DETAILED ACTION
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 13 April 2026 has been entered.
Response to Arguments
Applicant's arguments with respect to the rejection of claims 1-6, 8-16, 19, and 20 under 35 U.S.C. § 101 have been fully considered but they are not persuasive.
Applicant argues on p. 6 that “even if, arguendo, the amended claims could be interpreted as being directed to an alleged abstract idea, the amended claims recite significantly more than the alleged abstract idea that integrates into a practical application in industrial machine learning.” This argument is not persuasive. The claims do recite abstract ideas at Step 2A Prong One, the abstract ideas are not integrated into a practical application at Step 2A Prong Two, and the claims do not recite an inventive concept that makes the claims amount to significantly more than the abstract ideas. The claims are therefore ineligible as being directed to abstract ideas, as set forth in the rejection of the claims under 35 U.S.C. § 101 below.
The rejection of the claims under 35 U.S.C. § 112 is withdrawn, in light of the amendments to the claims.
Applicant's arguments with respect to the rejection of claims 1-6, 8-16, 19, and 20 under 35 U.S.C. § 102 have been fully considered but they are not persuasive.
Applicant argues on p. 7 that “the cited art of record does not disclose or suggest each and every feature of the amended claims.” This argument is not persuasive. The cited art of record discloses and/or suggests each and every feature of the amended claims, as set forth below in the rejection of claims 1-6, 8-16, 19, and 20 under 35 U.S.C. § 102.
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 1-6, 8-16, 19, and 20 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to abstract ideas without significantly more, as set forth below.
The following analysis is performed in accordance with the 2019 Revised Patent Subject Matter Eligibility Guidance (hereinafter 2019 PEG), as set forth in MPEP § 2106.
Step 1
Step 1 of the 2019 PEG asks whether the claim is to a process, machine, manufacture, or composition of matter.
Claims 1-6 and 8-16 are directed to a method.
Claims 19 and 20 are directed to an apparatus.
Step 2A Prong One
Step 2A Prong One of the 2019 PEG asks whether the claim recites an abstract idea, law of nature, or natural phenomenon.
The examiner has identified the following judicial exceptions in the claims:
Claim 1 recites:
receiving historical telemetry data associated with one or more refrigeration systems, wherein the historical telemetry data includes measured temperature values, historical equipment failures, and setpoint temperature values;
training a machine learning model using the historical telemetry data to derive an anomaly score, wherein the machine-learning model includes an encoding network and a decoding network, and wherein the machine-learning model is trained based on an evaluation between an input to the encoding network and an output from the decoding network; and
incrementing an anomaly count in response to determining that the anomaly score is greater than a threshold;
determining that the anomaly count is greater than a threshold count value;
providing an automatically determined indication in response to determining that the anomaly count is greater than a threshold count value; and
validating the machine-learning model using the historical equipment failures.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 1 therefore recites abstract ideas.
Claims 2-5 recite all of the limitations of claim 1, and therefore also recite abstract ideas.
Claim 6 recites:
processing the historical telemetry data using at least one of: transforming categorical variables, forward filling, determining relative values, or normalizing values, wherein the machine-learning model executes using the processed historical telemetry data.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 6 therefore recites abstract ideas.
Claim 8 recites:
generating an anomaly alert in response to the automatically determined indication.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 8 therefore recites abstract ideas.
Claim 9 recites:
wherein the machine-learning model is trained s using self-supervised learning.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 9 therefore recites abstract ideas.
Claim 10 recites:
wherein the machine-learning model includes an autoencoder.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 10 therefore recites abstract ideas.
Claim 11 recites:
processing the anomaly score to predict a likelihood of an equipment failure within a threshold failure time.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 11 therefore recites abstract ideas.
Claim 12 recites:
wherein providing the automatically determined indication includes outputting the automatically determined indication to a user interface of a diagnostic tool.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 12 therefore recites abstract ideas.
Claim 13 recites:
wherein providing the automatically determined indication includes outputting, on a user interface, anomaly data and refrigeration-dependent data.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 13 therefore recites abstract ideas.
