DETAILED ACTION
Notice of AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 1-40 are cancelled.
Claims 41-62 are pending and are rejected.
Priority
Provisional:
Acknowledgment is made of applicant’s claim for priority to provisional application no. 63484810 filled on 02/14/2023.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 06/05/2026, 04/03/2026, 02/25/2025, 07/24/2024 and 05/29/2024 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement(s) is/are being considered by the examiner.
Drawings
Drawings filled on 02/08/2024 are acceptable for the examination purpose.
Claim Objections
Claim 41 is objected to because of the following informalities:
Claim 41 recites, the one or more predicted metrics are useable to characterize an output of a process performed by a process equipment. The use of word useable makes the claim limitation vague, because it makes it unclear whether the one or more predicted metrics were actually ever getting used to characterize the output. The word useable means can be used, but doesn’t say that it is actually being used.
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims fall within the judicial exception of an abstract idea:
Claims 41-62 are rejected under 35 U.S.C. 101 because the claimed subject matter is directed to an abstract idea without significantly more.
Step 1:
Claims 41-62 are directed to a system that falls within one of the four statutory categories (i.e., process, machine, manufacture, or composition of matter).
Step 2A:
The claims 41-62 fall within the judicial exception of an abstract idea. Specifically, Mental processes (see MPEP § 2106.04(a)(2), subsection III) such that concepts performed in the human mind or with pen and paper including observation, evaluation, judgment and opinion and/or Mathematical concepts (see MPEP § 2106.04(a)(2), subsection I) such as mathematical relationships, mathematical formulas or equations, mathematical calculations.
Step 2A – Prong 1:
Claim 41:
…process a plurality of data types and datasets from a plurality of different sources for generating training data;
…optimizing a model; and
…use the model for generating one or more predicted metrics substantially in real-time, wherein the one or more predicted metrics are useable to characterize an output of a process performed by a process equipment.
These limitations describe, receiving dataset to optimize a model and then using the optimized model to predict metrics.
These limitations given its broadest reasonable interpretation in light of the specification is a Mental Process since this procedure can be performed in the human mind and/or using pen and paper and is an observation, analysis/ evaluation/ calculation and determination; and is Mathematical concepts such as mathematical calculations (generating metrics).
Claim 42:
… generating the one or more predicted metrics…
This limitation describes, determination of predicted metrics.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical calculations (generating metrics).
Claim 46:
… generating the one or more predicted metrics…
This limitation describes the claim 41 generic model being a virtual model and doesn’t further narrow the claim with any particular algorithm or mathematical technique.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical calculations (mathematical model).
Claim 51:
… validate the historical process data and the historical measurement data against the operation data and the equipment specification data...
This limitation describes, validating by comparing two different data sets.
These limitations given its broadest reasonable interpretation in light of the specification is a Mental Process since this procedure can be performed in the human mind and/or using pen and paper and is an observation and analysis/ evaluation (validating data by observing/analyzing/comparing one set of data with another).
Claim 54:
… process the plurality of data types or datasets by generating a component hierarchical structure of the process equipment, wherein the component hierarchical structure comprises a nested structure of (i) the process equipment and (ii) one or more components that are used within or in conjunction with the process equipment...
This limitation describes, process data to generate hierarchical structure.
These limitations given its broadest reasonable interpretation in light of the specification is a Mental Process since this procedure can be performed in the human mind and/or using pen and paper and is an observation and analysis/ evaluation (process data to generate hierarchical structure).
Claim 55:
… process the plurality of data types or datasets by generating a step-operation hierarchical structure of a recipe for the process, wherein the recipe comprises a plurality of steps, and wherein each step of the plurality of steps comprises a plurality of different sub-operations...
This limitation describes, process data to generate step-operation hierarchical structure.
These limitations given its broadest reasonable interpretation in light of the specification is a Mental Process since this procedure can be performed in the human mind and/or using pen and paper and is an observation and analysis/ evaluation (process data to generate step-operation hierarchical structure hierarchical structure).
Claim 56:
… process the plurality of data types or datasets by removing one or more data outliers;
pre-process and remove data outliers from the process data…
These limitations describe, processing/preprocessing data such that removing filtering certain data.
These limitations given its broadest reasonable interpretation in light of the specification is a Mental Process since this procedure can be performed in the human mind and/or using pen and paper and is an observation, analysis/ evaluation, and determination (processing/preprocessing data such that removing filtering certain data).
Claim 57:
… (i) feature engineering,
(ii) time-aware data normalization, and/or
(iii) an adaptive learning algorithm,.…
These limitations describe, generic mathematical operations, mathematical data manipulation, common mathematical data normalization/scaling, and dynamically adjusting models.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical relationships, mathematical formulas or equations, mathematical calculations (generic mathematical operations, mathematical data manipulation, common mathematical data normalization/scaling, and adjusting models continuously to update or make the previous model better).
Claim 58:
the feature engineering comprises
(i) an extraction of a plurality of features from raw trace data or sensor data within the training data and
(ii) use of an algorithm to select one or more features from a list of extracted features, based at least in part on local relationships between process data and measurement data
These limitations describe, feature extraction from data and select features by observing extracted data by using mathematical algorithm/equation.
These limitations given its broadest reasonable interpretation in light of the specification is a Mental Process since this procedure can be performed in the human mind and/or using pen and paper and is an observation, analysis/ evaluation, and determination (feature extraction from data and feature selection).
