DETAILED ACTION
Notice of Pre-AIA or 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 .
Status of Claims
This action is responsive to remarks filed 10/14/2025. Claims 1 and 9 are amended. No additional claims have been cancelled, and there are no new claims.
Claims 1, 3–9, and 11–16 are pending for examination.
Response to Arguments
In reference to 35 USC § 101
Applicant’s arguments, filed on 10/14/2025, with respect to the § 101 rejections have been fully considered but are not persuasive.
Applicant argues, beginning on Pg. 11 of the Remarks, that “obtaining metrology inspection data - including virtual measurement data that is generated through a simulation procedure - cannot be practically or conceptually performed by the human mind.” Examiner respectfully agrees. However, this limitation is not found to be a mental process. This limitation is insignificant extra-solution activity of mere data gathering and is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Applicant argues, beginning on Pg. 11 of the Remarks, that “The claimed steps … constitutes a practical application of a technical improvement in metrology data generation and process monitoring.” Examiner respectfully disagrees. See MPEP 2106.04(d)(1) and MPEP 2106.05(a) which respectively state:
if the specification explicitly sets forth an improvement but in a conclusory manner (i.e., a bare assertion of an improvement without the detail necessary to be apparent to a person of ordinary skill in the art), the examiner should not determine the claim improves technology. Second, if the specification sets forth an improvement in technology, the claim must be evaluated to ensure that the claim itself reflects the disclosed improvement. That is, the claim includes the components or steps of the invention that provide the improvement described in the specification. The claim itself does not need to explicitly recite the improvement described in the specification (e.g., "thereby increasing the bandwidth of the channel").
An indication that the claimed invention provides an improvement can include a discussion in the specification that identifies a technical problem and explains the details of an unconventional technical solution expressed in the claim, or identifies technical improvements realized by the claim over the prior art. For example, in McRO, the court relied on the specification’s explanation of how the particular rules recited in the claim enabled the automation of specific animation tasks that previously could only be performed subjectively by humans, when determining that the claims were directed to improvements in computer animation instead of an abstract idea. McRO, 837 F.3d at 1313-14, 120 USPQ2d at 1100-01. In contrast, the court in Affinity Labs of Tex. v. DirecTV, LLC relied on the specification’s failure to provide details regarding the manner in which the invention accomplished the alleged improvement when holding the claimed methods of delivering broadcast content to cellphones ineligible. 838 F.3d 1253, 1263-64, 120 USPQ2d 1201, 1207-08 (Fed. Cir. 2016).
In the instant application, Examiner contends the purported improvements are not improvements to a computer but are instead improvements directed to the abstract ideas themselves (i.e., obtaining a total prediction, classifying, and selecting) and therefore, do not integrate the abstract ideas into a practical application. Without details related to how the computer functions/technology have been improved (i.e., absent claim language and/or arguments directed to the novel technical aspects), the abstract ideas are not integrated into a practical application and the additional elements to do not amount to significantly more. See detailed analysis of the newly amended claims in § 101 below.
In reference to 35 USC § 103
Applicant’s arguments filed on 10/14/2025, with respect to the claims and the newly amended limitations have been fully considered but are not persuasive.
Applicant argues, beginning on Pg. 9 of the Remarks, that “Jung only discloses ‘production-related data, equipment-related data, and quality data’.” Examiner respectfully disagrees. Examiner notes that the BRI of metrology inspection data includes measurement data and Jung teaches that the data extraction unit extracts (i.e., obtains) quality data (i.e., metrology inspection data) because Jung’s quality data includes “data which is obtained by measuring quality of an intermediate product from the time when the manufacturing of the semiconductor starts until the present time, and may be divided into fault data and measurement data.” Jung Col. 4, lines 24–28. See § 103 below for a detailed analysis.
Thus, examiner maintains the § 103 rejections.
Applicant’s arguments filed on 10/14/2025, with respect to the newly amended claim limitations have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f):
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f). The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f). The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) because the claim limitations uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitations are:
“a total prediction unit […]” in claim 9.
“a filtering unit […]” in claim 9.
Because these claim limitations are being interpreted under 35 U.S.C. 112(f), they are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have these limitation(s) interpreted under 35 U.S.C. 112(f) applicant may: (1) amend the claim limitation(s) to avoid them being interpreted under 35 U.S.C. 112(f) (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f).
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 1, 3–9, and 11–16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1:
Step 1 — Is the claim to a process, machine, manufacture, or composition of matter?
Yes, claim 1 is directed to a method i.e., a process.
Step 2A — Prong 1 Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes, the claim recites an abstract idea.
“obtaining a total prediction result according to the first prediction result, the second prediction result and the third prediction result”
“classifying a plurality of sensing factors into a number of groups according to a correlation matrix wherein numbers of the sensing factors in two of the groups are different, more than one of the sensing factors whose relationship coefficient is greater than a predetermined threshold are classified into identical group, and the relationship coefficient of any two of the sensing factors classified into different groups is not greater than the predetermined threshold”
“selecting only one from the sensing factors in each of the groups, wherein some of the sensing factors in the groups having more than one sensing factors are not selected”
These limitations, under their broadest reasonable interpretation, cover mental processes, concepts performed in the human mind (including an observation, evaluation, judgment, opinion). See MPEP 2106.04(a)(2). In particular, with the aid of pen and paper, a human can obtain a total prediction result according to other prediction results (e.g. a voting method, see present disclosure para [0031]), classify data into groups, and select a factor from the groups.
Step 2A — Prong 2 — Does the claim recite additional elements that integrate the judicial exception into a practical application?
No, there are no additional elements that integrate the judicial exception into a practical application. The additional elements:
“obtaining a plurality of equipment recipe data of a plurality of pieces of equipment” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
“inputting the equipment recipe data into a first Neural Network model, to obtain a first prediction result” — This limitation amounts to no more than mere instructions to apply the exception and is the equivalent to mere instruction to implement the abstract idea on a computer. See MPEP 2106.05(f). Processing data with a neural network merely invokes computers or other machinery as a tool to perform an existing process.
“obtaining a plurality of equipment sensing data” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
“filtering out part of the equipment sensing data according to correlations among the equipment sensing data” — This limitation amounts to no more than mere instructions to apply the exception and is the equivalent to mere instruction to implement the abstract idea on a computer. See MPEP 2106.05(f). Filtering data merely invokes computers or other machinery as a tool to perform an existing process.
“inputting the equipment sensing data into a second Neural Network model, to obtain a second prediction result” — This limitation amounts to no more than mere instructions to apply the exception and is the equivalent to mere instruction to implement the abstract idea on a computer. See MPEP 2106.05(f). Processing data with a neural network merely invokes computers or other machinery as a tool to perform an existing process.
“obtaining a plurality of metrology inspection data, wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, and the virtual measurement data is obtained via a simulation procedure” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
“inputting the metrology inspection data into a third Neural Network model, to obtain a third prediction result” — This limitation amounts to no more than mere instructions to apply the exception and is the equivalent to mere instruction to implement the abstract idea on a computer. See MPEP 2106.05(f). Processing data with a neural network merely invokes computers or other machinery as a tool to perform an existing process.
Step 2B — Does the claim recite additional elements that amount to significantly more than the judicial exception?
No, there are no additional elements that amount to significantly more than the judicial exception.
