DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Responsive to the communication dated 4/20/2023.
Claims 1 – 20 presented for examination.
Priority
ADS dated 3/1/2023 claims foreign priority to IN 2022110739078 dated 12/20/2022.
Electronic priority documents submitted 4/30/2023.
Information Disclosure Statement
No IDS provided.
Drawings
The drawings dated 3/1/2023 have been reviewed. They are accepted.
Specification
The abstract has 133 words, 10 lines, and no legal phraseology. The abstract is accepted.
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.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Mishra_2021 (US 11,017,321 B1) in view of Alkadi_2021 (US 2021/0017926 A1).
Claim 1. Mishra_2021 makes obvious “An apparatus comprising at least one processor and at least one non-transitory memory comprising program product code stored thereon, wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, cause the apparatus to at least: (FIG. 1 Block 104: “processor”; COL 5 lines 24 – 30: “… a non-transitory computer readable storage medium stores instructions that, when executed by one or more processors, cause the one or more processors to perform operations for event categorization and maintenance action recommendation using machine learning…”) receive current operating conditions data associated with current operation of one or more operational systems (FIG. 1 block 136: “operating characteristics data”, block 152 “sensor(s)”; FIG. 4 block 406: “Real-Time and Historical Operational Data”; COL 9 lines 39 – 60: “… receive input data that indicates operating characteristics associated with the equipment asset 150 and event data indicating the events… the monitoring device 102 may receive operating characteristics data 136 rom the sensors 152… the operating characteristics data 136, which may also be referred to herein as operating conditions data, may indicate various operating characteristics or sensor readings associated with the equipment… such as temperature, pressures, vibrations, and the like…”); Generate, using a prediction model, at least one fprediction corresponding to the current operation of the one or more operational systems (FIG. 4 block 450: “insight and recommendation engine”, block 452: “insights”; COL 2 lines 52 – 57: “… the operating characteristics data and high-priority events (e.g., “worthy” events) may be analyzed using artificial intelligence and machine learning processes to determine a status (e.g., an insight) associated with the equipment asset…”; COL 3 lines 42 – 44: “… ML model that are configured to determine an “insight” such as status, associated with the equipment asset…”)“Wherein the prediction model comprises a machine learning model trained based at least in part on historical operating conditions data associated with past operation of the one or more operational systems and historical data associated with the past operation of the one or more operational systems” (FIG. 1 block 126: “Machine Learning Model(s)”; Block 130: “Machine Learning Model(s); Block 134: “Machine Learning Model(s) 134”; FIG. 3 illustrates the training of a artificial intelligence or machine learning model; FIG. 5 block 596: “provide… input data… to a first machine learning (ML) model to identify a status associated with the industrial machinery…”; COL 1 lines 10 – 15: “… leveraging machine learning and artificial intelligence to automatically analyze and categorize events, recommend equipment status (e.g., “insights”) based on the events, and generate commands or recommendations for maintenance actions…”; COL 3 lines 23 – 35: “… ML models may include or correspond to neural networks (NNs), support vector machines (SVMs), decision trees, random forest, regression models, Bayesian networks (BNs), dynamic Bayesian networks (DBNs), naïve Bayesian (NB) models, Gaussian processes, hidden Markov Models (IMMIs), and the like, that are configured to perform clustering. The first set of ML models may be trained using training data that is generated based on historical operating characteristics data, historical event data, ratings associated with historical events (e.g., from the knowledge base), and the like…”; COL 3 lines 55 – 60: “… ML model may be trained using training data that is generated based on historical operating characteristic data, historical event data, statuses (e.g., insights) associated with the historical events, and operating characteristics, and the like…”; COL 4 lines 5 – 15: “… ML model may include or correspond to NNs, SVMs, decision trees, random forests, regression models, BNs, DBNs, NB models, Gaussian processes, HMMs, and the like… may be trained using training data that is generated based on historical operating characteristics data, historical work orders (e.g., data indicative of previously performed maintenance actions), historical event data, historical insight-action relationship data, and the like…”) Wherein the prediction model is configured to generate the prediction based at least in part on the current operating conditions data (COL 32 lines 16 – 20: “… the method 300 includes model deployment, at 324. For example, the selected ML model from the candidate ML models may be deployed, such as being integrated within a monitoring device or an expert system, for generating real-time predictions associated with the equipment asset…”; COL 1 lines 33 – 35: “… predicting faults associated with industrial machinery before the faults occur and performing maintenance to prevent, or reduce the impact… predictive maintenance can include identifying issues that may result in future faults…”); And output the at least one prediction (FIG. 1: block 138 “output” shown on Block 140 “Display Device”; COL 42: “… FIG. 1 may provide the output 138 to the display device 140 to initiate display of the status 114, the priority events 112, at least a portion of the operating characteristics data 136, or a combination thereof…”; FIG. 4 block 450: Insights and Recommendation Engine”; Block 452: “Insights”; COL 7 lines 61 – 65: “… an equipment status or status of an equipment asset may refer to any equipment related insight that may be inferred predicted, recommended, or determined based on detected events, operating characteristics of the equipment asset, or other information or a combination thereof…”)