Claim 59:
… the time-aware data normalization comprises a decomposition of time series data into one or more components including smoothing data, trend data, and/or detrend data.
These limitations describe, data normalization and decomposition using various mathematical techniques.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical relationships, mathematical formulas or equations, mathematical calculations (data normalization using mathematical relationships/calculations).
Claim 60:
… the adaptive learning algorithm comprises one or more adaptive online ensemble learning algorithms.
These limitations describe, specific mathematical algorithm that is adaptive.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical relationships, mathematical formulas or equations, mathematical calculations (update mathematical model).
Claim 61:
optimize the model using at least in part hyperparameter optimization
These limitations describe, model optimization using mathematical relationships.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical relationships, mathematical formulas or equations, mathematical calculations (mathematical optimization).
Claim 62:
train the model with a set of hyperparameters on an output from the machine learning pipeline;
evaluate a performance of the model based on validation data, wherein the validation data is split from the training data for the hyperparameter optimization;
use a hyperparameter optimization algorithm to select a set of hyperparameters for a next iteration based on past performance, so as to increase or improve the performance of the model; and
repeat (i)-(iii) iteratively until the performance of the model meets termination criteria optimize the model using at least in part hyperparameter optimization
These limitations describe, generic mathematical optimization loop and iterate until convergence.
These limitations given its broadest reasonable interpretation in light of the specification is Mathematical concepts such as mathematical relationships, mathematical formulas or equations, mathematical calculations (generic mathematical optimization loop).
Step 2A – Prong 2 and Step 2B:
This judicial exception is not integrated into a practical application because the additional elements including the following limitations database (claim 52), and modules (various different modules claimed in claims 41-45, 51, 54-56 and 61-62) as recited in the claims are mere instructions to implement an abstract idea on a general purpose computer and the modules perform generic computer functions (receive, process, train and predict) (apply it; corresponding structure disclosed in the specification is a general purpose computer implementing the claimed functions characterized as abstract ideas above. See MPEP 2106.05(a)). Data processing module (claims 41, 51, 54-56), training and optimization module (claims 41 and 61-62) and inference module (claims 41-43 and 56) implemented via generic machine learning techniques implemented using the general purpose computer. The claim limitations are implemented on these generic elements such that the following are merely applying the abstract idea on a generic computer: determining, evaluating, analyzing, and determining etc.
Claim(s) 41 recites, “A system for process monitoring and control, comprising:” This is generally linking the use of a judicial exception to a particular technological environment or field of use. See MPEP 2106.05(h).
Claim(s) 43 recites, “provide the one or more predicted metrics for the process control, or for process monitoring, improvement or trouble-shooting.” This is generally linking the use of a judicial exception to a particular technological environment or field of use, because these determined predicted metrics are provided for control, monitoring, and trouble-shooting, but actual control, monitoring, and trouble-shooting is not being performed. See MPEP 2106.05(h).
Claim(s) 44 recites, “use the one or more predicted metrics to detect, correct, or mitigate a drift, a shift, or a deviation in the process or the process equipment.” This is generally linking the use of a judicial exception to a particular technological environment or field of use, because these determined predicted metrics are used for to detect, correct, or mitigate a drift, a shift, or a deviation, but actual control is not being performed. See MPEP 2106.05(h).
Claim(s) 45 recites, “use the one or more predicted metrics to improve process productivity via integration with run-to-run control.” This is generally linking the use of a judicial exception to a particular technological environment or field of use, because these determined predicted metrics are used to improve process productivity via integration with run-to-run control, but actual control is performed to achieve the improved process productivity. See MPEP 2106.05(h).
Claim(s) 46 recites, “the model comprises a virtual metrology (VM) model.” This is generally linking the use of a judicial exception to a particular technological environment or field of use (linking the virtual model to a semiconductor field of use). See MPEP 2106.05(h).
Claim(s) 47 recites, “the process equipment, wherein the process equipment comprises a semiconductor process equipment,” claim 48 recites, “the output of the process comprises a deposited or fabricated structure, wherein the deposited or fabricated structure comprises a film, a layer, or a substrate, and wherein the one or more predicted metrics comprise one or more dimensions or properties of the film, the layer, or the substrate, and optionally wherein the deposited or fabricated structure is etched, patterned, polished, or cleaned;” and claim 49 recites, “the system is configured to be used or deployed in a manufacturing environment.” This is generally linking the use of a judicial exception to a particular technological environment or field of use, because these elements tie the system to specific semiconductor field. See MPEP 2106.05(h).
Claim(s) 50 recites, “the plurality of data types and datasets comprise: (1) historical process data, (2) current process data, (3) historical measurement data of the one or more predicted metrics, (4) current measurement data of the one or more predicted metrics, (5) operation data, and/or (6) equipment specification data;” Claim(s) 52 recites, “(i) at least the historical process data or the historical measurement data or (ii) at least the operation data or the equipment specification data;” and Claim(s) 53 recites, “the plurality of different sources comprises (i) the process equipment or (ii) a measurement equipment configured to collect the current measurement data.” This is generally linking the use of a judicial exception to a particular technological environment or field of use (various types of data and data sources in the semiconductor field). See MPEP 2106.05(h).