“obtaining a plurality of equipment recipe data of a plurality of pieces of equipment” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
“obtaining a plurality of equipment sensing data” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
“obtaining a plurality of metrology inspection data, wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, and the virtual measurement data is obtained via a simulation procedure” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Regarding claim 3:
The claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim is dependent on claim 1 which included an abstract idea (see rejection for claim 1 above). This claim merely recites a further limitation on the obtaining a plurality of equipment recipe data limitation which is directed to insignificant extra-solution activity of mere data gathering. The additional limitation:
“wherein each of the equipment recipe data is discrete numerical data” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d)I.), failing Step 2A Prong 2.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B.
“wherein each of the equipment recipe data is discrete numerical data” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Thus, the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (see MPEP 2106.05(I.), failing step 2B.
Regarding claim 4:
The claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim is dependent on claim 1 which included an abstract idea (see rejection for claim 1 above). This claim merely recites a further limitation on the obtain a plurality of equipment sensing data limitation which is directed to insignificant extra-solution activity of mere data gathering. The additional limitation:
“wherein each of the equipment sensing data is continuous numerical data” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d)I.), failing Step 2A Prong 2.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B.
“wherein each of the equipment sensing data is continuous numerical data” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Thus, the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (see MPEP 2106.05(I.), failing step 2B.
Regarding claim 5:
The claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim is dependent on claim 1 which included an abstract idea (see rejection for claim 1 above). This claim merely recites a further limitation on the obtaining a plurality of metrology inspection data limitation which is directed to insignificant extra-solution activity of mere data gathering. The additional limitation:
“wherein each of the metrology inspection data is discrete numerical data” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d)I.), failing Step 2A Prong 2.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B.
“wherein each of the metrology inspection data is discrete numerical data” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Thus, the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (see MPEP 2106.05(I.), failing step 2B.
Regarding claim 6:
The claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim is dependent on claim 1 which included an abstract idea (see rejection for claim 1 above). This claim merely recites a further limitation on the obtaining a plurality of equipment recipe data of a plurality of pieces of equipment limitation which is directed to mere data gathering. The additional limitation:
“wherein a plurality of processes executed by the pieces of equipment are continuously executed” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d)I.), failing Step 2A Prong 2.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B.
“wherein a plurality of processes executed by the pieces of equipment are continuously executed” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Thus, the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (see MPEP 2106.05(I.), failing step 2B.
Regarding claim 7:
The claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim is dependent on claim 1 which included an abstract idea (see rejection for claim 1 above). This claim merely recites a further limitation on the obtaining a plurality of metrology inspection data limitation which is directed to insignificant extra-solution activity of mere data gathering. The additional limitation:
“wherein the metrology inspection data includes a plurality of actual measurement data and a plurality of virtual measurement data, and the virtual measurement data is obtained by performing a simulation procedure according to the actual measurement data and the equipment sensing data” — This limitation is insignificant extra-solution activity and is merely data gathering. See MPEP 2106.05(g).
Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d)I.), failing Step 2A Prong 2.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B.
“wherein the metrology inspection data includes a plurality of actual measurement data and a plurality of virtual measurement data, and the virtual measurement data is obtained by performing a simulation procedure according to the actual measurement data and the equipment sensing data” — This limitation is directed to the activity of data gathering which is not an inventive concept because it is insignificant extra-solution activity of mere data outputting. See Mayo, 566 U.S. at 79, 101 USPQ2d at 1968; OIP Techs., Inc. v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1092-93 (Fed. Cir. 2015); MPEP 2106.05(g)(3). This limitation is well-understood, routine, and conventional because it involves transmitting information over a network. MPEP 2106.05(d)(II).
Thus, the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (see MPEP 2106.05(I.), failing step 2B.
Regarding Claim 8:
Step 1 — Is the claim to a process, machine, manufacture, or composition of matter?
Yes, claim 8 depends from claim 1 (see analysis of claim 1 above) which is directed to an method i.e., a process.
Step 2A — Prong 1 Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes, the claim recites an abstract idea.
“wherein in the step of obtaining the total prediction result according to the first prediction result, the second prediction result and the third prediction result, the total prediction result is obtained through a voting procedure”
These limitations, under their broadest reasonable interpretation, cover mental processes, concepts performed in the human mind (including an observation, evaluation, judgment, opinion). See MPEP 2106.04(a)(2). In particular, with the aid of pen and paper, a human can obtain a total prediction result according to other prediction results through a voting procedure.
Step 2A — Prong 2 — Does the claim recite additional elements that integrate the judicial exception into a practical application?
No, there are no additional elements that integrate the judicial exception into a practical application.
Step 2B — Does the claim recite additional elements that amount to significantly more than the judicial exception?
No, there are no additional elements that amount to significantly more than the judicial exception.
Regarding Claim 9:
Step 1 — Is the claim to a process, machine, manufacture, or composition of matter?
Yes, claim 1 is directed to an apparatus i.e., a machine.
Step 2A — Prong 1 Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes, the claim recites an abstract idea.
“a total prediction unit, configured to obtain a plurality of total prediction result according to the first prediction result, the second prediction result and the third prediction result”
“classify a plurality of sensing factors into a number of groups according to a correlation matrix”
“select one from the sensing factors in each of the groups, wherein numbers of the sensing factors in two of the groups are different, more than one of the sensing factors whose relationship coefficient is greater than a predetermined threshold are classified into identical group, the relationship coefficient of any two of the sensing factors classified into different groups is not greater than the predetermined threshold, and some of the sensing factors in the groups having more than one sensing factors are not selected”
These limitations, under their broadest reasonable interpretation, cover mental processes, concepts performed in the human mind (including an observation, evaluation, judgment, opinion). See MPEP 2106.04(a)(2). In particular, with the aid of pen and paper, a human can obtain a total prediction result according to other prediction results (e.g. a voting method, see present disclosure para [0031]), classify data into groups, and select a factor from the groups. The BRI of total prediction unit is code, instructions, or software capable of being executed by a processor or circuitry (see present disclosure para [0015]).
Step 2A — Prong 2 — Does the claim recite additional elements that integrate the judicial exception into a practical application?
No, there are no additional elements that integrate the judicial exception into a practical application. The additional elements:
“a first database, configured to storing a plurality of equipment recipe data of a plurality of pieces of equipment” — This limitation is reciting generic computer components at a high-level of generality (i.e., as a generic computer component performing a generic computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component. See MPEP 2106.05(f).
“a first Neural Network model, configured to receive the equipment recipe data to obtain a first prediction result” — This limitation is reciting generic computer components at a high-level of generality (i.e., as a generic computer component performing a generic computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component. See MPEP 2106.05(f).
“a second database, configured to storing a plurality of equipment sensing data” — This limitation is reciting generic computer components at a high-level of generality (i.e., as a generic computer component performing a generic computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component. See MPEP 2106.05(f).
“a filtering unit, configured to filter out part of the equipment sensing data according to correlations among the equipment sensing data” — This limitation amounts to no more than mere instructions to apply the exception and is the equivalent to mere instruction to implement the abstract idea on a computer. See MPEP 2106.05(f). Filtering data merely invokes computers or other machinery as a tool to perform an existing process.
“a second Neural Network model, configured to receive the equipment sensing data to obtain a second prediction result” — This limitation is reciting generic computer components at a high-level of generality (i.e., as a generic computer component performing a generic computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component. See MPEP 2106.05(f).
“a third database, configured to storing metrology inspection data including a plurality of actual measurement data and a plurality of virtual measurement data, wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, and the virtual measurement data is obtained via a simulation procedure” — This limitation is reciting generic computer components at a high-level of generality (i.e., as a generic computer component performing a generic computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component. See MPEP 2106.05(f).