Claim 14 recites:
wherein the refrigeration-dependent data includes work order data.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 14 therefore recites abstract ideas.
Claim 15 recites:
wherein the automatically determined indication is provided on a graph.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 15 therefore recites abstract ideas.
Claim 16 recites:
wherein providing the automatically determined indication includes displaying information associated with a user-selected point in time on the graph.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 16 therefore recites abstract ideas.
Claim 19 recites:
historical telemetry data of one or more refrigeration systems, wherein the historical telemetry data include measured temperature values, historical equipment failures, and setpoint temperature values;
training a machine learning model using the historical telemetry data to derive an anomaly score, wherein the machine-learning model includes an encoding network and a decoding network, and wherein the machine-learning model is trained based on an evaluation between an input to the encoding network and an output from the decoding network; and
incrementing an anomaly count in response to determining that the anomaly score is greater than a threshold;
determining that the anomaly count is greater than a threshold count value;
providing an automatically determined indication in response to determining that the anomaly count is greater than a threshold count value; and
validating the machine-learning model using the historical equipment failures.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 19 therefore recites abstract ideas.
Claim 20 recites:
receiving historical telemetry data associated with one or more refrigeration systems, wherein the historical telemetry data includes measured temperature values, historical equipment failures, and setpoint temperature values;
training a machine learning model using the telemetry data to derive an anomaly score, wherein the machine-learning model includes an encoding network and a decoding network, and wherein the machine-learning model is trained based on an evaluation between an input to the encoding network and an output from the decoding network; and
incrementing an anomaly count in response to determining that the anomaly score is greater than a threshold;
determining that the anomaly count is greater than a threshold count value;
providing an automatically determined indication in response to determining that the anomaly count is greater than a threshold count value; and
validating the machine-learning model using the historical equipment failures.
These claim limitations are abstract ideas of mathematical concepts, as discussed in MPEP §§2106.04(a)(2)(I), and/or mental processes, as discussed in MPEP §2106.04(a)(2)(III). Under the broadest reasonable interpretation, the mental processes cover performance of the limitations in the mind, and/or with pen and paper, but for the recitation of generic computer components that are used merely as a tool to implement the abstract ideas. That is, other than reciting a processor, nothing in the claim precludes the mental process steps from practically being performed in the human mind. Additionally, the mere nominal recitation of a generic processor does not take the claim limitations out of the mental processes grouping.
Claim 20 therefore recites abstract ideas.
Step 2A Prong Two
Step 2A Prong Two of the 2019 PEG asks whether a claim recites additional elements that integrate the judicial exception into a practical application.
Claims 1, 6, and 8-16 recite the additional elements of:
using one or more hardware processors.
The one or more hardware processors merely represent a generic computer performing generic computer functions to implement the abstract ideas on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. Accordingly, the additional element of a generic computer does not integrate the abstract ideas into a practical application, because it does not impose any meaningful limits on practicing the abstract idea.
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claims 1, 6, and 8-16 are directed to the judicial exception of abstract ideas.
Claim 2 recites the additional elements of:
wherein the telemetry data is collected by one or more sensors associated with the one or more refrigeration systems.
The one or more sensors associated with the one or more refrigeration systems merely represent the insignificant extra-solution activity of data gathering that is necessary for use of the recited judicial exceptions. Therefore, these additional elements represent insignificant extra-solution activity, as set forth in MPEP §2106.05(g). Additionally, the one or more sensors associated with the one or more refrigeration systems are merely an attempt to generally link the abstract ideas to a particular technological environment or field of use, as set forth in MPEP §2106.05(h).
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claim 2 is directed to the judicial exception of abstract ideas.
Claim 3 recites the additional elements of:
wherein at least one of the one or more sensors is a component included in the one or more refrigeration systems.
The one or more sensors associated with the one or more refrigeration systems merely represent the insignificant extra-solution activity of data gathering that is necessary for use of the recited judicial exceptions. Therefore, these additional elements represent insignificant extra-solution activity, as set forth in MPEP §2106.05(g). Additionally, the one or more sensors associated with the one or more refrigeration systems are merely an attempt to generally link the abstract ideas to a particular technological environment or field of use, as set forth in MPEP §2106.05(h).