Claim(s) 57 recites, “the machine learning pipeline” and “wherein the machine learning pipeline is configured to apply the training data through the two or more components sequentially or simultaneously;” Claim(s) 58 recites, “wherein the feature engineering comprises;” Claim(s) 59 recites, “wherein the time-aware data normalization comprises;” and Claim(s) 60 recites, “the adaptive learning algorithm comprises.” This is generally linking the use of a judicial exception to a particular technological environment or field of use (linking the use of the generic machine learning techniques that is used for model optimization (abstract idea) in the semiconductor processing field). See MPEP 2106.05(h).
Claim(s) 60 recites, “the adaptive learning algorithm comprises one or more adaptive online ensemble learning algorithms.” This is generally linking the use of a judicial exception to a particular technological environment or field of use (linking the use of the generic machine learning techniques that is used for model optimization (abstract idea) in the semiconductor processing field). See MPEP 2106.05(h).
Claim(s) 61 recites, “the training and optimization module is configured to optimize the model using at least in part hyperparameter optimization.” This is generally linking the use of a judicial exception to a particular technological environment or field of use (linking the use of model optimization in the semiconductor field). See MPEP 2106.05(h).
Claim(s) 62 describes, train the model; then evaluate performance of the model; and then repeat/iterate to improve the model. This is generally linking the use of a judicial exception to a particular technological environment or field of use (linking the use of iterative model optimization in the semiconductor field). See MPEP 2106.05(h).
The claim(s) recite the additional elements of,
(Claims 41 and 54-56 ) “receive”… “plurality of data types and datasets”
(Claim 41) “provide the training data to a machine learning pipeline for training and optimizing a model”
(Claim 42) “receive” “process data” “the process data is received from the process equipment substantially in real-time as the process is performed”
(Claim 52) “database or a log that is configured to store”
(Claim 56) “process data is input to the model in the inference module”
These additional elements are recited at a high level of generality and as a form of insignificant extra solution activity recognized by court as well-understood, routine, conventional activity.
The receiving, providing and inputting data are claimed in a merely generic manner (e.g., at a high level of generality) and as an insignificant extra solution activity such as mere data gathering, input and output that are recognized by the court as well-understood, routine, and conventional MPEP 2106.05(d)(ii)).
The storing of data are in a database claimed in a merely generic manner (e.g., at a high level of generality) and as an insignificant extra solution activity for data storing as computer function such that storing of data are recognized by the court as well-understood, routine, and conventional MPEP 2106.05(d)(ii)).
Accordingly, these additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. Therefore, the claims are directed to an abstract idea.
The additional elements in the claim amount to no more than insignificant extra solution activity and do not amount to significantly more than the judicial exception because, storing is mere data storing, and receiving, providing, inputting are data gathering, inputting/outputting (MPEP 2106.05(d)(ii)). As explained above, the claim limitations are implemented on these generic elements such that the following are merely applying the abstract idea on a generic computer: determining, evaluating, analyzing, calculating, selecting and mathematical calculation steps be done by at least one processor. Further, the use of the claimed invention in a semiconductor processing field is simply an attempt to limit the use of the abstract idea to a particular technological environment (MPEP 2106.05(h)). The claim does not include any further additional elements that are sufficient to amount to significantly more than the judicial exception.
Even when combined with all of the claim limitations as a whole, it is still directed to the abstract idea of mental process. Therefore, the claims are not patent eligible.
Dependent claim(s) when analyzed as a whole are held to be patent ineligible under 35 U.S.C. 101 because the additional recited limitation(s) fail(s) to establish that the claim(s) is/are not directed to an abstract idea, as they recite further embellishment of the judicial exception.
Viewed as a whole, these additional claim element(s) do not provide meaningful limitation(s) to transform the abstract idea into a patent eligible application of the abstract idea such that the claim(s) amounts to significantly more than the abstract idea itself. Claims 41-62 do not include any further additional elements that are sufficient to amount to significantly more than the judicial exception.
Therefore, the claim(s) 41-62 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Claim Interpretation
It is noted that, at this time, the claims 41-62 are not examined under the claim interpretation 35 USC § 112(f) and therefore the 35 USC §101 rejections are applied. See the 35 USC §101 rejections as applied above.
Claim Rejections - 35 USC § 112
35 U.S.C. 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 41-62 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
-Unclear limitations and insufficient antecedent basis:
Claim 41-42:
Claim 41 recites, generating one or more predicted metrics substantially in real-time.
Claim 42 recites, the process data is received from the process equipment substantially in real-time.
The term “substantially in real-time” in these claims is a relative term which renders the claim indefinite. The term “substantially in real-time” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. It’s not clear if by substantially it means the time used in the claims can deviate from the real-time such as can be closer to real-time or can be further away from real-time.
For the examination purpose, in light of the specification, in cannot be construed what it meant by substantially in real-time.
Claim 43:
Claim recites, the process control.
There is insufficient antecedent basis for this limitation in the claim, because parent claim 41 describes “process” and “process equipment;” however doesn’t recite any process control. Therefore, it isn’t clear what exact process it is refereeing to.
For the examination purpose, in light of the specification, the above described limitation is construed as, process control.
Appropriate correction is required.
Claims 42-62:
Based on their dependencies in claim 41, claims 42-62 are rejected under 35 U.S.C. 112(b) for the same reasons.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 41-44 and 47-53 is/are rejected under 35 U.S.C. 102(a)(1)/102(a)(2) as being anticipated by Ummethala et al. (US20220066411A1) [hereinafter Ummethala].