“a third Neural Network model, configured to receive the metrology inspection data to obtain a third prediction result” — This limitation is reciting generic computer components at a high-level of generality (i.e., as a generic computer component performing a generic computer function) such that it amounts no more than mere instructions to apply the exception using a generic computer component. See MPEP 2106.05(f).
Step 2B — Does the claim recite additional elements that amount to significantly more than the judicial exception?
No, there are no additional elements that amount to significantly more than the judicial exception.
Regarding claims 11–16, although varying in scope, the limitations of claims 11–16 are substantially the same as the limitations of claims 1 and 3–8, respectively. Thus, claims 11–16 are rejected using the same reasoning and analysis as claims 1 and 3–8 above, respectively.
Claim Rejections - 35 USC § 103
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 filing 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:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 3–5, 7, 9, 11–13, and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Tsutsui et al., (US 20210166121 A1), hereinafter “Tsutsui” in view of in view of Husain et al., (US-20150262060-A1), hereinafter “Husain”, and in view of Jung et al., (US 11494636 B2), hereinafter “Jung”, and further in view of Doh Seung Yong et al., (KR-20090001148-A), hereinafter “Yong”.
Regarding claim 1, Tsutsui teaches:
obtaining a plurality of equipment recipe data of a plurality of pieces of equipment (Tsutsui Fig. 1, ¶0087: “Alternatively, the time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe I, as the time series data set 1. The time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe II, as the time series data set 2. Further, the time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe III, as the time series data set 3”; see also Tsutsui Figs. 2A–B, ¶0059: “Meanwhile, FIG. 2B illustrates a case in which a process performed in a single chamber (in the example of FIG. 2B, the “chamber B”) is defined as a unit of process 120. In this case, a wafer before processing 110 refers to a wafer that has been processed in the chamber A and that is to be processed in the chamber B, and a wafer after processing 130 refers to a wafer that has been processed in the chamber B and is to be processed in the chamber C”—[wherein the time series data sets 1, 2, and 3 are measured in chamber A, B, and C according to recipes)]);
inputting the equipment recipe data into a first Neural Network model, to obtain a first prediction result (Tsutsui Fig. 7, ¶¶0090–0091: “The branch section 710 controls input to the network sections of the first network section 720_1 to the M-th network section 720_M, so that the time series data sets and the device state information are processed by the network sections of the first network section 720_1 to the M-th network section 720_M. The first to M-th network sections (720_1 to 720_M) are configured based on a convolutional neural network (CNN), which include multiple layers”; see also Tsutsui Fig. 7, ¶0096: “By performing the machine learning, model parameters of each of the first to M-th network sections 720_1 to 720_M and the model parameters of the concatenation section 730 are optimized to predict device state information for adjustment of processes used in the manufacture of a processed substrate”—[wherein the time series data (i.e., equipment recipe data) is input into the first network section 720_1 (i.e., first neural network) to predict device state (i.e., first prediction result)]).
Tsutsui does not appear to explicitly teach:
obtaining a plurality of equipment sensing data;
filtering out part of the equipment sensing data according to correlations among the equipment sensing data, wherein the step of filtering out part of the equipment sensing data includes: classifying a plurality of sensing factors into a number of groups according to a correlation matrix,
wherein numbers of the sensing factors in two of the groups are different, more than one of the sensing factors whose relationship coefficient is greater than a predetermined threshold are classified into identical group, and the relationship coefficient of any two of the sensing factors classified into different groups is not greater than the predetermined threshold;
selecting only one from the sensing factors in each of the groups,
wherein some of the sensing factors in the groups having more than one sensing factors are not selected;
inputting the equipment sensing data into a second Neural Network model, to obtain a second prediction result;
obtaining a plurality of metrology inspection data, wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data;
inputting the metrology inspection data into a third Neural Network model, to obtain a third prediction result; and
obtaining a total prediction result according to the first prediction result, the second prediction result and the third prediction result
However, Husain teaches:
filtering out part of the equipment sensing data according to correlations among the equipment sensing data, wherein the step of filtering out part of the equipment sensing data includes: classifying a plurality of sensing factors into a number of groups according to a correlation matrix (Husain Figs. 1, 3, ¶0014: “After extracting the features, in step 120, correlation coefficients are calculated for each column by determining a moving average of the vibration signals and time. One such correlation coefficient could be a Spearman's rank correlation, which can be used to assess how strong a monotonic relationship is between a vibration signal and time. The plurality of columns may therefore be ranked according to the strength of this relationship”—[wherein the features are extracted (i.e., classified from sensor data) into columns and ranked (i.e., filtered into a number of groups) according to the correlation matrix (e.g., Spearman’s rank correlation)]); and
wherein numbers of the sensing factors in two of the groups are different, more than one of the sensing factors whose relationship coefficient is greater than a predetermined threshold are classified into identical group, and the relationship coefficient of any two of the sensing factors classified into different groups is not greater than the predetermined threshold (Husain Figs. 2–3, ¶0027: “FIG. 3 illustrates a high-level block diagram of a process for training an artificial neural network according to an exemplary embodiment of the present invention. In step 210, a training set may be formed from a set of run-to-failure bearing data. After forming the training set, in step 220 a correlation analysis can be performed to determine which of the measurements may be most correlated with the bearing degradation. After the correlation analysis, in step 230, for each condition monitoring measurement in a failure history, the method first uses the Generalized Weibull-FR function to fit the measurement series, and then uses the fitted measurement values to train the ANN. The ANN may be validated by using the data from all bearings, minus the bearing data that led to the original training of the ANN, as input to the trained ANN and comparing the output to the expected output. In step 240, the ANN attempts to model noise factors, which affect the prediction accuracy and generalization capability. In step 250, the mean square errors are calculated from the training and validation set as more neurons and more measurements are added. The mean square error can drop early on in the training process because the ANN is learning the relationship between the inputs and the outputs by modifying the trainable weights based on the training set. After a certain point, the mean square error for the validation set will start to increase, because the ANN starts to model the noise in the training set. After the mean square error starts to increase, in step 260, the training process can be stopped, and a trained ANN with good modeling and generalization capability is produced”—[(emphasis added) wherein correlation analysis is performed to determine which of the measurements may be most correlated with the bearing degradation (i.e., more than one relationship coefficient greater than the threshold are grouped together). Examiner notes that Tsutsui teaches predetermined thresholds as well and that a person having ordinary skill in the art before the filing of the instant application would have known that correlations can be grouped based on thresholds.]);
selecting only one from the sensing factors in each of the groups (Husain ¶0015: “After calculating correlation coefficients and ranking the columns, in step 130, the ranked columns are inputted into a series of artificial neural networks (ANN) such that for N columns, we create N ANNs. The first ANN is trained only with the first, most highly ranked column. The second is trained with the first and the second columns. And so on. In one embodiment, the ANNs take as input the ratio of run time to subsequent measurements and multiple condition monitoring measurements at the current and previous inspection points”; see also Husain ¶0026: “Because it can be difficult to determine the true inherent health condition of a piece of equipment based on condition monitoring measurements, in one embodiment the method utilizes a measure that has a monotonic mapping relationship with the true inherent health condition. It can be appreciated that the mapping between the inherent health condition and the life percentage is monotonically non-decreasing. It can also be appreciated that the life percentage is able to indicate when the failure occurs, that is, failure occurs when the life percentage reaches 100%”; see also Husain ¶0027: “In step 210, a training set may be formed from a set of run-to-failure bearing data. After forming the training set, in step 220 a correlation analysis can be performed to determine which of the measurements may be most correlated with the bearing degradation. After the correlation analysis, in step 230, for each condition monitoring measurement in a failure history, the method first uses the Generalized Weibull-FR function to fit the measurement series, and then uses the fitted measurement values to train the ANN”—[wherein the BRI of selecting is any action taken to choose the most suitable choice, and wherein a group can be fitted to map the monotonic relationship (i.e., selecting one relationship) from the group]),
wherein some of the sensing factors in the groups having more than one sensing factors are not selected (Husain ¶0027: “In step 210, a training set may be formed from a set of run-to-failure bearing data. After forming the training set, in step 220 a correlation analysis can be performed to determine which of the measurements may be most correlated with the bearing degradation. After the correlation analysis, in step 230, for each condition monitoring measurement in a failure history, the method first uses the Generalized Weibull-FR function to fit the measurement series, and then uses the fitted measurement values to train the ANN”—[wherein the measurements that are not most correlated are not chosen]);
The system of Tsutsui, the teachings of Husain, and the instant application are analogous art because they pertain to making predictions about semiconductor processing using neural networks.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the system of Tsutsui with the teachings of Husain to provide sensor data filtering by separating the data into groups according to a correlations and then selecting them to be used for the prediction. One would be motivated to do so to determine the relationship to the features and the equipment to better understand the system (Husain ¶0014: “After extracting the features, in step 120, correlation coefficients are calculated for each column by determining a moving average of the vibration signals and time. One such correlation coefficient could be a Spearman's rank correlation, which can be used to assess how strong a monotonic relationship is between a vibration signal and time. The plurality of columns may therefore be ranked according to the strength of this relationship”).