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claim 3 is directed to the judicial exception of abstract ideas.
Claim 4 recites the additional elements of:
wherein at least one of the one or more sensors is configured to measure an ambient condition external to the one or more refrigeration systems.
The one or more sensors merely represent the insignificant extra-solution activity of data gathering that is necessary for use of the recited judicial exceptions. Therefore, these additional elements represent insignificant extra-solution activity, as set forth in MPEP §2106.05(g). Additionally, the one or more sensors are merely an attempt to generally link the abstract ideas to a particular technological environment or field of use, as set forth in MPEP §2106.05(h).
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claim 4 is directed to the judicial exception of abstract ideas.
Claim 5 recites the additional elements of:
wherein the historical telemetry data is collected periodically and continuously.
The data collection merely represents the insignificant extra-solution activity of data gathering that is necessary for use of the recited judicial exceptions. Therefore, these additional elements represent insignificant extra-solution activity, as set forth in MPEP §2106.05(g). Additionally, the data collection is merely an attempt to generally link the abstract ideas to a particular technological environment or field of use, as set forth in MPEP §2106.05(h).
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claim 5 is directed to the judicial exception of abstract ideas.
Claim 19 recites the additional elements of:
a communication interface configured to receive data;
a processor coupled to the communication interface.
The processor and communication interface merely represent a generic computer performing generic computer functions to implement the abstract ideas on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. Accordingly, the additional element of a generic computer does not integrate the abstract ideas into a practical application, because it does not impose any meaningful limits on practicing the abstract idea.
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claim 19 is directed to the judicial exception of abstract ideas.
Claim 20 recites the additional elements of:
a non-transitory computer readable medium storing instructions that when executed by a processor, cause the processor to perform operations.
The computer program product merely represents a generic computer performing generic computer functions to implement the abstract ideas on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. Accordingly, the additional element of a generic computer does not integrate the abstract ideas into a practical application, because it does not impose any meaningful limits on practicing the abstract idea.
Whether considered individually, or as an ordered combination with other claim elements, the additional elements do not integrate the abstract ideas into a practical application under any of the indicia set forth in MPEP § 2106.04(d), or improve the functioning of a computer, or any other technology or technical field as set forth in MPEP § 2106.05(a).
Therefore, claim 20 is directed to the judicial exception of abstract ideas.
Step 2B
Step 2B of the 2019 PEG asks whether the claim provide an inventive concept, i.e., whether the claim recites additional element(s) or a combination of elements that amount to significantly more than the judicial exception in the claim.
Regarding claims 1-6 and 8-16, as discussed with respect to Step 2A Prong Two, the additional elements of one or more hardware processors amount to no more than mere instructions to apply the exception using a generic computer component. The same analysis applies here in 2B, i.e., mere instructions to apply an exception using a generic computer component cannot integrate a judicial exception into a practical application at Step 2A, or provide an inventive concept to make the claim amount to significantly more than the judicial exceptions in Step 2B.
The various sensors and data represent the insignificant extra-solution activity of data gathering necessary to perform the abstract ideas and are recited at a high level of generality, and therefore fail to provide an inventive concept, as set forth in MPEP §§ 2106.05(g). Additionally, the various sensors and data measurement are merely an attempt to generally link the abstract ideas to a particular technological environment or field of use, as set forth in MPEP §2106.05(h).
Whether considered individually, or as an ordered combination with other claim elements, these additional elements represent mere instructions to apply an exception and insignificant extra-solution activity, which do not provide an inventive concept that makes the claims amount to significantly more than the abstract ideas.
For these reasons, there are no inventive concepts in claims 1-6 and 8-16, and claims 1-6 and 8-16 are therefore ineligible as being directed to judicial exceptions of abstract ideas.