While Mishra_2021 teaches that operating states includes insights and insights are prediction made by a machine learning model and further recites, in the paragraphs in the body of the document, “leakage crossing the tolerance limit” (COL 30) and that the sensors include “leakage sensors” (COL 15), the paragraphs in the body of the document to not explicitly recite “fugitive leak.” Because TABLE 3, titled: “Example Insights for a Control Valve” includes “Fugitive emissions leak from the valve into the atmosphere, posting potential environmental and safety hazards”, the Examiner finds that, it may properly be found that one of ordinary skill in the art would find it obvious that the insight predictions made by the machine learning model would include “fugitive emissions.”
Nevertheless, Alkadi_2021 clearly teaches predicting “fugitive emissions” (Par 2: “monitoring and detection of gas leaks is commonly performed… gas leaks can occur as a result of equipment failures which can cause the release of unplanned, or fugitive gaseous emissions… localized weather patterns can alter the concentration and distribution of the gas emission making it difficult to accurately determine an emission source associated with the gas leak…”; Par 3: “… the method can include receiving near-field sensor data and far-field sensor data from one or more sensors… the method can further include determining gas concentration data associated with the gas emission. The method can include determining an emission rate corresponding to the gas emission. The method can also include generating emission data corresponding to the gas emission. The emission data can include the determined emission rate and one or more source locations associated with the gas emission. The method can further include providing the emission data…”; Par 24: “… gas emissions 120 occurring during unexpected, anomalous operating conditions can be referred to as fugitive emissions… gas emissions 120 can include common gas species such as methane, ethane, propane, butane, hexane, and other hydrocarbons such as natural gas liquids (C5, C6, C8-10), mixtures of alkanes, sour gases include H2S, SOX, carbon disulfide, unsaturated HCs/petrochemicals such as ethylene, propylene, or the like which can be emitted from various emission sources 115 and/or detected vis clients 125…”; Par 28: “… the emission analyzer 135 can also include a prediction module 145 configured to predict an emission rate for a particular gas emission… in some embodiments, the optimization module 150 can be configured in a machine learning process to perform the data-driven modeling and statistical analysis of the gas emissions 120… in the event of determining anomalous venting or fugitive emissions 120, the control module 155 can execute instructions to control the operating parameters of the emission source…”; Par 80: “… prediction system described herein addresses the technical problem of determining an emission rate of a gas emission emanating from a gas source based on received sensor data… determining emission rates for expected, vented emissions as well as unexpected, fugitive emissions…”).
Mishra_2021 and Alkadi_2021 are analogous art because they are from the same field of endeavor called monitoring and detecting status/condition of equipment including leaks. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Mishra_2021 and Alkadi_2021, The rationale for doing so would have been that Mishra_2021 teaches to monitor and predict potential failures of equipment and to make recommendations to repair personnel in order to mitigate and/or prevent the failures that include leaks by use of a machine learning algorithm. Alkadi_2021 teaches that gas leaks can occur as a result of equipment failures which can cause the release of unplanned, or fugitive gaseous emission and further that weather patterns can alter the concentration, location, and distribution of those gasses making it difficult to accurately determine an emission source associated with the gas leak (par 2). Alkadi_2021 further teaches that gas leaks in the environment can create hazardous operating conditions for personnel assigned to operate, maintain, and repair the industrial assets and can reduce production rates (par 2). Alkadi_2021 further teaches to use a machine learning model to predict fugitive gaseous emissions. Therefore, it would have been obvious to combine the machine learning algorithms and asset monitoring and recommendation system as taught by Mishra_2021 with the fugitive emissions prediction as taught by Alkadi_2021 for the benefit of monitoring and predicting potential hazardous fugitive emissions in order to prevent hazardous operating conditions for personnel and to help maintain production rates to obtain the invention as specified in the claims.