Tsutsui in view of Husain does not appear to explicitly teach:
obtaining a plurality of equipment sensing data;
inputting the equipment sensing data into a second Neural Network model, to obtain a second prediction result;
obtaining a plurality of metrology inspection data including a plurality of actual measurement data and a plurality of virtual measurement data, wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, and the virtual measurement data is obtained via a simulation procedure;
inputting the metrology inspection data into a third Neural Network model, to obtain a third prediction result; and
obtaining a total prediction result according to the first prediction result, the second prediction result and the third prediction result.
However, Jung teaches:
obtaining a plurality of equipment sensing data (Jung Col. 4, lines 5–28: “Data that is extracted, accumulated, and processed by the data extraction unit 110, and then is learned by the learning unit 120 will be described. The production-related data refers to historical data regarding overall production, in which data regarding a process ID, an equipment ID, a work space in equipment, an equipment work specification ID, etc. performed from a time when the manufacturing of a semiconductor starts until a present time are arranged in time series. The equipment-related data refers to data which is measured through an internal/external sensor in equipment during processes from the time when the manufacturing of the semiconductor starts until the present time, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like. That is, the equipment-related data may be referred to as input data regarding operating/state of equipment in which various physical/electrical state values regarding equipment are arranged in time series”—[wherein the data extraction unit obtains the equipment sensing data internal/external sensors (i.e., sensed equipment data)]);
inputting the equipment sensing data into a second Neural Network model, to obtain a second prediction result (Jung Col. 3, lines 42–46: “The data extracted by the data extraction unit 110 may be divided into 1) production-related data, 2) equipment-related data, and 3) quality data. The 3) quality-related data may be divided into 31) fault data and 32) measurement data. These types of data will be described below in detail”; see also Jung Col. 3, lines 51–63: “The learning unit 120 repeats a procedure of generating neural network models by performing machine learning with respect to the data extracted/processed by the data extraction unit 110, and of updating the neural network models according to input of additional data. The neural network models generated/updated by the learning unit 120 are stored in the storage unit 130. Since the neural network models are generated/updated differently according to types of data as described above, the storage unit 130 may include a production data neural network model 131, an equipment data neural network model 132, a fault data neural network model 133, and a measurement data neural network model 134”; see also Jung Col. 5, lines 52–62: “The learning unit 120 may train the equipment data neural network model 132 by inputting, as a cause value (X), time-series operating/state data of equipment measured through an internal/external sensor of the equipment, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like, to an input node of the equipment data neural network model 132, and by positioning, as an effect value (Y), a real semiconductor manufacturing yield value at an output node of the equipment data neural network model 1132”—[wherein the different types of data (e.g., equipment related data; i.e., equipment sensing data) are processed (i.e., input) by the different neural network models according to the data type (e.g., equipment data neural network model; i.e., second neural network) to predict the effect value Y]);
obtaining a plurality of metrology inspection data [including a plurality of actual measurement data and a plurality of virtual measurement data] (Jung Col. 3, lines 42–46: “The data extracted by the data extraction unit 110 may be divided into 1) production-related data, 2) equipment-related data, and 3) quality data. The 3) quality-related data may be divided into 31) fault data and 32) measurement data. These types of data will be described below in detail”—[wherein the BRI of metrology inspection data includes measurement data and wherein the data extraction unit extracts (i.e., obtains) quality data (i.e., metrology inspection data)]),
wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, [and the virtual measurement data is obtained via a simulation procedure] (Jung Col. 4, lines 14–23: “The equipment-related data refers to data which is measured through an internal/external sensor in equipment during processes from the time when the manufacturing of the semiconductor starts until the present time, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like. That is, the equipment-related data may be referred to as input data regarding operating/state of equipment in which various physical/electrical state values regarding equipment are arranged in time series”; see also Jung Col. 4, lines 24–28: “The quality data refers to data which is obtained by measuring quality of an intermediate product from the time when the manufacturing of the semiconductor starts until the present time, and may be divided into fault data and measurement data”—[wherein the BRI of heterogeneous data is a dataset composed of different data types, structures, formats, or sources, and wherein recipe data, equipment related data, and quality data (i.e., recipe data, equipment sensing data, and metrology data) each include different data types, structures, formats, or sources (e.g., according to recipe I, II, III, temperature, pressure, ratio of materials, fault data and measurement data)]);
inputting the metrology inspection data into a third Neural Network model, to obtain a third prediction result (Jung Col. 3, lines 42–46: “The data extracted by the data extraction unit 110 may be divided into 1) production-related data, 2) equipment-related data, and 3) quality data. The 3) quality-related data may be divided into 31) fault data and 32) measurement data. These types of data will be described below in detail”; see also Jung Col. 3, lines 51–63: “The learning unit 120 repeats a procedure of generating neural network models by performing machine learning with respect to the data extracted/processed by the data extraction unit 110, and of updating the neural network models according to input of additional data. The neural network models generated/updated by the learning unit 120 are stored in the storage unit 130. Since the neural network models are generated/updated differently according to types of data as described above, the storage unit 130 may include a production data neural network model 131, an equipment data neural network model 132, a fault data neural network model 133, and a measurement data neural network model 134”; see also Jung Col. 6, lines 17–25: “The learning unit 120 may train the measurement data neural network model 134 by inputting, as a cause value (X), imaged or time-series measurement data indicating a line width, thickness, and misalignment of a circuit to an input node of the measurement data neural network model 134, and by positioning, as an effect value (Y), a real semiconductor manufacturing yield value at an output node of the measurement data neural network model 134”—[wherein the different types of data (e.g., equipment related data; i.e., equipment sensing data) are processed (i.e., input) by the different neural network models according to the data type (e.g., measurement data neural network model; i.e., third neural network) to predict the effect value Y]); and
obtaining a total prediction result according to the first prediction result, the second prediction result and the third prediction result (Jung Col. 3, lines 64–67 – Col. 4, lines 1–3: “The prediction unit 140 predicts a semiconductor manufacturing yield by using the neural network models 131, 132, 133, 134 stored in the storage unit 130. The semiconductor manufacturing yield is predicted when there is a request for prediction of a yield with data from external systems 11, 12, 13, and a result of prediction may be returned to a corresponding external system”—[wherein the prediction unit predicts the yield the outputs from each model]).
