DETAILED ACTION
This Action is responsive to Claims filed 04/17/2026.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 04/17/2026 has been entered.
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 the Claims
Claims 1-6, 7, 9, 12-19, and 21 have been amended. Claims 8 and 20 have been canceled. Claims 1-7, 9, 12-19, and 21 are currently pending.
Response to Arguments
Applicant's arguments, see Pages 8-15, filed 04/17/2026, with regards to the 35 U.S.C. 101 Rejection of Claims 1-9 and 12-21 have been fully considered but they are not persuasive.
The newly amended limitations point to an algorithmic series of data manipulation or operational steps but for the recitation of a generic computing environment performing the steps, or instructions to apply the results of the algorithmic steps.
The claimed “communication interface” is highly generic, and amounts to a means to receive data. The claimed “a storage that stores…” is highly generic, and amounts to a means to store data. The claimed “processor” is highly generic, and amounts to a mere generic computing component or processing element.
The “extracting…” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from performing a spectral subtraction technique in mixed spectrum analysis on optical spectrum data in order to generically extract spectrum data. This limitation is merely performed by a generic processor.
The “converts” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from using a calibration curve to convert the extracted spectrum data into a concentration. This limitation is merely performed by a generic processor.
The “acquires…” and subsequent “inputting…” step amounts to the normal operation of a generic machine learning model (data is input, a prediction result is acquired). The data input is manipulated by the aforementioned abstract idea mental process steps. The generic processor generically acquires the result. The step amounts to mere data transmittal, and the “inputting…” step amounts to instruction to apply the data manipulated by the abstract idea mental process steps.
The “generates” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from calculating a difference between the optical spectrum data and the spectrum data of the physical component. This limitation is merely performed by a generic processor.
The second “acquires…” step again amounts to typical, generic machine learning model behavior. Data is input, and a prediction result is obtained. The data input is manipulated by the aforementioned abstract idea mental process steps. The generic processor generically acquires the result.
The second “generates” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from generating total information obtained by adding multiple prediction results generically acquired by a generic processor. This limitation is merely performed by a generic processor.
The “transmits” step amounts to mere data transmittal. This limitation is merely performed by a generic processor.
The “the control system controls…” limitation, likewise, amounts to instructions to apply the data generically generated by a generic processing component in the second “generates…” step.
Given the above interpretation, the Examiner respectfully disagrees with the Applicant regarding the Claims’ eligibility. As presently drafted, the central limitations of the claim are a set of algorithmic data manipulation or calculation steps without specific structure or implementation precluding a human mind with or without the aid of pen and paper from completely them. The steps are performed by a generic processing element (amounting to instructions to apply), and the recitation of the control system and neural network(s) are highly generic, only producing a particular or specific output due to the manipulations performed on the data input to them.
Because of the highly generic nature of the of the instructions to apply the algorithmic steps, any improvement realized must be rooted in the specific data manipulations performed within the abstract idea mental process steps, before it is applied to the output of the neural network(s) and/or control system. Per MPEP 2106.05(a), the specific improvement cannot come from the abstract idea. See the updated 35 U.S.C. 101 Rejection below.
Applicant’s arguments, filed 04/17/2026, with regards to the 35 U.S.C. 103 Rejection(s) of Claims 1-9 and 12-21 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 Rejections - 35 USC § 101
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-7, 9, 12-19, and 21 rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more; and because the claims as a whole, considering all claim elements both individually and in combination, do not amount to significantly more than the abstract idea, see Alice Corporation Pty. Ltd. v. CLS Bank International, et al, 573 U.S. (2014). In determining whether the claims are subject matter eligible, the Examiner applies the 2019 USPTO Patent Eligibility Guidelines. (2019 Revised Patent Subject Matter Eligibility Guidance, 84 Fed. Reg. 50, Jan. 7, 2019.)
Step 1:
Claims 1-6, 9, and 12-19, and 21 recite an information processing device, which falls under the statutory category of a machine. Claim 7 recites an information processing method, which falls under the statutory category of a process.
Step 2A – Prong 1:
Claim 1 recites an abstract idea, law of nature, or natural phenomenon. The limitations of “extracts…”, “converts…”, “generates…”, and “generates…” under the broadest reasonable interpretation, cover a mental process including an observation, evaluation, judgment or opinion that could be performed in the human mind or with the aid of pencil and paper.
