detection analysis based on the operational data associated with the first process and second process using a common fault detection unit.

Fault detection spanning multiple processes

A method includes defining a model of a process for manufacturing a device, the process including a plurality of steps. A plurality of inline process targets are defined for at least a subset of the process steps. The model relates the inline process targets to a plurality of process output parameters. A first set of probabilistic constraints for the inline process targets is defined. A second set of probabilistic constraints for the process output parameters is defined. An objective function is defined based on the model and the plurality of process output parameters. A trajectory of the process output parameters is determined by optimizing the objective function subject to the first and second sets of probabilistic constraints for each process step to determine values for the inline process targets at each process step, the optimization being iterated after completion of each process step for the remaining process steps.

Probability constrained optimization for electrical fabrication control

A method and an apparatus for adjusting a process controller based upon a fault detection analysis. A process step upon a workpiece is performed using a processing tool. Manufacturing data relating to processing of the workpiece is acquired. The manufacturing data may include metrology data relating to the processed workpiece and/or tool state data relating to the tool state of a processing tool. A metrology/tool state data integration process is performed based upon the acquired manufacturing data. The metrology/tool state data integration process includes performing an assessment of a tool health related to the processing tool and adjusting an emphasis of the metrology data based upon the assessment of the tool health.

Updating process controller based upon fault detection analysis

A method and an apparatus for dynamic targeting for a process control system. Inline parameter data relating to a processed workpiece is received. A determination is made whether the inline parameter would result in a value of a device operation parameter within a predetermined range. At least one process operation performed upon the workpiece is adjusted in response to a determination that the inline parameter would not result in a value of the device operation parameter.

Correlating an inline parameter to a device operation parameter

A method and an apparatus for the determination of a process flow based upon fault detection. A process step upon a workpiece is performed. Fault detection analysis based upon the process step performed upon the workpiece is performed. A workpiece routing process is performed based upon the fault detection analysis. The wafer routing process includes using a controller to perform one or a rework process routing, a non-standard process routing, a fault verification process routing, a normal process routing, or a termination process routing, based upon the fault detection analysis.

Growing community of inventors

Austin, TX, United States of America

Gregory A Cherry