Claim 11. The limitations of claim 11 are substantially the same as those of claim 1 and are rejected due to the same reasons as outlined above for claim 1. Additionally, Mishra_2021 makes obvious the further limitations of “A computer-implemented method comprising: Receiving current operating conditions data associated with current operation of one or more operational systems” (FIG. 1 Block 104: “processor”; COL 5 lines 24 – 30: “… a non-transitory computer readable storage medium stores instructions that, when executed by one or more processors, cause the one or more processors to perform operations for event categorization and maintenance action recommendation using machine learning…”).
Claim 20. The limitations of claim 20 are substantially the same as those of claim 1 and are rejected due to the same reasons as outlined above for claim 1. Additionally, Mishra_2021 makes obvious the further limitations of “a computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising an executable portion configured to: (FIG. 1 Block 104: “processor”; COL 5 lines 24 – 30: “… a non-transitory computer readable storage medium stores instructions that, when executed by one or more processors, cause the one or more processors to perform operations for event categorization and maintenance action recommendation using machine learning…”) receive current operating conditions data associated with current operation of one or more operational systems” (FIG. 4 block 406: “real-time and historical operational data”; COL 21 lines 60: “… monitoring operating characteristics and other real-time data provided by the equipment asset 150 and the sensors 152 to identify the priority event 112, determine the status 114… to prevent (or reduce a severity of) a fault at the equipment asset 150. By using the trained ML models 126…”).
Claim 2, 12. Mishra_2021 makes obvious “wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, further cause the apparatus to at least: Generate at least one repair alert for one or more components of the one or more operational systems based at least in part on the fugitive leak predictions, wherein to output the at least one fugitive leak prediction the apparatus is caused to at least output the at least one repair alert” (COL 21 lines 1 – 10: “… initiate a warning (e.g., an alert message, a visual warning, an audio warning, or the like), initiate a shutdown of the equipment asset 150, and the like, as non-limiting examples…”; COL 26 lines 10 – 16: “as another example, the maintenance performance system 208 may automatically perform one or more of the maintenance actions, such an emitting an audible alert, ordering a replacement part for the equipment asset, performing diagnostics on particular parameters of the operating characteristic data, or the like…”).
Alkadi_2021 also makes obvious “wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, further cause the apparatus to at least: Generate at least one repair alert for one or more components of the one or more operational systems based at least in part on the fugitive leak predictions, wherein to output the at least one fugitive leak prediction the apparatus is caused to at least output the at least one repair alert” (par 81: “… the client 125 can include an improved display or graphical user interface (GUI) that provides more efficient visualization and execution of emission data… the improved GUI can also provide enhanced visualization for alerts or notifications of gas emissions, planning maintenance or repair procedures for emission sources, or for managing production rates of the gas…”).
Claim 3, 13. Alkadi_2021 also makes obvious “wherein each of the fugitive leak predictions comprises one or more predicted fugitive emissions values representing predicted fugitive emissions associated with the one or more operational systems” (Figure 5 block 530, 535, 540: determine emission rate, generate emission date, provide emission data; par 68: “… the prediction module 145 determines the emission rate associated with the received sensor data corresponding to a particular emission source…”).