The “extracts…” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from performing a spectral subtraction technique in mixed spectrum analysis on optical spectrum data in order to generically extract spectrum data. This limitation is merely performed by a generic processor.
The “converts” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from using a calibration curve to convert the extracted spectrum data into a concentration. This limitation is merely performed by a generic processor.
The “generates” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from calculating a difference between the optical spectrum data and the spectrum data of the physical component. This limitation is merely performed by a generic processor.
The second “generates” step is interpretable as an abstract idea mental process step. There is nothing recited in the limitation precluding a human mind from generating total information obtained by adding multiple prediction results generically acquired by a generic processor. This limitation is merely performed by a generic processor.
Step 2A – Prong 2:
The additional elements of claim 9 do not integrate the abstract idea into a judicial exception. The claim recites the additional elements “A control system”, “an information processing device”, “a communication interface”, “a storage” and “total information” are recognized as generic computer components recited at a high level of generality. Although they have and execute instructions to perform the abstract idea itself, this also does not serve to integrate the abstract idea into a practical application as it merely amounts to instructions to "apply it." (See MPEP 2106.04(d)(2) indicating mere instructions to apply an abstract idea does not amount to integrating the abstract idea into a practical application).
The additional elements of “a hybrid model”, “a physical model”, “a raw material”, “a state of one of the plants”, and “machine learning model” are recognized as non-generic computer components, but are recited at a high level of generality and are found to generally link the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)).
The limitations of “receives optical spectrum data of the raw material from the plant;”, “stores a hybrid model, including a physical model and machine learning models, generated based on data concerning the plant;”, “…inputting the concentration of the principal component of the raw material obtained from the optical spectrum data into the physical model,” and “…inputting the first prediction result and the difference spectrum data into the machine learning models, wherein the other prediction results include information that corrects an output of the physical model,” are found to be mere data transmittal extra-solution activity steps (See MPEP 2106.05(g)).
The limitations of “acquires a first prediction result that indicates a state of the plant…”, “acquires other prediction results that indicate a state of the plant…”, and “the control system controls at least one of a temperature, a pressure, and a flow rate in the plant by operating one or more actuators of the plant based on the operation content.” are found to be mere instructions to apply the abstract idea of generating the total information See MPEP 2106.05(f)).
Step 2B:
The only limitation on the performance of the described method is a limitation reciting “A control system”, “an information processing device”, “a communication interface”, “a storage” and “total information” These elements are insufficient to transform a judicial exception to a patentable invention because the recited elements are considered insignificant extra-solution activity (generic computer system, processing resources, links the judicial exception to a particular, respective, technological environment). The claim thus recites computing components only at a high-level of generality such that it amounts to no more than mere instructions to apply the exception using generic computer components; mere instructions to apply an exception using a generic computer component cannot provide an inventive concept (see MPEP 2106.05(f)).
The additional elements of “a hybrid model”, “a physical model”, “a raw material”, “a state of one of the plants”, and “machine learning model” are recognized as non-generic computer components, but are recited at a high level of generality and are found to generally link the abstract idea to a particular technological environment or field of use (See MPEP 2106.05(h)).
The limitations of “receives optical spectrum data of the raw material from the plant;”, “stores a hybrid model, including a physical model and machine learning models, generated based on data concerning the plant;”, “…inputting the concentration of the principal component of the raw material obtained from the optical spectrum data into the physical model,” and “…inputting the first prediction result and the difference spectrum data into the machine learning models, wherein the other prediction results include information that corrects an output of the physical model,” are found to be mere data transmittal and are well-understood, routine, or conventional activity (See MPEP 2106.05(d)(II)(i)).
The limitations of “acquires a first prediction result that indicates a state of the plant…”, “acquires other prediction results that indicate a state of the plant…”, and “the control system controls at least one of a temperature, a pressure, and a flow rate in the plant by operating one or more actuators of the plant based on the operation content.” are found to be mere instructions to apply the abstract idea of generating the total information See MPEP 2106.05(f)).
Taken alone or in ordered combination, these additional elements do not amount to significantly more than the above-identified abstract idea. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation.