Regarding claim 41:
Ummethala discloses, A system for process monitoring and control, comprising: [¶4: system including a memory and a processing device…processing device is to provide, as input to a trained machine learning model, data associated with processing each of a first set of substrates at a manufacturing system];
a data processing module configured to receive and process a plurality of data types and datasets from a plurality of different sources for generating training data; [¶30: a training set generator 172 that is capable of generating training data sets (e.g., a set of data inputs and a set of target outputs) to train, validate, and/or test a machine learning model 190…
¶5: generating first training data for the machine learning model. The first training data includes historical data…and a first set of historical metrology measurement values…
¶16: the processing device may receive data from sensors at a processing chamber, a transfer chamber, a load lock, a factory interface,…
¶17: provide the data associated with processing the set of substrates and/or the data associated with the substrate before or after the substrates are processed as input the trained machine learning model…
Examiner notes that the limitation, data types and datasets from a plurality of different sources is broad and can be any data from any sources];
a training and optimization module configured to provide the training data to a machine learning pipeline for training and optimizing a model; and [¶5: training a machine learning model to predict a modification for a particular process recipe for a current substrate being processed at a manufacturing system… providing the first training data, the second training data, and the third training data to train the machine learning model to predict, for the particular process recipe for the current substrate being processed at the manufacturing system, which modification to the process recipe is to satisfy a drift criterion for a subsequent set of substrates that is to be processed after the current substrate.];
an inference module configured to use the model for generating one or more predicted metrics substantially in real-time, wherein the one or more predicted metrics are useable to characterize an output of a process performed by a process equipment. [¶5: to predict, for the particular process recipe for the current substrate being processed at the manufacturing system, which modification to the process recipe is to satisfy a drift criterion for a subsequent set of substrates that is to be processed after the current substrate…
¶18: The processing device may determine, from the output of the machine learning model, an amount of drift of the set of metrology measurement values for the set of substrates from a target metrology measurement value….processing device may determine that a modification to the process recipe satisfies a drift criterion in response to determining the respective modification is predicted to reduce an amount of substrate process drift…identify a respective modification having a level of confidence that satisfies a confidence criterion and update the process recipe based on the identified modification.
Examiner notes the claim objections related to the term “useable” and the 35 USC 112(b) rejections related to the term “substantially in real-time” set forth in the current office action.
Examiner notes that claim language is broad and output can be any output, and useable means it’s useable but doesn’t necessarily means it is used, and substantially in real-time is vague and can mean real time or can be near real time (may be not exactly real-time)].
Regarding claim 42:
Ummethala discloses, The system of claim 41, and further discloses, wherein the inference module is configured to receive and provide process data to the model for generating the one or more predicted metrics, wherein the process data is received from the process equipment substantially in real-time as the process is performed. [¶16: A processing device…may receive data associated with processing a set of substrates at a manufacturing system according to a process recipe. The data may be received from sensors at various stations of a manufacturing system….
¶17: The processing device may provide the data associated with processing the set of substrates and/or the data associated with the substrate before or after the substrates are processed as input the trained machine learning model and obtain one or more outputs from the machine learning model….may also provide a set of metrology measurement values (e.g., an etch rate, an etch rate uniformity, a critical dimension, a critical dimension uniformity, edge to edge placement error, etc.) for the set of substrates as input to the machine learning model.
Examiner notes the 35 USC 112(b) rejections related to the term “substantially in real-time” set forth in the current office action.
Examiner notes that claim language is broad and process data can be any process data, and substantially in real-time is vague and can mean real time or can be near real time (may be not exactly real-time)].
Regarding claim 43:
Ummethala discloses, The system of claim 41, and further discloses, wherein the inference module is configured to provide the one or more predicted metrics for the process control, or for process monitoring, improvement or trouble-shooting. [Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Ummethala discloses, provide the one or more predicted metrics for process control/improvement/troubleshooting etc.
¶18: processing device may determine that a modification to the process recipe satisfies a drift criterion in response to determining the respective modification is predicted to reduce an amount of substrate process drift for a set of substrates subsequently processed at the manufacturing system….identify a respective modification having a level of confidence that satisfies a confidence criterion and update the process recipe based on the identified modification.
Examiner notes the 35 USC 112(b) rejections set forth in the current office action.].
Regarding claim 44:
Ummethala discloses, The system of claim 41, and further discloses, further comprising a process control module configured to use the one or more predicted metrics to detect, correct, or mitigate a drift, a shift, or a deviation in the process or the process equipment. [Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Ummethala discloses, use the one or more predicted metrics to detect/correct drift/deviation.
¶18: processing device may determine that a modification to the process recipe satisfies a drift criterion in response to determining the respective modification is predicted to reduce an amount of substrate process drift for a set of substrates subsequently processed at the manufacturing system].
Regarding claim 47:
Ummethala discloses, The system of claim 41, and further discloses, further comprising the process equipment, wherein the process equipment comprises a semiconductor process equipment. [¶15: systems detecting and correcting substrate process drift using a machine learning model…
¶16: the processing device may receive data from sensors at a processing chamber, a transfer chamber, a load lock, a factory interface,…
¶21: the machine learning model may detect when substrate process drift has occurred and predict a modification that is likely to correct the drift for substrates subsequently processed at the manufacturing system….update a process recipe based on the predicted modification. By updating a process recipe based on the predicted modification, the amount of substrate process drift at a manufacturing system will decrease, causing a number of defects that occur within the manufacturing system to decrease, improving overall system efficiency…
Also see fig. 3].