Claim 4, 14. Mishra_2021 makes obvious “wherein the fugitive leak predictions are generated based at least in part on correlations between one or more sensed operating conditions values of the historical operating conditions data and one or more sensed fugitive emissions values of the historical fugitive emissions data, wherein each sensed operating conditions values of the one or more sensed operating conditions values represents an operating condition of the one or more operational systems at an instance of time during the past operations of the one or more operational systems, wherein each sensed fugitive emissions values of the one or more sensed fugitive emissions values represents fugitive emissions at an instance of time during the past operation of the one or more operational systems, and wherein the fugitive leak prediction model is trained based at least in part on the correlations” (FIG. 4 block 432: “analytic engine (e.g., correlation, regression, clustering, survival)”’ COL 28 line 59 – COL 29 line 5: “… feature selection may include extracting features from the subset of data that are highly influential in predicting the target value… such that substantially irrelevant features are not included in the training data, which may improve performance of the ML Models and reduce complexity of the ML models. To features that are selected may be based on statistical analysis of the subset of data and the target variables, for example, using methods such as correlation…” EXAMINER NOTE: the above teaches to use correlation to have correlated training data for the Machine Learning model for the purpose of reducing complexity. Accordingly, predictions are generated based at least in part on correlations and the prediction model is trained based at least in part on the correlations. COL 36 lines 46 – 66: “… correlation analysis may also indicate the strength of linear relationships between different variables… correlation analysis may be performed to assess the strength of relationships for all existing and derived features…”; lines 23 – 35: “… ML models may be trained using training data that is generated based on historical operating characteristics data, historical event data, ratings associated with historical events (e.g., from the knowledge base), and the like…”; COL 3 lines 55 – 60: “… ML model may be trained using training data that is generated based on historical operating characteristic data, historical event data, statuses (e.g., insights) associated with the historical events, and operating characteristics, and the like…”; COL 4 lines 5 – 15: “… ML model may include or correspond to NNs, SVMs, decision trees, random forests, regression models, BNs, DBNs, NB models, Gaussian processes, HMMs, and the like… may be trained using training data that is generated based on historical operating characteristics data, historical work orders (e.g., data indicative of previously performed maintenance actions), historical event data, historical insight-action relationship data, and the like…”; COL 33 lines 24 – 25: “… the data sources 402 include one or more sources of data for use in training ML models and analyzing by the insight platform… the data sources 402 may include data from one or more other devices or location that is associated with an equipment asset, such as the equipment asset 150 of FIG. 1. For example… real-time and historical operational data 406, historical events 408…”).
Claim 5, 15. Mishra_2021 makes obvious “wherein the at least one non-transitory memory and the program product code are configured to, with the at least one processor, further cause the apparatus to at least: train the fugitive leak prediction model based at least in part on the historical operating conditions data and the historical fugitive emissions data” (FIG. 3; COL 3 lines 23 – 35: “… ML models may include or correspond to neural networks (NNs), support vector machines (SVMs), decision trees, random forest, regression models, Bayesian networks (BNs), dynamic Bayesian networks (DBNs), naïve Bayesian (NB) models, Gaussian processes, hidden Markov Models (IMMIs), and the like, that are configured to perform clustering. The first set of ML models may be trained using training data that is generated based on historical operating characteristics data, historical event data, ratings associated with historical events (e.g., from the knowledge base), and the like…”; COL 3 lines 55 – 60: “… ML model may be trained using training data that is generated based on historical operating characteristic data, historical event data, statuses (e.g., insights) associated with the historical events, and operating characteristics, and the like…”; COL 4 lines 5 – 15: “… ML model may include or correspond to NNs, SVMs, decision trees, random forests, regression models, BNs, DBNs, NB models, Gaussian processes, HMMs, and the like… may be trained using training data that is generated based on historical operating characteristics data, historical work orders (e.g., data indicative of previously performed maintenance actions), historical event data, historical insight-action relationship data, and the like…”)
Alkadi_2021 also makes obvious “wherein the at least one non-transitory memory and the program product code are configured to, with the at least one processor, further cause the apparatus to at least: train the fugitive leak prediction model based at least in part on the historical operating conditions data and the historical fugitive emissions data” (par 67, 68).
Claim 6, 16. Alkadi_2021 makes obvious wherein the fugitive leak prediction model is trained based at least in part on simulated fugitive emissions data associated with simulated operation of the one or more operational systems, and the fugitive leak predictions are generated based at least in part on correlations between one or more simulated operating conditions values of the simulated fugitive emissions data and one or more estimated fugitive emissions values of the simulated fugitive emissions data associated with the one or more simulated operating conditions values, wherein the fugitive leak prediction model is trained based at least in part on the correlations, and Wherein each simulated operating conditions value of the one or more simulated operating conditions values represents an operating condition of the one or more operational systems during the simulated operation of the one or more operational systems, and wherein each estimated fugitive emissions value of the one or more estimated fugitive emissions values represents estimated fugitive emissions resulting from the one or more simulated operating conditions during the simulated operation of the one or more operational systems” (Par 57: “… in predicting gas emission rates… in some embodiments, the Far-Field model can be generated via reduced order modeling and/or via a machine learning model that has been trained using computational fluid dynamic (CFD) simulation datasets. The CFD simulation datasets represent wind vector datasets including wind velocity fields on selected Far-Field wind conditions, such as boundary conditions for a particular CFD simulation. The CFD simulation datasets also include bluff body shape factors that represent typical oil production facility object shapes…”; Par 67: “… data-driven modeling approach can be devised. This data-driven model can estimate the near-field wind information to be used in the dispersion model… to build such a model that can operate over a wide range of wind conditions, CFD models of the flow around a subset of representative infrastructures and different values of the wind vector (wind speed and directions) can be discretized… from this ensemble of simulations, a database of input and output features corresponding to open (far) field and near-field velocity measurements can be generated for training a machine learning (ML) model… the data was split into training and validation datasets…”).