For the reasons above, claim 1 is rejected as being directed to non-patentable subject matter under §101. This rejection applies equally to independent claim 7.
Claim 7 recites similar limitations to claim 9, with the exception of “An information processing method comprising:” (generic computer components), therefore both claims are similarly rejected.
Dependent Claims:
Claim 2 recites abstract idea mental process steps “generated based on…” and “provide…” and data transmittal step “acquires…”
Claim 3 recites abstract idea mental process steps “generated based on…”, “generated based on…” and “obtained by…” and instructions to apply/data transmittal steps “acquires…” and “acquires…”
Claim 4 recites refinements to the instructions to apply step of Claim 1.
Claim 5 recites an abstract idea mental process step “generating…”
Claim 6 recites data transmittal extra-solution activity steps.
Claim 9 recites an abstract idea mental process step “generates…”
Claim 12 recites extra-solution activity data collection steps.
Claim 13 recites refinements to the data types
Claim 14 recites refinements to the data types
Claim 15 recites refinements to the data types
Claim 16 recites refinements to the data types
Claim 17 recites refinements to the data types
Claim 18 recites refinements to the data types
Claim 19 recites refinements to the data types
Claim 21 recites data transmittal steps.
Claim Rejections - 35 USC § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-2, 7, 13-17, 19, and 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chan et al. (US 11,853,032 B2, filed 05/06/2020), hereinafter Chan; Montavon et al. (Machine learning of molecular electronic properties in chemical compound space, 2013), hereinafter Montavon; Thomas et al. (Comparison of Multivariate Calibration Methods for Quantitative Spectral Analysis, 1990), hereinafter Thomas; and Mehrubeoglu et al. (Detection and identification of plastics using SWIR hyperspectral imaging, 2020), hereinafter Mehrubeoglu.
In regards to claim 1: The present invention claims: “A plant operation assistance system comprising: a control system that controls operation of a plant that…and an information processing device comprising: a communication interface…a storage that stores a hybrid model, including a physical model and machine learning models generated based on data concerning the plant; and a processor that:” Chan Figures 1A, 12, 13, and 14 at least demonstrate these technical features or hardware implementations. Chan also teaches “The hybrid models 116, 516, 716 predict, with improved accuracy, the progress and physical characteristics/conditions of the subject chemical process 124.”
“acquires a first prediction result that indicates a state of the plant…” Chan teaches “In turn, module 106 automatically generates a list of features based on first principles knowledge (such as thermodynamic and chemical engineering domain knowledge) and models of chemical processes. These features could include physical properties such as density, viscosity, heat capacity, dimensionless numbers corresponding to phenomena occurring in the process as well results calculated by a theoretical model of the process. The automatically generated list of features transforms the raw input measurements from module 104 into more reliable and representative inputs into the machine learning training model 108. The resulting new features or augmented input data (XA) of module 106 can be a transform of an existing input, a combination of inputs, or a calculation from a model.” (Column 7, Lines 7-21). Mapping “first principle knowledge” to “a physical model” based on “These "full-scale" models may consist on thousands to millions of mathematical equations representing physical and chemical properties as well as mass and energy balances in a chemical process under consideration.” (Chan Column 1, Lines 19-22) and Applicant’s Specification [0028]).
“acquires other prediction results that indicate a state of the plant…wherein the other prediction results include information that corrects an output of the physical model,” Chan teaches “In embodiments of the present invention, the process modeling system 130 generates and deploys hybrid models 116, 516, 716 (detailed later) of the subject chemical process 124 by combining a first principles model 126 and a machine learning model 108, 508, 706.” (Column 6, Lines 20-24). Chan, Column 8, Lines 25-34 detail the uses and purposes of the predictive outputs within the system.