The system of Tsutsui in view of Husain, the teachings of Jung, and the instant application are analogous art because they pertain to making predictions about semiconductor processing using neural networks.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the system of Tsutsui in view of Husain with the teachings of Jung to provide sensor data from the equipment to the prediction model for semiconductor processing. One would be motivated to do so to predict results that are appropriate for characteristics of the data and to provide the ability to adjust the flow of processes to improve productivity and yield (Jung Col. 2, lines 44–61: “According to embodiments of the present disclosure as described above, it is possible to exactly predict a result value by ensuring a neural network model of a structure appropriate for characteristics of data by applying different neural network models according to types of data. In addition, when embodiments of the present disclosure are applied to prediction of a semiconductor manufacturing yield, the flow of processes can be adjusted to enhance productivity/yield, and a fault can be detected early and thus it is possible to take appropriate response in advance. In addition, when embodiments of the present disclosure are applied to prediction of a semiconductor manufacturing yield, a product that is predicted as not having a good yield can be specially observed/managed, and the yield prediction function can be integrated into a platform and can be used for common use in various manufacturing processes. In addition, various feedback results may be combined and analyzed and thus can be utilized in various fields”).
Tsutsui in view of Husain and Jung does not appear to explicitly teach:
[obtaining a plurality of metrology inspection data] including a plurality of actual measurement data and a plurality of virtual measurement data; and
[wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data,] and the virtual measurement data is obtained via a simulation procedure.
However, Yong teaches:
[obtaining a plurality of metrology inspection data] including a plurality of actual measurement data and a plurality of virtual measurement data (Yong Pg. 4: “The virtual measurement method and the reliability estimation method according to the present invention for achieving the above object generates a virtual measurement model using the sensor data of the process equipment and the measured value of the measurement equipment, the reliability corresponding to the virtual measurement model A model is generated, the sensor data at the time of performing the process is substituted into the virtual measurement model to predict the measurement value, and the reliability of the measurement value is estimated”—[(emphasis added)]); and
[wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data,] and the virtual measurement data is obtained via a simulation procedure (Yong Pg. 4: “In addition, the estimation of the reliability may be performed by loading the generated reliability model to estimate the reliability of the predicted measured value. If the estimated reliability is greater than or equal to the reference value, the actual measurement is omitted, and if the estimated reliability is less than the reference value, Request measurement”—[(emphasis added)]).
The system of Tsutsui in view of Husain and Jung, the teachings of Yong, and the instant application are analogous art because they pertain to making predictions about semiconductor processing using neural networks.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the system of Tsutsui in view of Husain and Jung with the teachings of Yong to provide sensor data from the equipment and from a simulation when making predictions with a model for semiconductor processing. One would be motivated to do so to strengthen the measurement system used to detect abnormalities in the manufacturing process in order to reduce cost and process cycles, improving manufacturing yield (Yong Pg. 2: “the object of the present invention is to predict the measurement information for all the wafers in real time during the process to reduce the overall process cost and shorten the process cycle, defective wafer It is to provide a virtual measurement system and a virtual measurement method that can improve manufacturing yield through prior discovery and action”).
Regarding claim 3 Tsutsui in view of Husain, Jung, and Yong teaches all the limitations of claim 1.
Tsutsui teaches:
wherein each of the equipment recipe data is discrete numerical data (Tsutsui Fig. 1, ¶0087: “Alternatively, the time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe I, as the time series data set 1. The time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe II, as the time series data set 2. Further, the time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe III, as the time series data set 3”; see also Tsutsui ¶0072: “The GPU 404 is an arithmetic operation processing device for image processing. When the CPU 401 executes the predicting program, the GPU 404 performs high-speed calculation of various image data (i.e., the time series data sets in the present embodiment) by using parallel processing. The GPU 404 includes an internal memory (GPU memory) to temporarily retain information needed to perform parallel processing of the various image data”—[wherein the time series data associated with the recipe is image data (i.e., discrete numerical data)]).
Regarding claim 4 Tsutsui in view of Husain, Jung, and Yong teaches all the limitations of claim 1.
Jung teaches:
wherein each of the equipment sensing data is continuous numerical data (Jung Col. 5, lines 52–62: “The learning unit 120 may train the equipment data neural network model 132 by inputting, as a cause value (X), time-series operating/state data of equipment measured through an internal/external sensor of the equipment, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like, to an input node of the equipment data neural network model 132, and by positioning, as an effect value (Y), a real semiconductor manufacturing yield value at an output node of the equipment data neural network model 1132”—[wherein the equipment data is temperature data (i.e. continuous numerical data)]).
The same motivation that was utilized for combining Tsutsui in view of Husain with Jung, as set forth in claim 1, is equally applicable to claim 4.
Regarding claim 5 Tsutsui in view of Husain, Jung, and Yong teaches all the limitations of claim 1.
Jung teaches:
wherein each of the metrology inspection data is discrete numerical data (Jung Col. 4, lines 24–28: “The quality data refers to data which is obtained by measuring quality of an intermediate product from the time when the manufacturing of the semiconductor starts until the present time, and may be divided into fault data and measurement data. The fault data refers to data indicating the number of faults, a fault size, a fault position, a fault shape, or the like. The fault data may be image data or may use text data. The measurement data refers to data indicating a line width, thickness, and misalignment of a circuit. Like the fault data, the measurement data may use text data as well as image data”—[wherein the quality data (i.e., metrology inspection data) includes discrete numerical measurement data (e.g., line width and thickness)]).
The same motivation that was utilized for combining Tsutsui in view of Husain with Jung, as set forth in claim 1, is equally applicable to claim 5.
Regarding claim 7 Tsutsui in view of Husain, Jung, and Yong teaches all the limitations of claim 1.
Yong teaches:
wherein the metrology inspection data includes a plurality of actual measurement data and a plurality of virtual measurement data, and the virtual measurement data is obtained by performing a simulation procedure according to the actual measurement data and the equipment sensing data (Yong Pg. 4: “The virtual measurement method and the reliability estimation method according to the present invention for achieving the above object generates a virtual measurement model using the sensor data of the process equipment and the measured value of the measurement equipment, the reliability corresponding to the virtual measurement model A model is generated, the sensor data at the time of performing the process is substituted into the virtual measurement model to predict the measurement value, and the reliability of the measurement value is estimated … In addition, the estimation of the reliability may be performed by loading the generated reliability model to estimate the reliability of the predicted measured value. If the estimated reliability is greater than or equal to the reference value, the actual measurement is omitted, and if the estimated reliability is less than the reference value, Request measurement”—[(emphasis added)]).
The same motivation that was utilized for combining Tsutsui in view of Husain and Jung with Yong, as set forth in claim 1, is equally applicable to claim 7.