While Alkadi_2021 teaches to simulate at least fugitive leak data known as far-field wind data that is used to create a database of emission values used for training a machine learning model, and while it may be obvious to those of ordinary skill in the art that the predicted emission is correlated to the wind field data and accordingly the predicted emission is based on correlation and the training is also based on this correlation. Nevertheless, Alkadi_2021 does not clearly discuss correlation.
Mishra_2021, however, makes obvious “predictions are generated based at least in part on correlations” and “prediction model is trained based at least in part on the correlations” (FIG. 4 block 432: “analytic engine (e.g., correlation, regression, clustering, survival)”’ COL 28 line 59 – COL 29 line 5: “… feature selection may include extracting features from the subset of data that are highly influential in predicting the target value… such that substantially irrelevant features are not included in the training data, which may improve performance of the ML Models and reduce complexity of the ML models. To features that are selected may be based on statistical analysis of the subset of data and the target variables, for example, using methods such as correlation…” EXAMINER NOTE: the above teaches to use correlation to have correlated training data for the Machine Learning model for the purpose of reducing complexity. Accordingly, predictions are generated based at least in part on correlations and the prediction model is trained based at least in part on the correlations. COL 36 lines 46 – 66: “… correlation analysis may also indicate the strength of linear relationships between different variables… correlation analysis may be performed to assess the strength of relationships for all existing and derived features…” COL 3 discusses: “… machine learning models (ML) that are configured to group the events into clusters based on categories… ML models may include or correspond to… support vector machines (SVMs)…” EXAMINER NOTE: When a Support Vector Machine (SVM) finds a hyperplane that maximizes the margin between classes, it inherently relies on the premise that items within the same class share underlying similarities—which can be interpreted as a form of correlation).
Claim 7, 17. Alkadi_2021, at paragraph 67, teaches to train a machine learning model to predict fugitive emissions and to use an ensemble of simulated data to build a database that is used to train a machine learning model and that the training data is split into training and validation datasets. Further, Mishra_2021, at FIG. 3, illustrates that training and validation sets are applied to candidate machine learning models to evaluate the effectiveness of the training. Because the validation set (e.g., golden set) is used as ground-truth to evaluate the machine learning models predicted fugitive emissions, it is obvious to those of ordinary skill in the art to generate the simulated fugitive emissions data used to evaluate the candidate model during training by generating the one or more estimated fugitive emissions values based at least in part on the one or more simulated operating conditions values.
Therefore, the combination of Mishra_2021 and Alkadi_2021 makes obvious wherein the at least one non-transitory memory and the program code are configured to, with the at least one processor, further cause the apparatus to at least: generate the simulated fugitive emissions data by generating the one or more estimated fugitive emissions values based at least in part on the one or more simulated operating conditions values.