Regarding claim 48:
Ummethala discloses, The system of claim 41, and further discloses, wherein the output of the process comprises a deposited or fabricated structure, wherein the deposited or fabricated structure comprises a film, a layer, or a substrate, wherein the one or more predicted metrics comprise one or more dimensions or properties of the film, the layer, or the substrate, and optionally wherein the deposited or fabricated structure is etched, patterned, polished, or cleaned.[¶15: methods and systems detecting and correcting substrate process drift using a machine learning model. Substrate process drift refers to a drift of a quality and/or consistency of substrates processed at a manufacturing system according to a particular process recipe. Substrate process drift may be detected based on a drift of metrology measurement…metrology measurement value may include…etch rate uniformity (i.e., a variation of an etch rate at two or more portions of the surface of the substrate), a critical dimension (i.e., a unit of measurement for measuring a dimension of elements of a substrate,…
¶25: metrology data may include a value of one or more of film property data (e.g., wafer spatial film properties),].
Regarding claim 49:
Ummethala discloses, The system of claim 41, and further discloses, wherein the system is configured to be used or deployed in a manufacturing environment. [¶15: methods and systems detecting and correcting substrate process drift using a machine learning model. Substrate process drift refers to a drift of a quality and/or consistency of substrates processed at a manufacturing system according to a particular process recipe.].
Regarding claim 50:
Ummethala discloses, The system of claim 41, and further discloses, wherein the plurality of data types and datasets comprise: (1) historical process data, (2) current process data, (3) historical measurement data of the one or more predicted metrics, (4) current measurement data of the one or more predicted metrics, (5) operation data, and/or (6) equipment specification data. [Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Ummethala discloses, plurality of data types and datasets comprise: (1) historical process data, (2) current process data etc.
¶16: A processing device (e.g., a system controller for the manufacturing system) may receive data associated with processing a set of substrates at a manufacturing system according to a process recipe….the processing device may receive data from sensors at a processing chamber, a transfer chamber, a load lock, a factory interface, and so forth… processing device may receive, from a substrate measurement sub-system, data associated with the substrates before or after the substrates are processed at the manufacturing system].
Regarding claim 51:
Ummethala discloses, The system of claim 50, and further discloses, wherein the data processing module is configured to validate the historical process data and the historical measurement data against the operation data and the equipment specification data. [¶46: At block 216, processing logic generates first training data based on historical data associated with the first set of substrates processed according to the first process recipe and the metrology measurement for each of the first set of substrates. At block 218, processing logic generates second training data based on historical data associated with the first set of substrates processed according to the first process recipe and the metrology measurement for each of the first set of substrates. At block 220, processing logic generates third training data including an indication of a difference between the first process recipe and the second process recipe.
Examiner notes that claim is broad and validate means it can be validated any manner because claim doesn’t provide any particular validation process, comparison requirements, threshold or outcome of the validation].
Regarding claim 52:
Ummethala discloses, The system of claim 50, and further discloses, wherein the plurality of different sources comprise a database or a log that is configured to store (i) at least the historical process data or the historical measurement data or (ii) at least the operation data or the equipment specification data. [Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Ummethala discloses, store (i) at least the historical process data or the historical measurement data or (ii) at least the operation data or the equipment specification data.
¶27: Data store 140 may be…a database system… may store data associated with processing a substrate at manufacturing equipment 124….store data collected by sensors 126 at manufacturing equipment 124 before, during, or after a substrate process (referred to as process data). Process data can refer to historical process data (e.g., process data generated for a previous substrate processed at the manufacturing system) and/or current process data (e.g., process data generated for a current substrate processed at the manufacturing system)…data store may store metrology data including historical metrology data (e.g., metrology measurement values for a prior substrate processed at the manufacturing system)].
Regarding claim 53:
Ummethala discloses, The system of claim 50, and further discloses, wherein the plurality of different sources comprises (i) the process equipment or (ii) a measurement equipment configured to collect the current measurement data.
[Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Ummethala discloses, different sources comprises (i) the process equipment or (ii) a measurement equipment configured to collect the current measurement data.
¶16: A processing device…receive data…data may be received from sensors at various stations of a manufacturing system…processing device may receive data from sensors at a processing chamber, a transfer chamber, a load lock, a factory interface, and so forth….receive, from a substrate measurement sub-system, data associated with the substrates… receive spectral data associated with a profile of each substrate after each substrate is etched at a processing chamber.].
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filling date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 45-46 is/are rejected under 35 U.S.C. 103 as being unpatentable over UMMETHALA and further in view of Baseman (US20140031969A1) [hereinafter Baseman].
Regarding claim 45:
Ummethala discloses, The system of claim 44, and further discloses, wherein the process control module is configured to use the one or more predicted metrics to improve process productivity [¶49: After block 228, machine learning model 190 may be used to predict, for a particular process recipe for a current substrate being processed at the manufacturing system, which modification to the process recipe is to satisfy a drift criterion for a subsequent set of substrates that is to be processed after the current substrate.], but doesn’t explicitly disclose, and
Baseman discloses, use the one or more predicted metrics to improve process productivity via integration with run-to-run control. [¶62: An R2R controller 530 is operative to receive measurement prediction results from virtual metrology module 522, prediction error results from prediction error module 524, and estimated metrology error and variation results from module 510….R2R controller 530 is operative in a feedback control arrangement to control one or more manufacturing tools 532…The manufacturing tools 532 are used to produce wafers and generate processing variables 534, which can be used by subsequent processing steps to adjust the manufacturing process to meet prescribed parameters.].