Regarding claim 19, as discussed with respect to Step 2A Prong Two, the additional elements of processor and communication interface amount to no more than mere instructions to apply the exception using a generic computer component. The same analysis applies here in 2B, i.e., mere instructions to apply an exception using a generic computer component cannot integrate a judicial exception into a practical application at Step 2A, or provide an inventive concept to make the claim amount to significantly more than the judicial exceptions in Step 2B.
Whether considered individually, or as an ordered combination with other claim elements, these additional elements represent mere instructions to apply an exception and insignificant extra-solution activity, which do not provide an inventive concept that makes the claims amount to significantly more than the abstract ideas.
For these reasons, there are no inventive concepts in claim 19, and claim 19 is therefore ineligible as being directed to judicial exceptions of abstract ideas.
Regarding claim 20, as discussed with respect to Step 2A Prong Two, the additional element of computer program product amounts to no more than mere instructions to apply the exception using a generic computer component. The same analysis applies here in 2B, i.e., mere instructions to apply an exception using a generic computer component cannot integrate a judicial exception into a practical application at Step 2A, or provide an inventive concept to make the claim amount to significantly more than the judicial exceptions in Step 2B.
Whether considered individually, or as an ordered combination with other claim elements, these additional elements represent mere instructions to apply an exception and insignificant extra-solution activity, which do not provide an inventive concept that makes the claims amount to significantly more than the abstract ideas.
For these reasons, there are no inventive concepts in claim 20, and claim 20 is therefore ineligible as being directed to judicial exceptions of abstract ideas.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-6, 8-16, 19, and 20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Trinh et al., US 20200379454 A1 (hereinafter Trinh).
Regarding claim 1, Trinh teaches a method, comprising:
receiving historical telemetry data (“receiving a set of sensor data generated from sensors associated with equipment, one of the sensors being a target sensor, the set of sensor data comprising measured values of the target sensor,” ¶ 0004; “The machine learning model may be trained by a training data set that includes historical measurements of sensor data,” ¶ 0071) associated with one or more refrigeration systems, wherein the historical telemetry data includes measured temperature values, historical equipment failures, and setpoint temperature values (“a sensor 154 may monitor the temperature of a particular component of a piece of equipment. For every predetermined period of time (e.g., a second, a few seconds, a minute, an hour, etc.), the sensor 154 generates a data point,” ¶ 0041; “an operator may set a target temperature of a particular component or location of the equipment 150. The sensor readings may be fed back to the equipment 150, as represented by the arrows 158, to maintain the measurements of the components in the proximity of the target measurements specified in the settings. In another case, a setting value may be dynamic. For example, a target temperature of a particular component or location of the equipment 150 may be dynamically changed based on other conditions of the equipment 150. The values of the settings 152 may also be reported as a time series and transmitted to the data store 120 through the controller 160 or directly,” ¶ 0043; “The VAE model may be an unsupervised learning model that is trained based on training data that does not include labels or only includes a small number of labels on whether a piece of equipment is normal or defective or on the repair history of the equipment,” ¶ 0081);
training, using one or more hardware processors a machine learning model using telemetry data to derive an anomaly score, wherein the machine-learning model includes an encoding network and a decoding network, and wherein the machine-learning model is trained based on an evaluation between an input to the encoding network and an output from the decoding network (“The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152. The predictive maintenance server 110 may train one or more machine learning models that assign anomaly scores to a piece of equipment 150. The anomaly scores may include an overall anomaly score and individual anomaly scores each corresponding to a component, a measurement, or an aspect of the equipment 150,” ¶ 0044; “the neural network 900 may be a multi-layer neural network that might include an encoder 920, one or more bottleneck layers 930, and decoder 950,” ¶ 0083; “As another example, to train a model to perform a regression task, during the creation/training phase, one or more data elements of the training data are input to the model being trained, and the model generates output indicating a predicted value of one or more other data elements of the training data. The predicted values of the training data are compared to corresponding actual values of the training data, and the computer modifies the model until the model accurately and reliably (e.g., within some specified criteria) predicts values of the training data,” ¶ 0049);