“generates, based on a total information amount obtained by adding the first prediction result and the other prediction results, operation content that specifies an update of at least one of a temperature setting, a pressure setting, and a flow-rate setting for controlling the plant, and transmits the operation content to the control system, wherein, the control system controls at least one of a temperature, a pressure, and a flow rate in the plant by operating one or more actuators of the plant based on the operation content.” Chan teaches “By combining these methods (machine learning and first principles modeling)“ (Column 5, Lines 54-55) and “The hybrid models 116, 516, 716 predict, with improved accuracy, the progress and physical characteristics/conditions of the subject chemical process 124.” (Column 6, Lines 25-27). Mapping the combination of first principles modeling (“a first prediction result that indicates a state of the plant…”) and machine learning (“other prediction results…”) resulting in an output. Chan also teaches “The inferentials could include concentrations and flows of byproducts, temperature, or pressures inside the equipment etc.” (Column 12, Lines 34-40).”
Chan fails to explicitly teach “extracts spectrum data of a principal component from the optical spectrum data by a spectral subtraction technique in mixed spectrum analysis.” However, Mehrubeoglu teaches “Each pixel’s spectrum represents the (mixed) material spectral signature observed at the spatial location of the pixel. Spatial visualization, on the other hand represents the scene with the objects’ physical and relative placement in space, more specifically, the 2D projection of the 3D scene. Spatial information at different spectral bands can be extracted from hyperspectral data cubes as image frames corresponding to a narrow wavelength range based on the manufacturer’s system and internal calibration.” (Section 2.2).
Mehrubeoglu teaches “Most plastics are typically transparent in the visible spectral range, rendering them challenging to detect using silicon based vision sensors. In this work a SWIR hyperspectral imaging system is used to collect the SWIR hyperspectral signatures as well as spatial information of a variety of plastics outdoors to test this technology for plastic debris detection and identification in future marine and environmental applications.” (Abstract). It would have been obvious to one of ordinary skill in the art at the time of the Applicant’s filing to use both imaging and spectrographic techniques when analyzing a chemical or other industrial output to address the challenges highlighted above.
“generates difference spectrum data by calculating a difference between the optical spectrum data and the spectrum data of the principal component,” Mehrubeoglu teaches “The authors show principal component analysis (PCA) score plots for first and second principal components and use partial least-squares discriminant analysis to differentiate among polymers such as PP, PE, and expanded Polystyrene.” (Page 2, Section 1.2) and “The authors make use of distinctive absorption bands as spectral indices (ratio or difference of two bands) for the three materials together with correlation matrix analysis for a real-time application. After continuum removal, the authors identify the spectral index of 1170:1650 (ratio 1) to separate PET from PLA and PS, and 1160:1140 (ratio 2) to separate PLA from PS after that, using a decision tree procedure.” (Page 3).
“…by inputting the first prediction result and the difference spectrum data into the machine learning models,” Chan teaches “Next, module 104 feeds the input values (X) into a simulation model 506 to predict the output (YS). Additionally, the input values (X) can be augmented with module 106 (discussed above) before developing the simulation model. In tum, module 507 calculates the residual (R) as the 102 from the chemical process 124 and difference between the simulation prediction (YS) and the observed output (Y). Training step 508 trains and develops a machine learning model 516 for the residual. In this case, the resulting machine learning model 516 is not trying to capture all of the underlying physics of the subject chemical process but instead only modeling the portion of the industrial system/chemical process not described by first principles.” (Column 11, Lines 10-22).
The combination of Chan and Mehrubeoglu fails to explicitly teach: “converts the extracted spectrum data into a concentration of the principal component by using a calibration curve indicating a relationship between an intensity and a concentration,” Thomas, which provides more detail into Principal Component Analysis for spectral analysis, teaches “PLS and PCR are both factor-based methods that are capable of being full-spectrum methods. These methods have been explained and contrasted recently by Haaland and Thomas(6). Like ILS, PLS and PCR can be employed even when only one component is known in the calibration samples. Both PLS and PCR methods factor the spectral data calibration matrix into the product of two smaller matrices. This amounts to a data compression step where the intensities at all frequencies used in the analysis are compressed to a small number of intensities in a new full-spectrum coordinate system.” (Pages 1091-1092), see also Table I (Page 1092) for Factor Levels used in PCA, including calibration, concentration, and intensities.
Thomas demonstrates the concepts recited related to spectrographic analysis would have been known to one of ordinary skill in the art at the time of the Applicant’s filing date. A person of ordinary skill in the art would have known to leverage known methods to realize known benefits in a combination of Chan and Mehrubeoglu.