Therefore, it would have been obvious to one of ordinary skill in the art before the filling date of the claimed invention to have combined the capability of use the one or more predicted metrics to improve process productivity via integration with run-to-run control in order to facilitate timely process control, detect faulty wafers early in the process, and improve productivity taught by Baseman with the system taught by UMMETHALA as discussed above in order to have reasonable expectation of success such as to facilitate timely process control, detect faulty wafers early in the process, and improve productivity [Baseman, ¶21: facilitate timely process control, detect faulty wafers early in the process, and improve productivity].
Regarding claim 46:
UMMETHALA and Baseman disclose, The system of claim 41, and
Baseman further discloses, wherein the model comprises a virtual metrology (VM) model. [¶53: The predictive modeling module 408 is operative to…generate therefrom a prediction model…Any one or more of the virtual metrology models discussed above are suitable for use with the predictive modeling module 408. The predictive modeling module 408 is operative to generate an output…
¶55: the sampling and optimization module 406 is operative to determine when a confidence in the virtual metrology (i.e., predictive model) has fallen below some prescribed target value so as to require taking actual metrology measurements to correct the predictive model].
Claim(s) 54 is/are rejected under 35 U.S.C. 103 as being unpatentable over UMMETHALA and further in view of Patel et al. (US20020107599A1) [hereinafter Patel].
Regarding claim 54:
Ummethala discloses, The system of claim 41, and further discloses, wherein the data processing module is configured to receive and process the plurality of data types or datasets… [¶16: the processing device may receive data from sensors at a processing chamber, a transfer chamber, a load lock, a factory interface,], but doesn’t explicitly disclose, and
Patel discloses, …receive and process the plurality of data types or datasets by generating a component hierarchical structure of the process equipment,
wherein the component hierarchical structure comprises a nested structure of (i) the process equipment and (ii) one or more components that are used within or in conjunction with the process equipment. [¶17: Work cell 10 includes a scheduler 12, a buffer 14, a dispatcher 16, and a plurality of machines 18. Each machine 18 may have an associated controller 20. However, the controller 20 functions may be performed by the dispatcher 16. By using a controller 20 for each machine 18, the processes of the present invention may be distributed between controller 20 and dispatcher 16…
Examiner notes that, as shown in Patel fig. 1; component hierarchical structure of the process equipment; nested structure of process equipment such as buffer Lot 1-n, machine 1-n, dispatcher etc. and one or more components such as controllers 1-n etc.].
Therefore, it would have been obvious to one of ordinary skill in the art before the filling date of the claimed invention to have combined the capability of receiving and processing the plurality of data types or datasets by generating a component hierarchical structure of the process equipment, wherein the component hierarchical structure comprises a nested structure of (i) the process equipment and (ii) one or more components that are used within or in conjunction with the process equipment in order to minimize the influence of process and inventory disturbances and to optimize process performance taught by Patel with the system taught by UMMETHALA as discussed above in order to have reasonable expectation of success such as to minimize the influence of process and inventory disturbances and to optimize process performance [Patel, ¶4: minimizes the influence of process and inventory disturbances and optimizes process performance.].
Claim(s) 55 is/are rejected under 35 U.S.C. 103 as being unpatentable over UMMETHALA and further in view of Hsu et al. (US20120215337A1) [hereinafter Hsu].
Regarding claim 55:
Ummethala discloses, The system of claim 41, and further discloses, the data processing module is configured to receive and process the plurality of data types or datasets by generating a step-operation…structure of a recipe for the process,
wherein the recipe comprises a plurality of steps, and wherein each step of the plurality of steps comprises a plurality of different sub-operations. [¶28: Contextual data can include a recipe name, recipe step number,…
¶34: providing data associated with processing each of a set of substrates at a manufacturing system according to a process recipe as input to trained machine learning model 190… determining process recipe modification data from the output of the trained machine learning model 190 and using the process recipe modification data to predict a modification to process recipe…
¶75: process recipe modification data identifying one or more modifications to the process recipe and, for each of the modifications, an indication of a level of confidence that a respective modification satisfies a drift criterion for a second set of substrates.
Examiner notes that, claim limitations related to “recipe” for the processing is broad and can be any recipe. Ummethala implicitly discloses, receiving and processing data by generating modified recipe, such that one of the ordinary skilled in the art will understand that a substrate processing recipe (i.e.; taught by Ummethala) will include plurality of steps with different sub-operations (e.g.; preparation: (select, slice, polish); oxidation: (dry/wet); CVD deposition (atmospheric/low pressure/plasma enhanced etc.); coating, etching etc.), thus, the modified recipes include steps with different sub operations…
Examiner notes that, even though Ummethala teaches generation of substrate recipe where the recipe steps and sub operations are well known in the art; however, for the purpose of compact prosecution, a secondary prior art teaching the “hierarchical structure of a recipe for the process” has been provided in combination with the teachings of Ummethala], but Ummethala doesn’t explicitly disclose, and
Hsu discloses, receive and process the plurality of data types or datasets by generating a step-operation hierarchical structure of a recipe for the process, wherein the recipe comprises a plurality of steps, and
wherein each step of the plurality of steps comprises a plurality of different sub-operations. [¶15: block 25 in which data is gathered. The data may come from any semiconductor device product, for example an Integrated Circuit (IC) chip. The data gathered is with respect to a specific process or recipe, for example an etching process, an oxidation process, a diffusion process, a deposition process, a lithography process, or another applicable process. As long as the process or recipe is the same, the data may come from different products or wafers. The gathered data may also come from different processing chambers…
¶59: the advanced processing control includes adjusting the…recipes of one processing tool applicable to the relevant wafers].