determining that the anomaly count is greater than a threshold count value; incrementing an anomaly count in response to determining that the anomaly score is greater than a threshold (“When the anomaly scores are determined to be beyond a specific range such as above a predetermined threshold, the predictive maintenance server 110 identifies a particular facility site 140 and a particular piece of equipment 150 and provides an indication that the equipment 150 may need an inspection and possible repair. The predictive maintenance server 110 may also train additional models such as classifiers and regressors that can identify a specific component of the equipment 150 that may need an inspection, repair and/or replacement,” ¶ 0044; “The anomaly score is normalized so that the baseline anomaly score may be 1, which represents the average error in predicting the measurements of a vital using the machine learning model. A high dissimilarity metric value or anomaly score value such as a value within a range of 4 to 5 (even 8 to 9 at some points) indicates the loss of predictive power of other sensor channels to predict the values of the vitals. This indicates a time period that the equipment 150 may operate in an abnormal state. The predictive maintenance server 110 may send an alert to the facility site 140 and/or may display the alert in the front-end interface to indicate to an operator that maintenance may be needed for the equipment 150,” ¶ 0080);
providing an automatically determined indication in response to determining that the anomaly count is greater than a threshold count (“generating an anomaly score for the equipment based on the differences. The process may further include generating, based on anomaly score, an alert for the equipment,” ¶ 0004); and
validating the machine-learning model using the historical equipment failures (“The predictive maintenance server 110 may also process and validate data in a small batch for the scoring data 422 in block 440. If data validation fails, the predictive maintenance server 110 may trigger an alert in block 452 indicating potential problems might have occurred,” ¶ 0064; “After the machine learning model is trained, a series of output values (predicted values) may be generated for a historical period of time (e.g., one month). The series of predicted values of the vital may be compared to the actual values of the vital,” ¶ 0071).
Regarding claim 2, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein the historical telemetry data is collected by one or more sensors associated with the one or more refrigeration systems (“A piece of equipment 150 may include one or more settings 152 and one or more sensors 154 that are equipped to monitor one or more measures of the equipment 150, articles on which the equipment 150 operate, or the environment of the equipment 150. Example measurements that are monitored by various sensors 154 may include temperature,” ¶ 0040).
Regarding claim 3, Trinh teaches the invention of claim 2, as set forth in the rejection of claim 2 above. Trinh also teaches wherein at least one of the one or more sensors is a component is included in the one or more refrigeration systems (“A piece of equipment 150 may include one or more settings 152 and one or more sensors 154 that are equipped to monitor one or more measures of the equipment 150, articles on which the equipment 150 operate, or the environment of the equipment 150. Example measurements that are monitored by various sensors 154 may include temperature,” ¶ 0040).
Regarding claim 4, Trinh teaches the invention of claim 2, as set forth in the rejection of claim 2 above. Trinh also teaches wherein at least one of the one or more sensors is configured to measure an ambient condition external to the one or more refrigeration systems (“A piece of equipment 150 may include one or more settings 152 and one or more sensors 154 that are equipped to monitor one or more measures of the equipment 150, articles on which the equipment 150 operate, or the environment of the equipment 150. Example measurements that are monitored by various sensors 154 may include temperature,” ¶ 0040).
Regarding claim 5, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein the historical telemetry data is collected periodically and continuously (“a sensor 154 may monitor the temperature of a particular component of a piece of equipment. For every predetermined period of time (e.g., a second, a few seconds, a minute, an hour, etc.), the sensor 154 generates a data point,” ¶ 0041).
Regarding claim 6, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches processing the historical telemetry data using at least one of: transforming categorical variables, forward filling, determining relative values, or normalizing values; wherein the machine learning model executes using the processed historical telemetry data (“The predictive maintenance server 110 may normalize 550 the differences using one of the reference metrics of the machine learning model that represents the statistics of the training error of the machine learning model for a particular period of training data,” ¶ 0073).