Regarding claim 9, Tsutsui teaches:
a first database, configured to store a plurality of equipment recipe data of a plurality of pieces of equipment (Tsutsui Fig. 1, ¶0035: “When a wafer before processing 110 is processed at the unit of process 120, device state information is acquired, and the device state information is stored, as training data (input data), in the training data storage unit 163 of the predicting device 160, in association with the time series data sets”; see also Tsutsui Fig. 1, ¶0087: “Alternatively, the time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe I, as the time series data set 1. The time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe II, as the time series data set 2. Further, the time series data acquiring devices 140_1 to 140_n may acquire time series data measured during processing in accordance with the process recipe III, as the time series data set 3”; see also Tsutsui Figs. 2A–B, ¶0059: “Meanwhile, FIG. 2B illustrates a case in which a process performed in a single chamber (in the example of FIG. 2B, the “chamber B”) is defined as a unit of process 120. In this case, a wafer before processing 110 refers to a wafer that has been processed in the chamber A and that is to be processed in the chamber B, and a wafer after processing 130 refers to a wafer that has been processed in the chamber B and is to be processed in the chamber C”—[wherein the data is stored in the training data storage unit 163 (i.e., a first database) and wherein the time series data sets 1, 2, and 3 are measured in chamber A, B, and C according to recipes)]);
a first Neural Network model, configured to receive the equipment recipe data to obtain a first prediction result (Tsutsui Fig. 7, ¶¶0090–0091: “The branch section 710 controls input to the network sections of the first network section 720_1 to the M-th network section 720_M, so that the time series data sets and the device state information are processed by the network sections of the first network section 720_1 to the M-th network section 720_M. The first to M-th network sections (720_1 to 720_M) are configured based on a convolutional neural network (CNN), which include multiple layers”; see also Tsutsui Fig. 7, ¶0096: “By performing the machine learning, model parameters of each of the first to M-th network sections 720_1 to 720_M and the model parameters of the concatenation section 730 are optimized to predict device state information for adjustment of processes used in the manufacture of a processed substrate”—[wherein the time series data (i.e., equipment recipe data) is input into the first network section 720_1 (i.e., first neural network) to predict device state (i.e., first prediction result)]);
a second database, configured to store (Tsutsui ¶0068: “The predicting device 160 further includes an auxiliary storage device 405, a display device 406, an operating device 407, an interface (I/F) device 408, and a drive device 409. Each hardware element in the predicting device 160 is connected to each other via a bus 410”—[wherein the auxiliary storage device 405 is configured to store data (i.e., a second database)]);
Tsutsui does not appear to explicitly teach:
a plurality of equipment sensing data;
a filtering unit, configured to filter out part of the equipment sensing data according to correlations among the equipment sensing data, wherein the filtering unit is used to classify a plurality of sensing factors into a number of groups according to a correlation matrix;
select only one from the sensing factors in each of the groups;
wherein numbers of the sensing factors in two of the groups are different, more than one of the sensing factors whose relationship coefficient is greater than a predetermined threshold are classified into identical group, and the relationship coefficient of any two of the sensing factors classified into different groups is not greater than the predetermined threshold;
some of the sensing factors in the groups having more than one sensing factors are not selected;
a second Neural Network model, configured to receive the equipment sensing data to obtain a second prediction result;
a third database, configured to store metrology inspection data including a plurality of actual measurement data and a plurality of virtual measurement data,
wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, and the virtual measurement data is obtained via a simulation procedure;
a third Neural Network model, configured to receive the metrology inspection data to obtain a third prediction result; and
a total prediction unit, configured to obtain a plurality of total prediction results according to the first prediction result, the second prediction result and the third prediction result.
However, Husain teaches:
filtering out part of the equipment sensing data according to correlations among the equipment sensing data, wherein the step of filtering out part of the equipment sensing data includes: classifying a plurality of sensing factors into a number of groups according to a correlation matrix (Husain Figs. 1, 3, ¶0014: “After extracting the features, in step 120, correlation coefficients are calculated for each column by determining a moving average of the vibration signals and time. One such correlation coefficient could be a Spearman's rank correlation, which can be used to assess how strong a monotonic relationship is between a vibration signal and time. The plurality of columns may therefore be ranked according to the strength of this relationship”—[wherein the features are extracted (i.e., classified from sensor data) into columns and ranked (i.e., filtered into a number of groups) according to the correlation matrix (e.g., Spearman’s rank correlation)]); and
selecting only one from the sensing factors in each of the groups (Husain ¶0015: “After calculating correlation coefficients and ranking the columns, in step 130, the ranked columns are inputted into a series of artificial neural networks (ANN) such that for N columns, we create N ANNs. The first ANN is trained only with the first, most highly ranked column. The second is trained with the first and the second columns. And so on. In one embodiment, the ANNs take as input the ratio of run time to subsequent measurements and multiple condition monitoring measurements at the current and previous inspection points”; see also Husain ¶0026: “Because it can be difficult to determine the true inherent health condition of a piece of equipment based on condition monitoring measurements, in one embodiment the method utilizes a measure that has a monotonic mapping relationship with the true inherent health condition. It can be appreciated that the mapping between the inherent health condition and the life percentage is monotonically non-decreasing. It can also be appreciated that the life percentage is able to indicate when the failure occurs, that is, failure occurs when the life percentage reaches 100%”; see also Husain ¶0027: “In step 210, a training set may be formed from a set of run-to-failure bearing data. After forming the training set, in step 220 a correlation analysis can be performed to determine which of the measurements may be most correlated with the bearing degradation. After the correlation analysis, in step 230, for each condition monitoring measurement in a failure history, the method first uses the Generalized Weibull-FR function to fit the measurement series, and then uses the fitted measurement values to train the ANN”—[wherein the BRI of selecting is any action taken to choose the most suitable choice, and wherein a group can be fitted to map the monotonic relationship (i.e., selecting one relationship) from the group]); and
some of the sensing factors in the groups having more than one sensing factors are not selected (Husain ¶0027: “In step 210, a training set may be formed from a set of run-to-failure bearing data. After forming the training set, in step 220 a correlation analysis can be performed to determine which of the measurements may be most correlated with the bearing degradation. After the correlation analysis, in step 230, for each condition monitoring measurement in a failure history, the method first uses the Generalized Weibull-FR function to fit the measurement series, and then uses the fitted measurement values to train the ANN”—[wherein the measurements that are not most correlated are not chosen]).
The system of Tsutsui, the teachings of Husain, and the instant application are analogous art because they pertain to making predictions about semiconductor processing using neural networks.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the system of Tsutsui with the teachings of Husain to provide sensor data filtering by separating the data into groups according to a correlations and then selecting them to be used for the prediction. One would be motivated to do so to determine the relationship to the features and the equipment to better understand the system (Husain ¶0014: “After extracting the features, in step 120, correlation coefficients are calculated for each column by determining a moving average of the vibration signals and time. One such correlation coefficient could be a Spearman's rank correlation, which can be used to assess how strong a monotonic relationship is between a vibration signal and time. The plurality of columns may therefore be ranked according to the strength of this relationship”).
Tsutsui in view of Husain does not appear to explicitly teach:
obtain a plurality of equipment sensing data;
a second Neural Network model, configured to receive the equipment sensing data to obtain a second prediction result;
a third database, configured to store metrology inspection data [including a plurality of actual measurement data and a plurality of virtual measurement data],
wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, [and the virtual measurement data is obtained via a simulation procedure];
a third Neural Network model, configured to receive the metrology inspection data to obtain a third prediction result; and
a total prediction unit, configured to obtain a plurality of total prediction results according to the first prediction result, the second prediction result and the third prediction result.