Claim 8, 18. Mishra_2021 makes obvious wherein the fugitive leak prediction model is trained based at least in part on historical fugitive leak data associated with the one or more operational systems, wherein the historical fugitive leak data identifies one or more fugitive leaks detected within the one or more operational systems at one or more instances of time during the past operation of the one or more operational systems” (FIG. 3; COL 3 lines 23 – 35: “… ML models may include or correspond to neural networks (NNs), support vector machines (SVMs), decision trees, random forest, regression models, Bayesian networks (BNs), dynamic Bayesian networks (DBNs), naïve Bayesian (NB) models, Gaussian processes, hidden Markov Models (IMMIs), and the like, that are configured to perform clustering. The first set of ML models may be trained using training data that is generated based on historical operating characteristics data, historical event data, ratings associated with historical events (e.g., from the knowledge base), and the like…”; COL 3 lines 55 – 60: “… ML model may be trained using training data that is generated based on historical operating characteristic data, historical event data, statuses (e.g., insights) associated with the historical events, and operating characteristics, and the like…”; COL 4 lines 5 – 15: “… ML model may include or correspond to NNs, SVMs, decision trees, random forests, regression models, BNs, DBNs, NB models, Gaussian processes, HMMs, and the like… may be trained using training data that is generated based on historical operating characteristics data, historical work orders (e.g., data indicative of previously performed maintenance actions), historical event data, historical insight-action relationship data, and the like…”)
Claim 9, 19. Mishra_2021 makes obvious wherein the fugitive leak predictions are generated based at least in part on correlations between one or more sensed operating conditions values of the historical operating conditions data corresponding to the one or more instances of time at which the one or more fugitive leaks are detected and estimated fugitive emissions values determined with respect to the one or more sensed operating conditions values, and wherein the fugitive leak predictions model is trained at least in part on the correlations” (FIG. 4 block 432: “analytic engine (e.g., correlation, regression, clustering, survival)”’ COL 28 line 59 – COL 29 line 5: “… feature selection may include extracting features from the subset of data that are highly influential in predicting the target value… such that substantially irrelevant features are not included in the training data, which may improve performance of the ML Models and reduce complexity of the ML models. To features that are selected may be based on statistical analysis of the subset of data and the target variables, for example, using methods such as correlation…” EXAMINER NOTE: the above teaches to use correlation to have correlated training data for the Machine Learning model for the purpose of reducing complexity. Accordingly, predictions are generated based at least in part on correlations and the prediction model is trained based at least in part on the correlations. COL 36 lines 46 – 66: “… correlation analysis may also indicate the strength of linear relationships between different variables… correlation analysis may be performed to assess the strength of relationships for all existing and derived features…” COL 3 discusses: “… machine learning models (ML) that are configured to group the events into clusters based on categories… ML models may include or correspond to… support vector machines (SVMs)…” EXAMINER NOTE: When a Support Vector Machine (SVM) finds a hyperplane that maximizes the margin between classes, it inherently relies on the premise that items within the same class share underlying similarities—which can be interpreted as a form of correlation).
Claim 10. Mishra_2021 makes obvious “wherein the historical fugitive emissions data identifies one or more sensed fugitive emissions values representing fugitive emissions sensed by one or more fugitive emissions sensors during the past operation of the one or more operational systems” (FIG. 4 real-time historical operational data”; Fig. 1 sensors 152; lines 23 – 35: “… ML models may include or correspond to neural networks (NNs), support vector machines (SVMs), decision trees, random forest, regression models, Bayesian networks (BNs), dynamic Bayesian networks (DBNs), naïve Bayesian (NB) models, Gaussian processes, hidden Markov Models (IMMIs), and the like, that are configured to perform clustering. The first set of ML models may be trained using training data that is generated based on historical operating characteristics data, historical event data, ratings associated with historical events (e.g., from the knowledge base), and the like…”; COL 3 lines 55 – 60: “… ML model may be trained using training data that is generated based on historical operating characteristic data, historical event data, statuses (e.g., insights) associated with the historical events, and operating characteristics, and the like…”; COL 4 lines 5 – 15: “… ML model may include or correspond to NNs, SVMs, decision trees, random forests, regression models, BNs, DBNs, NB models, Gaussian processes, HMMs, and the like… may be trained using training data that is generated based on historical operating characteristics data, historical work orders (e.g., data indicative of previously performed maintenance actions), historical event data, historical insight-action relationship data, and the like…”; COL 33 lines 24 – 25: “… the data sources 402 include one or more sources of data for use in training ML models and analyzing by the insight platform… the data sources 402 may include data from one or more other devices or location that is associated with an equipment asset, such as the equipment asset 150 of FIG. 1. For example… real-time and historical operational data 406, historical events 408…”; COL 15 line 43: “… the sensors 152 may include… leakage sensors…”; TABLE 4: “… Opr_Leakage_Rate…”).
Alkadi_2021 makes obvious “wherein the historical fugitive emissions data identifies one or more sensed fugitive emissions values representing fugitive emissions sensed by one or more fugitive emissions sensors during the past operation of the one or more operational systems” (par 67, 68: “the prediction model 145 determines the emission rate associated with the received sensor data corresponding to a particular emission source…”).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN S COOK whose telephone number is (571)272-4276. The examiner can normally be reached 8:00 AM - 5:00 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Emerson Puente can be reached at 571-272-3652. 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.
/BRIAN S COOK/Primary Examiner, Art Unit 2187
Prosecution Insights