While the aforementioned figures of Chan read on a communication interface, Chan fails to explicitly teach: “processes waste plastic as a raw material;” and “a communication interface that receives optical spectrum data of a raw material from a plant;” However, Montavon teaches the use of machine learning for identifying structure-relationships of chemical compounds (Abstract, Page 3) and the use of optical spectrum data in the analysis of chemical compounds (Section 2.3, first paragraph and Page 10).
Montavon highlights the benefits of using machine learning in the analysis for the identification of meaningful, novel and predictive structure–property relationships in chemical compounds (Abstract). It would have been obvious to one of ordinary skill in the art at the time of the applicant’s filing to use relevant chemical compound data in an industrial analysis system like that of Chan to improve the overall accuracy and information robustness of the system.
“…by inputting a concentration of a principal component of the raw material obtained from the optical spectrum data into the physical model” Chan teaches “Second, the hybrid model is able to accurately predict quantities that are important for monitoring the process but may not have been measured (also referred to as inferentials) due to limitations of instrumentation and other factors. The inferentials could include concentrations and flows of byproducts, temperature, or pressures inside the equipment etc.” (Column 12, Lines 34-40).” In a system combining Chan and Montavon, it would be reasonable to measure or collect data on concentrations of chemicals.
In regards to claim 2: The present invention claims ”wherein the machine learning models include a machine learning model that is generated based on, as training data, data that provide an error to prediction of the physical model,” See the Rejection of claim 1 where Chan teaches the physical model and machine learning models. Chan also teaches “Instead, error calculation module or step 708 calculates the errors of the output prediction (YS) of combined machine learning and first principles models 706, 707 relative to the field measurement outputs (Y). If the calculated error (/(YS)-(Y)/) does not satisfy a threshold acceptability level, then error calculation module 708 propagates at 709 the calculated errors into the machine learning model 706 for training.” (Column 12, Lines 50-57, mapping to training the machine learning model on an error based on the physical model).
“and the processor acquires the other prediction results that include a prediction result that is provided based on the machine learning model.” Chan teaches “In embodiments, the computer-implemented method and system further comprises the step of or a configured module calculating a predicted value for output from a simulator, the first principles model forming the simulator. The method/system trains and develops the machine learning model to represent differences between observed output variable values from plant data and corresponding output variable values predicted by the simulator.” (Column 2, Lines 53-60).
In regards to claim 7: Claim 7 recites similar limitations to claim 1, with the exception of “An information processing method comprising:” therefore, both claims are similarly rejected.
In regards to claim 13: The present invention claims: “wherein the first prediction result is composition data.” Montavon teaches “We have introduced a ML model for predicting the electronic properties of molecules based on training deep multi-task artificial NNs in chemical space.” and “Due to its inherent first principles setup, we expect the overall approach to be equally applicable to molecules or materials of arbitrary size, configurations and composition—without any major modification.” A person of ordinary skill in the art combining the teachings of Chan and Montavon may have reasonably been making relevant predictions to a chemical’s composition.
In regards to claim 14: The present invention claims: “wherein the data that provide the error is foreign substance data specified by imaging analysis.” Mehrubeoglu teaches “The detection and identification of microplastic debris is commonly a three-step process: 1) Bulk or concentrated water samples are passed through a membrane filter to collect microplastic debris, 2) Non-synthetic debris (i.e., natural particulate organic matter) is digested using hydrogen peroxide, and 3) Particles and fibers are detected and identified visually using microscopy and/or chemically using spectroscopy (e.g., FTIR and Raman) or chromatography (e.g., pyrolysis GC/MS and HPLC)7 .” (Introduction).
In regards to claim 15: The present invention claims: “wherein the data that provide the error is foreign substance data specified by spectrum analysis.” Mehrubeoglu teaches “The detection and identification of microplastic debris is commonly a three-step process: 1) Bulk or concentrated water samples are passed through a membrane filter to collect microplastic debris, 2) Non-synthetic debris (i.e., natural particulate organic matter) is digested using hydrogen peroxide, and 3) Particles and fibers are detected and identified visually using microscopy and/or chemically using spectroscopy (e.g., FTIR and Raman) or chromatography (e.g., pyrolysis GC/MS and HPLC)7 .” (Introduction).