However, Jung teaches:
obtain a plurality of equipment sensing data (Jung Col. 4, lines 5–23: “Data that is extracted, accumulated, and processed by the data extraction unit 110, and then is learned by the learning unit 120 will be described. The production-related data refers to historical data regarding overall production, in which data regarding a process ID, an equipment ID, a work space in equipment, an equipment work specification ID, etc. performed from a time when the manufacturing of a semiconductor starts until a present time are arranged in time series. The equipment-related data refers to data which is measured through an internal/external sensor in equipment during processes from the time when the manufacturing of the semiconductor starts until the present time, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like. That is, the equipment-related data may be referred to as input data regarding operating/state of equipment in which various physical/electrical state values regarding equipment are arranged in time series”—[wherein the data extraction unit obtains the equipment sensing data internal/external sensors (i.e., sensed equipment data)]);
a second Neural Network model, configured to receive the equipment sensing data to obtain a second prediction result (Jung Col. 3, lines 42–46: “The data extracted by the data extraction unit 110 may be divided into 1) production-related data, 2) equipment-related data, and 3) quality data. The 3) quality-related data may be divided into 31) fault data and 32) measurement data. These types of data will be described below in detail”; see also Jung Col. 3, lines 51–63: “The learning unit 120 repeats a procedure of generating neural network models by performing machine learning with respect to the data extracted/processed by the data extraction unit 110, and of updating the neural network models according to input of additional data. The neural network models generated/updated by the learning unit 120 are stored in the storage unit 130. Since the neural network models are generated/updated differently according to types of data as described above, the storage unit 130 may include a production data neural network model 131, an equipment data neural network model 132, a fault data neural network model 133, and a measurement data neural network model 134”; see also Jung Col. 5, lines 52–62: “The learning unit 120 may train the equipment data neural network model 132 by inputting, as a cause value (X), time-series operating/state data of equipment measured through an internal/external sensor of the equipment, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like, to an input node of the equipment data neural network model 132, and by positioning, as an effect value (Y), a real semiconductor manufacturing yield value at an output node of the equipment data neural network model 1132”—[wherein the different types of data (e.g., equipment related data; i.e., equipment sensing data) are processed (i.e., input) by the different neural network models according to the data type (e.g., equipment data neural network model; i.e., second neural network) to predict the effect value Y]);
a third database, configured to store metrology inspection data [including a plurality of actual measurement data and a plurality of virtual measurement data] (Jung Col. 4, lines 63–67: “On the other hand, the input data received in real time for prediction of a semiconductor yield is accumulated in an in-memory database (DB) provided in the prediction unit. The in-memory DB may query about a memory block rapidly in an index method and may obtain data”; see also Jung Col. 3, lines 42–46: “The data extracted by the data extraction unit 110 may be divided into 1) production-related data, 2) equipment-related data, and 3) quality data. The 3) quality-related data may be divided into 31) fault data and 32) measurement data. These types of data will be described below in detail”—[wherein the real-time metrology data is stored in the in-memory database (i.e., a third database), and wherein the BRI of metrology inspection data includes measurement data and wherein the data extraction unit extracts (i.e., obtains) quality data (i.e., metrology inspection data)]),
wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data, [and the virtual measurement data is obtained via a simulation procedure] (Jung Col. 4, lines 14–23: “The equipment-related data refers to data which is measured through an internal/external sensor in equipment during processes from the time when the manufacturing of the semiconductor starts until the present time, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like. That is, the equipment-related data may be referred to as input data regarding operating/state of equipment in which various physical/electrical state values regarding equipment are arranged in time series”; see also Jung Col. 4, lines 24–28: “The quality data refers to data which is obtained by measuring quality of an intermediate product from the time when the manufacturing of the semiconductor starts until the present time, and may be divided into fault data and measurement data”—[wherein the BRI of heterogeneous data is a dataset composed of different data types, structures, formats, or sources, and wherein recipe data, equipment related data, and quality data (i.e., recipe data, equipment sensing data, and metrology data) each include different data types, structures, formats, or sources (e.g., according to recipe I, II, III, temperature, pressure, ratio of materials, fault data and measurement data)]);
a third Neural Network model, configured to receive the metrology inspection data to obtain a third prediction result (Jung Col. 3, lines 42–46: “The data extracted by the data extraction unit 110 may be divided into 1) production-related data, 2) equipment-related data, and 3) quality data. The 3) quality-related data may be divided into 31) fault data and 32) measurement data. These types of data will be described below in detail”; see also Jung Col. 3, lines 51–63: “The learning unit 120 repeats a procedure of generating neural network models by performing machine learning with respect to the data extracted/processed by the data extraction unit 110, and of updating the neural network models according to input of additional data. The neural network models generated/updated by the learning unit 120 are stored in the storage unit 130. Since the neural network models are generated/updated differently according to types of data as described above, the storage unit 130 may include a production data neural network model 131, an equipment data neural network model 132, a fault data neural network model 133, and a measurement data neural network model 134”; see also Jung Col. 6, lines 17–25: “The learning unit 120 may train the measurement data neural network model 134 by inputting, as a cause value (X), imaged or time-series measurement data indicating a line width, thickness, and misalignment of a circuit to an input node of the measurement data neural network model 134, and by positioning, as an effect value (Y), a real semiconductor manufacturing yield value at an output node of the measurement data neural network model 134”—[wherein the different types of data (e.g., equipment related data; i.e., equipment sensing data) are processed (i.e., input) by the different neural network models according to the data type (e.g., measurement data neural network model; i.e., third neural network) to predict the effect value Y]); and
a total prediction unit, configured to obtain a plurality of total prediction result according to the first prediction result, the second prediction result and the third prediction result (Jung Col. 3, lines 64–67 – Col. 4, lines 1–3: “The prediction unit 140 predicts a semiconductor manufacturing yield by using the neural network models 131, 132, 133, 134 stored in the storage unit 130. The semiconductor manufacturing yield is predicted when there is a request for prediction of a yield with data from external systems 11, 12, 13, and a result of prediction may be returned to a corresponding external system”—[wherein the prediction unit predicts the yield the outputs from each model, and wherein the BRI of total prediction unit is code, instructions, or software capable of being executed by a processor or circuitry (see present disclosure para [0015])]).
The system of Tsutsui in view of Husain, the teachings of Jung, and the instant application are analogous art because they pertain to making predictions about semiconductor processing using neural networks.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the system of Tsutsui in view of Husain with the teachings of Jung to provide sensor data from the equipment to the prediction model for semiconductor processing. One would be motivated to do so to predict results that are appropriate for characteristics of the data and to provide the ability to adjust the flow of processes to improve productivity and yield (Jung Col. 2, lines 44–61: “According to embodiments of the present disclosure as described above, it is possible to exactly predict a result value by ensuring a neural network model of a structure appropriate for characteristics of data by applying different neural network models according to types of data. In addition, when embodiments of the present disclosure are applied to prediction of a semiconductor manufacturing yield, the flow of processes can be adjusted to enhance productivity/yield, and a fault can be detected early and thus it is possible to take appropriate response in advance. In addition, when embodiments of the present disclosure are applied to prediction of a semiconductor manufacturing yield, a product that is predicted as not having a good yield can be specially observed/managed, and the yield prediction function can be integrated into a platform and can be used for common use in various manufacturing processes. In addition, various feedback results may be combined and analyzed and thus can be utilized in various fields”).
