# US 10,769,347 B1 · Predicting no-defect-found physical failure analysis results using Bayesian inference and generalized linear models > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,769,347 B1 - Document ID: US10769347B1 - Status: Granted - Kind: B1 - Application number: 16/397,022 - Filed: 2019-04-29 - Granted: 2020-09-08 - Earliest claimed: 2018-04-27 - Assignees: Synopsys, Inc. - Applicants: Synopsys, Inc - Inventors: Christopher W. Schuermyer - Web page: https://patentgrep.com/patent/US10769347B1 ## CPC - G06F 30/398: Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM] (optical proximity correction [OPC] design processes G03F1/36) - G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model - G06F 30/333: Design for testability [DFT], e.g. scan chain or built-in self-test [BIST] - G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks - G06N 20/00: Machine learning - G06F 2111/08: Probabilistic or stochastic CAD - G06F 2111/20: Configuration CAD, e.g. designing by assembling or positioning modules selected from libraries of predesigned modules - G06F 2119/02: Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA] - G06N 3/063: using electronic means - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A Physical Fault Analysis (PFA) outcome prediction tool utilizes previously-generated evaluation data and associated PFA outcome data to generate a Bayesian Generalized Linear Model (BGLM), and then utilizes the BGLM to generate a PFA outcome prediction for newly-submitted evaluation data that operably characterizes measured operating characteristics of an IC chip that is being developed. The BGLM generation methodology by utilizing a Generalized Linear Model (GLM) in a Bayesian framework to form a hierarchical model representing the evaluation data and associated PFA outcome data as a linear combination. The PFA outcome prediction includes a credible interval of a posterior distribution that effectively represents a cross-sectional portion of the BGLM corresponding to the newly-submitted evaluation data. The previously-generated evaluation data and associated PFA outcome data are stored in a training data library, which is updated to include newly-submitted evaluation data and associated PFA outcome after each PFA is performed. ## Claims ### Claim 1 An electronic design automation (EDA) tool in a computer, the EDA tool including instructions that, when executed by a processor of said computer, cause the processor to perform operations comprising: generating a Bayesian Generalized Linear Model using a training data library including a plurality of first training data entries, where each first training data entry includes corresponding first evaluation data that respectively operably characterizes one or more measured operating characteristics of a corresponding first integrated circuit chip, and corresponding first physical failure analysis result data that operably characterizes results of a corresponding physical failure analysis of said corresponding first integrated circuit chip; and utilizing the Bayesian Generalized Linear Model to generate a physical fault analysis (PFA) outcome prediction for a second evaluation data that operably characterizes said one or more measured operating characteristics of a corresponding second integrated circuit chip, wherein the PFA outcome prediction includes at least one of a posterior distribution and a credible interval, said posterior distribution being a portion of said Bayesian Generalized Linear Model corresponding to said second evaluation data, and said credible interval being a partial certainty region defined by the posterior distribution. ### Claim 2 (depends on 1) The EDA tool of claim 1, wherein generating the Bayesian Generalized Linear Model comprises forming a hierarchical model that utilizes said evaluation data and said associated PFA outcome data as a linear combination such that said Bayesian Generalized Linear Model provides a range of posterior distributions that vary as a function of said first evaluation data. ### Claim 3 (depends on 2) The EDA tool of claim 2, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction comprises utilizing said second evaluation data to select said posterior distribution from said range of posterior distributions. ### Claim 4 (depends on 2) The EDA tool of claim 2, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises calculating said credible interval using a predetermined portion of said posterior distribution. ### Claim 5 (depends on 1) The EDA tool of claim 1, wherein said operations further comprise updating said training data library to include both said a plurality of first training data entries and a second training data entry comprising said second evaluation data and a corresponding second physical failure analysis result data that operably characterizes results of a corresponding physical failure analysis of said corresponding said integrated circuit chip. ### Claim 6 (depends on 1) The EDA tool of claim 1, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises performing a random draw from a Bernoulli distribution, wherein a rate parameter of the Bernoulli distribution is estimated with a logistic link function applied to the first evaluation data. ### Claim 7 (depends on 1) The EDA tool of claim 1, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises utilizing a second training data library including said first physical failure analysis result data and corresponding fabrication data that respectively operably characterizes one or more measured fabrication process parameters utilized during fabrication of said corresponding first integrated circuit chip. ### Claim 8 A method for generating a Physical Fault Analysis (PFA) outcome prediction for a selected integrated circuit device, said method being executed by a processor of a computer and comprising: generating a Bayesian Generalized Linear Model using a training data library including a plurality of first training data entries, where each first training data entry includes