Regarding claim 8, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches generating an anomaly alert in response to the automatically determined indication (“generating an anomaly score for the equipment based on the differences. The process may further include generating, based on anomaly score, an alert for the equipment,” ¶ 0004, “Based on the anomaly scores such as by comparing the scores to one or more threshold values or some predetermined ranges, the maintenance recommendation engine 270 may select an appropriate alert or recommendation. For example, if an overall anomaly score exceeds a predetermined threshold value, the maintenance recommendation engine 270 may recommend an inspection of the particular equipment 150. In other cases, multiple machine learning models may generate different results. The alert may specify a component of the equipment 150. The alert may also specify the significance and acuteness of the situation and include information regarding the persistence and chronicity of the condition of the equipment 150,” ¶ 0052).
Regarding claim 9, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein the machine learning model is trained s using self-supervised learning (“The machine learning models may be unsupervised or semi-supervised to reduce the cost of determining whether the equipment is abnormal during the training of the machine learning model. The machine learning models may include a predictive power parity (PPP) model, a variational auto-encoder (VAE) model, and a Bayes-based histogram model,” ¶ 0033).
Regarding claim 10, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein the machine learning model includes an autoencoder (“Deep learning techniques such as neural networks, including recurrent neural networks and long short-term memory networks, may also be used. Other machine learning techniques, such as predictive power parity (PPP), variational auto-encoder (VAE), and Bayes-based histogram, may also be used,” ¶ 0048).
Regarding claim 11, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches processing the anomaly score to predict a likelihood of an equipment failure within a threshold failure time (“The models that are trained to classify failures may estimate failure probabilities of the equipment 150 or of a particular component of the equipment 150,” ¶ 0051).
Regarding claim 12, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein providing the automatically determined indication includes outputting the automatically determined indication to a user interface of a diagnostic tool (“The alert may be sent as a message or may be displayed on a user interface, for example, the user interface of an anomaly review application 466,” ¶ 0066).
Regarding claim 13, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein providing the automatically determined indication includes outputting, on a user interface, anomaly data and refrigeration-dependent data (“The alert may be sent as a message or may be displayed on a user interface, for example, the user interface of an anomaly review application 466,” ¶ 0066).
Regarding claim 14, Trinh teaches the invention of claim 13, as set forth in the rejection of claim 13 above. Trinh also teaches wherein the refrigeration-dependent data includes work order data (“includes a small number of labels on whether a piece of equipment is normal or defective or on the repair history of the equipment,” ¶ 0067).
Regarding claim 15, Trinh teaches the invention of claim 1, as set forth in the rejection of claim 1 above. Trinh also teaches wherein the automatically determined indication is provided on a graph (“FIG. 8 illustrates an example plot of anomaly scores of a piece of equipment in a period of time that is generated by a PPP model,” ¶ 0080).
Regarding claim 16, Trinh teaches the invention of claim 15, as set forth in the rejection of claim 15 above. Trinh also teaches wherein providing the automatically determined indication includes displaying information associated with a user-selected point in time on the graph (“The user interface allows a user to select a particular time sub-interval 1930 and inspect the data 1932 for that sub-interval in detail and identify dimensions 1934 (e.g., temperature, pressure, frequency, or other sensor channels) that have significant impact on that sub-interval,” ¶ 0111).