In regards to claim 16: The present invention claims: “wherein the data that provide the error indicate contamination of the raw material for waste plastic recycling.” Mehrubeoglu Section 1.2 goes into detail regarding plastics, plastics as contaminants, and plastic contamination in the context of spectrographic analysis. It would have been obvious to one of ordinary skill in the art when combining Chan, Montavon, and Mehrubeoglu to utilize similar data if it were relevant to the operation of the plant.
In regards to claim 17: The present invention claims: “wherein the data that provide the error indicate a production area of the raw material.” Chan teaches “Plant equipment includes distillation columns, various kinds of reactors and reactor tanks, evaporators, pipe systems, valves, heaters, etc. by way of illustration and not limitation. Plant data 102 represents inputs (feed amounts, values of certain variables, etc.) and outputs (products, residuals, physical operating characteristics/conditions, etc.) of the chemical process 124.” (Column 6, Lines 5-11).
In regards to claim 19: The present invention claims: “wherein the data that provide the error indicate an operation load of the plant.” Chan teaches “The step of modeling is automated and implemented by a processor including generating a model that predicts progress (e.g., operating conditions, physical properties, etc.) of the chemical process.” (Column 1, Lines 54-57).
In regards to claim 21: The present invention claims: “wherein the processor further acquires composition data including temperature and composition information output from the physical model based on an initial condition of the plant and the optical spectrum data.” See the rejection of Claim 13 where Montavon teaches collecting composition data, and the rejection of Claim 1 where Chan refers to the collection of temperature data. Chan reads on the collection or prediction of plant-based condition or operation data and how Montavon reads on the collection or observation of chemical-based sprectrographical data.
Claim(s) 3-6 and 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chan, Thomas, Montavon, and Mehrubeoglu as applied to claim 1 above, and further in view of Xu et al. (US 2018/0136617 A1, filed 11/08/2017), hereinafter Xu.
In regards to claim 3: While Chan teaches multiple models (column 6, Lines 20-25), and the machine learning model(s) of Chan reads on “a second machine learning model that is generated based on plant data that are generated at a target plant that is a prediction target,” and “and acquires a third prediction result based on the second machine learning model,” in “The interface is coupled to the modeling subsystem in a manner that enables improvements in performance of the chemical process at the subject 10 industrial plant based on predictions made by the generated model.” (Column 2, Lines 7-11)
Chan fails to explicitly teach “wherein the machine learning models include a first machine learning model that is generated based on data that are common to a plurality of plants” and “the processor acquires a second prediction result based on the first machine learning model” However, Xu, in a similar field of industrial system monitoring, teaches “A method of continuously modeling industrial asset performance includes an initial model build block creating a first model based on a combination of an industrial asset historical data, configuration data and training data…” (Abstract). Mapping the use of historical data to “data that are common to a plurality of plants” based on “The training data can also include historical data, which can include monitored data from sensors for the particular physical asset and monitored data from other industrial assets of the same type and nature. The historical data, asset configuration data and domain knowledge can be used to create an initial model.” ([0044]). Data collected regarding similar industrial sensors or assets would reasonably be common among plants of similar kinds.
“and total information is obtained by further adding the third prediction result.” Both Chan (referenced above) and Xu (Figure 1, item 155) reference outputting based on multiple models’ results.
Xu highlights the complexity of industrial systems and the difficulty standard machine learning practices can have modeling them ([0003]). It would have been obvious to one of ordinary skill in the art at the time of the Applicant’s filing to incorporate additional, similar, or historic data into the model(s) of Chan (which it is noted also performs augmentation of input data to increase accuracy) to increase predictive performance of the one or more models.
In regards to claim 4: The present invention claims “wherein the processor operates the target plant based on a correction result that is provided by correcting the first prediction result with the second prediction result and the third prediction result.” Chan teaches “Process modeling system 130 uses the corresponding residual model 516 to correct the first principle model predicted amounts of a physical condition or property. The resulting corrected physical condition prediction (resulting predicted amounts) is improved in accuracy, thus improving output (model of the chemical process of interest 124) of the process modeling system 130. As a consequence, controller 122 outputs improved in accuracy settings (values) 132 and updates thereto for controlling the chemical process 124 and industrial plant 120 operations.” (Column 11, Lines 57-66).