Therefore, it would have been obvious to one of ordinary skill in the art before the filling date of the claimed invention to have combined the capability of receiving and processing the plurality of data types or datasets by generating a step-operation hierarchical structure of a recipe for the process, wherein the recipe comprises a plurality of steps, and wherein each step of the plurality of steps comprises a plurality of different sub-operations in order to improve the efficiency of the process taught by Hsu with the system taught by UMMETHALA as discussed above in order to have reasonable expectation of success such as to improve the efficiency of the process and reduce fabrication/production time [Hsu, ¶61: the fabrication time can be reduced].
Claim(s) 56-60 is/are rejected under 35 U.S.C. 103 as being unpatentable over UMMETHALA and further in view of Cheon et al. (US20230237412A1) [hereinafter Cheon].
Regarding claim 56:
Ummethala discloses, The system of claim 50, but doesn’t explicitly disclose, and
Cheon discloses, wherein the data processing module is configured to:
(i) receive and process the plurality of data types or datasets by removing one or more data outliers;
(ii) pre-process and remove data outliers from the process data before the process data is input to the model in the inference module;
(iii) continuously update the training data with the current process data and the current measurement data: or
(iv) any combination of (i)-(iii). [Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Cheon discloses, steps i) or ii).
¶176: Data may be preprocessed to remove superfluous data (e.g., all inlier data may be removed). In some embodiments, dimensional reduction may be applied to simplify the remaining outlier data.].
Therefore, it would have been obvious to one of ordinary skill in the art before the filling date of the claimed invention to have combined the data processing module is configured to: (i) receive and process the plurality of data types or datasets by removing one or more data outliers; (ii) pre-process and remove data outliers from the process data before the process data is input to the model in the inference module; (iii) continuously update the training data with the current process data and the current measurement data: or (iv) any combination of (i)-(iii) in order to achieve more efficient substrate manufacturing process with minimized waste allowing quick processing time and reducing analysis complexity by removing unnecessary data taught by Cheon with the system taught by UMMETHALA as discussed above in order to have reasonable expectation of success such as to improve the efficiency of the process and reduce fabrication/production time [Cheon, ¶37: allows quick processing time and a reduction in communication bandwidth, complexity of analyzing full trace data, etc… ¶]41: results in a more efficient substrate manufacturing process with minimized waste)].
Regarding claim 57:
Ummethala discloses, The system of claim 41, and further discloses, wherein the machine learning pipeline comprises two or more components from a plurality of components [¶5: training a machine learning model to predict a modification for a particular process recipe for a current substrate being processed at a manufacturing system… providing the first training data, the second training data, and the third training data to train the machine learning model to predict, for the particular process recipe for the current substrate being processed at the manufacturing system, which modification to the process recipe is to satisfy a drift criterion for a subsequent set of substrates that is to be processed after the current substrate.], but doesn’t explicitly disclose, and
Cheon discloses, wherein the machine learning pipeline comprises two or more components from a plurality of components comprising of
(i) feature engineering,
(ii) time-aware data normalization, and/or
(iii) an adaptive learning algorithm, and
[Examiner notes that claim requires only one of the optional elements separated by “or,” and only one of them is given the patentable weight.
Accordingly, Cheon discloses, i) or ii) or iii).
¶47: Processing of the sensor data 142, metrology data 160,…may include generating features….features are a pattern in the sensor data 142, metrology data 160,…sensor data 142 may include features and the features may be used by the predictive component 114 for performing signal processing and/or for obtaining predictive data 168 for performance of a corrective action…
¶]124: preprocessing may be performed on the data, e.g., smoothing, interpolation, normalization, etc. Preprocessing may be performed at various stages of analysis…
¶95: the data set is used to train, validate, or test one or more models 190 (e.g., one of the models that are included in model 190, ensemble model 190, etc.)];
wherein the machine learning pipeline is configured to apply the training data through the two or more components sequentially or simultaneously. [¶47: sensor data 142 may include features and the features may be used by the predictive component 114 for performing signal processing and/or for obtaining predictive data 168 for performance of a corrective action…
¶50: predictive system 110 may generate predictive data 168 using machine learning, such as supervised machine learning (e.g., a machine learning model may be configured to produce labels associated with input data, such as metrology predictions].
Regarding claim 58:
Ummethala and Cheon disclose, The system of claim 57, and
Cheon further discloses, wherein the feature engineering comprises
(i) an extraction of a plurality of features from raw trace data or sensor data within the training data and [¶47: sensor data 142, metrology data 160,…processed (e.g., by the client device 120 and/or by the predictive server 112). Processing of the sensor data 142, metrology data 160,…may include generating features….features are a pattern in the sensor data 142, metrology data 160, and/or manufacturing parameters 150 (e.g., slope, width, height, peak, etc.) or a combination of values from the sensor data 142, metrology data 160,];
(ii) use of an algorithm to select one or more features from a list of extracted features, based at least in part on local relationships between process data and measurement data. [¶47: features and the features may be used by the predictive component 114 for performing signal processing and/or for obtaining predictive data 168 for performance of a corrective action.…
¶67: enable a processing device to connect sensor differences (e.g., a particular pattern or feature in data from a particular sensor) to faults. Processing logic may be configured to connect certain features or patterns in sensor data to certain types of faults, certain components aging, drifting, or failing, etc.].