Tsutsui in view of Husain and Jung does not appear to explicitly teach:
[a third database, configured to store metrology inspection data] including a plurality of actual measurement data and a plurality of virtual measurement data; and
[wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data,] and the virtual measurement data is obtained via a simulation procedure.
However, Yong teaches:
[a third database, configured to store metrology inspection data] including a plurality of actual measurement data and a plurality of virtual measurement data (Yong Pg. 4: “The virtual measurement method and the reliability estimation method according to the present invention for achieving the above object generates a virtual measurement model using the sensor data of the process equipment and the measured value of the measurement equipment, the reliability corresponding to the virtual measurement model A model is generated, the sensor data at the time of performing the process is substituted into the virtual measurement model to predict the measurement value, and the reliability of the measurement value is estimated”—[(emphasis added)]); and
[wherein the equipment recipe data, the equipment sensing data and the metrology inspection data are heterogeneous data,] and the virtual measurement data is obtained via a simulation procedure (Yong Pg. 4: “In addition, the estimation of the reliability may be performed by loading the generated reliability model to estimate the reliability of the predicted measured value. If the estimated reliability is greater than or equal to the reference value, the actual measurement is omitted, and if the estimated reliability is less than the reference value, Request measurement”—[(emphasis added)]).
The system of Tsutsui in view of Husain and Jung, the teachings of Yong, and the instant application are analogous art because they pertain to making predictions about semiconductor processing using neural networks.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the system of Tsutsui in view of Husain and Jung with the teachings of Yong to provide sensor data from the equipment and from a simulation when making predictions with a model for semiconductor processing. One would be motivated to do so to strengthen the measurement system used to detect abnormalities in the manufacturing process in order to reduce cost and process cycles, improving manufacturing yield (Yong Pg. 2: “the object of the present invention is to predict the measurement information for all the wafers in real time during the process to reduce the overall process cost and shorten the process cycle, defective wafer It is to provide a virtual measurement system and a virtual measurement method that can improve manufacturing yield through prior discovery and action”).
Regarding claims 11–13, although varying in scope, the limitations of claims 11–13 are substantially the same as the limitations of claims 3–5, respectively. Thus, claims 11–13 are rejected using the same reasoning and analysis as claims 3–5 above, respectively.
Regarding claim 15, although varying in scope, the limitations of claim 15 are substantially the same as the limitations of claim 7. Thus, claim 15 is rejected using the same reasoning and analysis as claim 7 above.
Claims 6 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Tsutsui in view of Husain, Jung, and Yong, and further in view of Foster et al., (US 5273588 A), hereinafter “Foster”.
Regarding claim 6 Tsutsui in view of Husain, Jung, and Yong teaches all the limitations of claim 1.
Tsutsui in view of Husain, Jung, and Yong does not appear to explicitly teach:
wherein a plurality of processes executed by the pieces of equipment are continuously executed.
However, Foster teaches:
wherein a plurality of processes executed by the pieces of equipment are continuously executed (Foster ¶0009: “Efficient commercial production of semiconductor wafers requires that the processing equipment function as continuously as possible. When deposits form on interior components of processing chambers, such as those of CVD reactors, they become ineffective and their use must be suspended for cleaning. Many reactors of the prior art require cleaning at an undesirable frequency, or are too difficult and too slow to clean, thus resulting in excessive reactor downtime. Accordingly, there is a continuing need for processing chambers such as those of CVD reactors that require less frequent cleaning of components, that reduce unwanted deposition on components, and that can be cleaned more rapidly”—[emphasis added]).
The methods of Tsutsui in view of Husain, Jung, and Yong, the teachings of Foster, and the instant application are analogous art because they pertain to apparatuses for industrial manufacturing.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the methods of Tsutsui in view of Husain, Jung, and Yong with the teachings of Foster to provide for semiconductor processes to function continuously. One would be motivated to do so to remain as efficient as possible (Foster ¶0009: “Efficient commercial production of semiconductor wafers requires that the processing equipment function as continuously as possible. When deposits form on interior components of processing chambers, such as those of CVD reactors, they become ineffective and their use must be suspended for cleaning. Many reactors of the prior art require cleaning at an undesirable frequency, or are too difficult and too slow to clean, thus resulting in excessive reactor downtime. Accordingly, there is a continuing need for processing chambers such as those of CVD reactors that require less frequent cleaning of components, that reduce unwanted deposition on components, and that can be cleaned more rapidly”).
Regarding claim 14, although varying in scope, the limitations of claim 14 are substantially the same as the limitations of claim 6. Thus, claim 14 is rejected using the same reasoning and analysis as claim 6 above.
Claims 8 and are 16 rejected under 35 U.S.C. 103 as being unpatentable over Tsutsui in view of Husain, Jung, and Yong, and further in view of Saqlain et al., ("A Voting Ensemble Classifier for Wafer Map Defect Patterns Identification in Semiconductor Manufacturing," in IEEE Transactions on Semiconductor Manufacturing, vol. 32, no. 2, pp. 171-182, May 2019, doi: 10.1109/TSM.2019.2904306.), hereinafter “Saqlain”.
Regarding claim 8 Tsutsui in view of Husain, Jung, and Yong teaches all the limitations of claim 1.
Tsutsui in view of Husain, Jung, and Yong does not appear to explicitly teach:
wherein in the step of obtaining the total prediction result according to the first prediction result, the second prediction result and the third prediction result, the total prediction result is obtained through a voting procedure.
However, Saqlain teaches:
wherein in the step of obtaining the total prediction result according to the first prediction result, the second prediction result and the third prediction result, the total prediction result is obtained through a voting procedure (Saqlain Pg. 172, Left Col.: “The goal of ensemble methods is to combine the prediction results of various ML models within given learning parameters and generate a final prediction result to improve the accuracy. Comparing the performance of individual classifiers, ensemble classifiers have shown more effective results regarding stability and robustness [18]. All the ensemble systems consist of three basic pillars, including diversity, training of all member classifiers of ensemble systems, and combining results of all these members using simple or weighted majority voting to get an aggregate result”—[(emphasis added)]).
The methods of Tsutsui in view Husain, Jung, and Yong, the teachings of Saqlain, and the instant application are analogous art because they pertain to using predictive modeling with industrial manufacturing processes.
It would be obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the methods of Tsutsui in view Husain, Jung, and Yong with the teachings of Saqlain to provide a voting procedure to determine the overall result. One would be motivated to do so to improve accuracy of the prediction (Saqlain Pg. 172, Left Col.: “The goal of ensemble methods is to combine the prediction results of various ML models within given learning parameters and generate a final prediction result to improve the accuracy. Comparing the performance of individual classifiers, ensemble classifiers have shown more effective results regarding stability and robustness [18]. All the ensemble systems consist of three basic pillars, including diversity, training of all member classifiers of ensemble systems, and combining results of all these members using simple or weighted majority voting to get an aggregate result).
Regarding claim 16, although varying in scope, the limitations of claim 16 are substantially the same as the limitations of claim 8. Thus, claim 16 is rejected using the same reasoning and analysis as claim 8 above.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NICHOLAS SHINE whose telephone number is (571)272-2512. The examiner can normally be reached M-F, 11am – 7pm ET.
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, David Yi can be reached on (571) 270-7519. 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.
/N.B.S./Examiner, Art Unit 2126
/DAVID YI/Supervisory Patent Examiner, Art Unit 2126