corresponding first evaluation data that respectively operably characterizes one or more measured operating characteristics of a corresponding first integrated circuit device, and corresponding first physical failure analysis result data that operably characterizes results of a corresponding physical failure analysis of said corresponding first integrated circuit device; and utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction for a second evaluation data that operably characterizes said one or more measured operating characteristics of said selected integrated circuit device, wherein the PFA outcome prediction includes at least one of a posterior distribution and a credible interval, said posterior distribution being a portion of said Bayesian Generalized Linear Model corresponding to said second evaluation data, and said credible interval being a partial certainty region defined by the posterior distribution. ### Claim 9 (depends on 8) The method of claim 8, wherein generating the Bayesian Generalized Linear Model comprises forming a hierarchical model that utilizes said evaluation data and said associated PFA outcome data as a linear combination such that said Bayesian Generalized Linear Model provides a range of posterior distributions that vary as a function of said first evaluation data. ### Claim 10 (depends on 9) The method of claim 9, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction comprises utilizing said second evaluation data to select said posterior distribution from said range of posterior distributions. ### Claim 11 (depends on 9) The method of claim 9, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises calculating said credible interval using a predetermined portion of said posterior distribution. ### Claim 12 (depends on 8) The method of claim 8, wherein said method further comprises updating said training data library to include both said a plurality of first training data entries and a second training data entry comprising said second evaluation data and a corresponding second physical failure analysis result data that operably characterizes results of a corresponding physical failure analysis of said corresponding said integrated circuit device. ### Claim 13 (depends on 8) The method of claim 8, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises performing a random draw from a Bernoulli distribution, wherein a rate parameter of the Bernoulli distribution is estimated with a logistic link function applied to the first evaluation data. ### Claim 14 (depends on 8) The method of claim 8, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises utilizing a second training data library including said first physical failure analysis result data and corresponding fabrication data that respectively operably characterizes one or more measured fabrication process parameters utilized during fabrication of said corresponding first integrated circuit device. ### Claim 15 A non-transitory computer readable medium comprising stored instructions, which when executed by a processor, cause the processor to: generate a Bayesian Generalized Linear Model using a training data library including a plurality of first training data entries, where each first training data entry includes corresponding first evaluation data that respectively operably characterizes one or more measured operating characteristics of a corresponding first integrated circuit chip, and corresponding first physical failure analysis result data that operably characterizes results of a corresponding physical failure analysis of said corresponding first integrated circuit chip; and utilize the Bayesian Generalized Linear Model to generate a physical fault analysis (PFA) outcome prediction for a second evaluation data that operably characterizes said one or more measured operating characteristics of a corresponding second integrated circuit chip, wherein the PFA outcome prediction includes at least one of a posterior distribution and a credible interval, said posterior distribution being a portion of said Bayesian Generalized Linear Model corresponding to said second evaluation data, and said credible interval being a partial certainty region defined by the posterior distribution. ### Claim 16 (depends on 15) The non-transitory computer readable medium of claim 15, wherein generating the Bayesian Generalized Linear Model comprises forming a hierarchical model that utilizes said evaluation data and said associated PFA outcome data as a linear combination such that said Bayesian Generalized Linear Model provides a range of posterior distributions that vary as a function of said first evaluation data. ### Claim 17 (depends on 16) The non-transitory computer readable medium of claim 16, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction comprises utilizing said second evaluation data to select said posterior distribution from said range of posterior distributions. ### Claim 18 (depends on 16) The non-transitory computer readable medium of claim 16, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises calculating said credible interval using a predetermined portion of said posterior distribution. ### Claim 19 (depends on 15) The non-transitory computer readable medium of claim 15, wherein said operations further comprise updating said training data library to include both said a plurality of first training data entries and a second training data entry comprising said second evaluation data and a corresponding second physical failure analysis result data that operably characterizes results of a corresponding physical failure analysis of said corresponding said integrated circuit chip. ### Claim 20 (depends on 15) The non-transitory computer readable medium of claim 15, wherein utilizing the Bayesian Generalized Linear Model to generate said PFA outcome prediction further comprises performing a random draw from a Bernoulli distribution, wherein a rate parameter of the Bernoulli distribution is estimated with a logistic link function applied to the first evaluation data.