Regarding claim 59:
Ummethala and Cheon disclose, The system of claim 57, and
Cheon further discloses, wherein the time-aware data normalization comprises a decomposition of time series data into one or more components including smoothing data, trend data, and/or detrend data. [¶124: trace data may be stored separated into process operations, and operation-separated data may be utilized to generate summary data….preprocessing may be performed on the data, e.g., smoothing, interpolation, normalization, etc. Preprocessing may be performed at various stages of analysis.].
Regarding claim 60:
Ummethala and Cheon disclose, The system of claim 57, and
Cheon further discloses, wherein the adaptive learning algorithm comprises one or more adaptive online ensemble learning algorithms. [¶95: data set generator 372 generates a data set (e.g., training set…that includes one or more data inputs 310 (e.g., training input,…Data inputs 310 may also be referred to as “features,”…data set generator 372 may provide the data set to the training engine 182,…where the data set is used to train, validate, or test one or more models 190 (e.g., one of the models that are included in model 190, ensemble model 190, etc.).].
Claim(s) 61-62 is/are rejected under 35 U.S.C. 103 as being unpatentable over UMMETHALA and further in view of KHAVRONIN et al. (US20220188700A1) [hereinafter KHAVRONIN].
Regarding claim 61:
Ummethala discloses, The system of claim 41, but doesn’t explicitly disclose, and
KHAVRONIN discloses, wherein the training and optimization module is configured to optimize the model using at least in part hyperparameter optimization. [¶32: The manager operates a model parameter and/or hyperparameter (“(H)P”) optimization process, and at each instance or epoch of the training process, the manager directs each worker to run model training with respective sets of (H)Ps. Each of the workers trains and tests a local ML model using their respective (H)P sets, in parallel. Each worker independently provides their tested (H)P sets with calculated performance scores back to the manager, which then performs additional optimizations on the (H)P sets to produce more optimal (H)P sets. These more optimal (H)P sets are then sent to available workers to train and test their local models using the updated (H)P sets. This process continues until convergence is met.].
Therefore, it would have been obvious to one of ordinary skill in the art before the filling date of the claimed invention to have combined the capability of optimizing the model using hyperparameter optimization in order to achieve faster optimization of ML models from relatively large volumes of data with fewer evaluation instances or epochs while also producing more optimal model parameters taught by KHAVRONIN with the system taught by UMMETHALA as discussed above in order to have reasonable expectation of success such as to achieve faster optimization of ML models and producing more optimal model parameters [KHAVRONIN, ¶31: optimizes ML models from relatively large volumes of data faster than the existing optimization approaches (e.g., requiring fewer evaluation instances or epochs) while also producing more optimal model parameters than the existing optimization approaches].
Regarding claim 62:
Ummethala and KHAVRONIN disclose, The system of claim 61, and
KHAVRONIN further discloses, wherein the training and optimization module is further configured to:
(i) train the model with a set of hyperparameters on an output from the machine learning pipeline;
(ii) evaluate a performance of the model based on validation data, wherein the validation data is split from the training data for the hyperparameter optimization;
(iii) use a hyperparameter optimization algorithm to select a set of hyperparameters for a next iteration based on past performance, so as to increase or improve the performance of the model; and
(iv) repeat (i)-(iii) iteratively until the performance of the model meets termination criteria. [¶198: The model optimizer 16 a 10 generates a new/different set of (H)Ps 16 a 08 using a suitable optimization process. The model optimizer 16 a 10 optimizes the (H)Ps 16 a 08 in an iterative process until a most optimal set of (H)Ps 16 a 08 are determined…
¶32: The manager operates a model parameter and/or hyperparameter (“(H)P”) optimization process, and at each instance or epoch of the training process, the manager directs each worker to run model training with respective sets of (H)Ps…trains and tests a local ML model using their respective (H)P sets, in parallel. Each worker independently provides their tested (H)P sets with calculated performance scores back to the manager, which then performs additional optimizations on the (H)P sets to produce more optimal (H)P sets. These more optimal (H)P sets are then sent to available workers to train and test their local models using the updated (H)P sets. This process continues until convergence is met.].
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure is listed in the PTO-892 Notice of Reference Cited document.
David (US20160148850A1): Process control techniques for semiconductor manufacturing processes:
¶80-¶82: FIG. 6, in step 608, filtering, normalization and/or cleansing steps can be performed on the input data….In step 610, a dimensionality reduction or feature selection step is performed…In step 612, the data is then fed into the algorithm for training.
Banna (US20200333774A1): Adaptive chamber matching in advanced semiconductor process control:
¶7: receiving in-line metrology data…receiving historic device data…converting the sensor trace data, the in-line metrology data and the historic device data into current device processing data; providing the current device processing data as input to a trained machine learning model; obtaining, from the trained machine-learning model, one or more outputs indicative of predictive data; and, causing, based on the one or more outputs indicative of the predictive data, performance of one or more corrective actions associated with respective chambers.
.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOHAMMED SHAFAYET whose telephone number is (571)272-8239. The examiner can normally be reached M-F 8:30 AM-5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kenneth Lo can be reached at (571) 272-9774. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/M.S./
Patent Examiner,
Art Unit 2116
/CHAD G ERDMAN/Primary Examiner, Art Unit 2116
Prosecution Insights