In regards to claim 5: Chan reads on “wherein the processor generates: the first machine learning model based on training data” and “and the second machine learning model based on training data that include plant data that are obtained at a target chemical plant that is a prediction target.“ in “The interface is coupled to the modeling subsystem in a manner that enables improvements in performance of the chemical process at the subject 10 industrial plant based on predictions made by the generated model.” (Column 2, Lines 7-11) and the use of multiple models;
“that include experiment data that are obtained by a study concerning a chemical plant that produces a new chemical product where a used chemical product is provided as the raw material thereof,” Montavon teaches the use of machine learning for identifying structure-relationships of chemical compounds (Abstract, Page 3).
In regards to claim 6: The present invention claims: “wherein the processor: inputs the optical spectrum data, the first prediction result, and the second prediction result into the second machine learning model to acquire the third prediction result.” Montavon teaches the use of optical spectrum data in the analysis of chemical compounds (Section 2.3, first paragraph and Page 10). It would have reasonable to one of ordinary skill in the art to include similar data into a system combining aspects of Chan and Montavon.
In regards to claim 9: The present invention claims: “wherein the processor: generates the physical model that predicts, by a simulation that uses data that are obtained from the raw material of a product that is produced by each of the plurality of plants, a state of one of the plurality of plants;” Chan teaches “In embodiments, the computer-implemented method and system further comprises the step of or a configured module calculating a predicted value for output from a simulator, the first principles model forming the simulator.” (Column 2, Lines 53-56).
Claim(s) 12 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chan, Thomas, Montavon, and Mehrubeoglu as applied to claims 1 and 2 above, and further in view of Anderson et al. (US 2017 /0284903 A1), hereinafter Anderson.
In regards to claim 12: The combination of Chan and Montavon fails to explicitly teach sensing or sensors as claimed in “wherein the optical spectrum data is obtained by spectroscopic sensing in the plant.” However, Anderson, in a similar field of endeavor of plant management teaches “Although network devices 204-209 are shown in FIG. 2 as a mobile phone, laptop computer, tablet computer, temperature sensor, motion sensor, and audio sensor respectively, the network devices may be or include sensors that are sensitive to detecting aspects of their environment. For example, the network devices may include sensors such as water sensors, power sensors, electrical current sensors, chemical sensors, optical sensors, pressure sensors, geographic or position sensors (e.g., GPS), velocity sensors, acceleration sensors, flow rate sensors, among others.” ([0060]). A person of ordinary skill in the art combining the industrial sensing mechanisms of Chan and/or Anderson and the spectrum analysis of Montavon would reasonably be obtaining spectrographic data from spectroscopic or optical sensors.
Anderson teaches “Machines can be used to perform various processes. For example, industrial plants that process chemicals can include machines, such as heaters, furnaces, and fired heaters, that perform various steps to process the chemicals. Various characteristics of the machines can be monitored using sensors. This may enable operators of the machines to detect failures of the machines, safety hazards, etc.” ([0003]). It would have been obvious to one of ordinary skill in the art at the time of the Applicant’s filing in combining Chan, Montavon, and Anderson to utilize similar data if it were relevant to the operation of the plant.
In regards to claim 18: The combination of Chan and Montavon fails to explicitly teach “wherein the data that provide the error indicate an external environment of the plant including average temperature and rainfall.” Anderson teaches “The sensors ll04a-d can be positioned to detect characteristics of the machine 1102, ambient conditions ( e.g., near to the machine 1102), or both of these. In an example in which the machine 1102 is a furnace, the sensors ll04a-d can detect a firing rate of the furnace, a feed rate of a material into or through the furnace, a temperature in a bridge-wall section of the furnace, a temperature in a stack section of the furnace, an atmospheric temperature, a humidity, a wind direction, or any combination of these.” ([0136]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GRIFFIN T BEAN whose telephone number is (703)756-1473. The examiner can normally be reached M - F 7:30 - 4:30.
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, Li Zhen can be reached at (571) 272-3768. 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.
/GRIFFIN TANNER BEAN/Examiner, Art Unit 2121
/Li B. Zhen/Supervisory Patent Examiner, Art Unit 2121
Prosecution Insights