Regarding claim 19, Trinh teaches a system, comprising:
a communication interface (“The data store 120 and the predictive maintenance server 110 may be in communication with one or more facility sites 140 through the network 130 such as the Internet,” ¶ 0037) configured to receive historical telemetry data (“receiving a set of sensor data generated from sensors associated with equipment, one of the sensors being a target sensor, the set of sensor data comprising measured values of the target sensor,” ¶ 0004; “The machine learning model may be trained by a training data set that includes historical measurements of sensor data,” ¶ 0071) of one or more refrigeration systems wherein the historical telemetry data includes measured temperature values, historical equipment failures, and setpoint temperature values (“a sensor 154 may monitor the temperature of a particular component of a piece of equipment. For every predetermined period of time (e.g., a second, a few seconds, a minute, an hour, etc.), the sensor 154 generates a data point,” ¶ 0041; “an operator may set a target temperature of a particular component or location of the equipment 150. The sensor readings may be fed back to the equipment 150, as represented by the arrows 158, to maintain the measurements of the components in the proximity of the target measurements specified in the settings. In another case, a setting value may be dynamic. For example, a target temperature of a particular component or location of the equipment 150 may be dynamically changed based on other conditions of the equipment 150. The values of the settings 152 may also be reported as a time series and transmitted to the data store 120 through the controller 160 or directly,” ¶ 0043; “The VAE model may be an unsupervised learning model that is trained based on training data that does not include labels or only includes a small number of labels on whether a piece of equipment is normal or defective or on the repair history of the equipment,” ¶ 0081);
and a processor (“Parts of the predictive maintenance server 110 may include one or more processors such as a CPU (central processing unit), a GPU (graphics processing unit), a TPU (tensor processing unit), a DSP (digital signal processor), a system on a chip (SOC), a controller, a state machine, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or any combination of these,” ¶ 0035) coupled to the communication interface and configured to:
training a machine learning model using historical telemetry data to derive an anomaly score, wherein the machine-learning model includes an encoding network and a decoding network, and wherein the machine-learning model is trained based on an evaluation between an input to the encoding network and an output from the decoding network (“The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152. The predictive maintenance server 110 may train one or more machine learning models that assign anomaly scores to a piece of equipment 150. The anomaly scores may include an overall anomaly score and individual anomaly scores each corresponding to a component, a measurement, or an aspect of the equipment 150,” ¶ 0044; “the neural network 900 may be a multi-layer neural network that might include an encoder 920, one or more bottleneck layers 930, and decoder 950,” ¶ 0083; “As another example, to train a model to perform a regression task, during the creation/training phase, one or more data elements of the training data are input to the model being trained, and the model generates output indicating a predicted value of one or more other data elements of the training data. The predicted values of the training data are compared to corresponding actual values of the training data, and the computer modifies the model until the model accurately and reliably (e.g., within some specified criteria) predicts values of the training data,” ¶ 0049);
determining that the anomaly count is greater than a threshold count value; incrementing an anomaly count in response to determining that the anomaly score is greater than a threshold (“When the anomaly scores are determined to be beyond a specific range such as above a predetermined threshold, the predictive maintenance server 110 identifies a particular facility site 140 and a particular piece of equipment 150 and provides an indication that the equipment 150 may need an inspection and possible repair. The predictive maintenance server 110 may also train additional models such as classifiers and regressors that can identify a specific component of the equipment 150 that may need an inspection, repair and/or replacement,” ¶ 0044; “The anomaly score is normalized so that the baseline anomaly score may be 1, which represents the average error in predicting the measurements of a vital using the machine learning model. A high dissimilarity metric value or anomaly score value such as a value within a range of 4 to 5 (even 8 to 9 at some points) indicates the loss of predictive power of other sensor channels to predict the values of the vitals. This indicates a time period that the equipment 150 may operate in an abnormal state. The predictive maintenance server 110 may send an alert to the facility site 140 and/or may display the alert in the front-end interface to indicate to an operator that maintenance may be needed for the equipment 150,” ¶ 0080);
providing an automatically determined indication in response to determining that the anomaly count is greater than a threshold count (“generating an anomaly score for the equipment based on the differences. The process may further include generating, based on anomaly score, an alert for the equipment,” ¶ 0004); and
validating the machine-learning model using the historical equipment failures (“The predictive maintenance server 110 may also process and validate data in a small batch for the scoring data 422 in block 440. If data validation fails, the predictive maintenance server 110 may trigger an alert in block 452 indicating potential problems might have occurred,” ¶ 0064; “After the machine learning model is trained, a series of output values (predicted values) may be generated for a historical period of time (e.g., one month). The series of predicted values of the vital may be compared to the actual values of the vital,” ¶ 0071).
Regarding claim 20, Trinh teaches a non-transitory computer readable medium storing instructions that when executed by a processor cause the processor to perform operations including:
receiving historical telemetry data associated with one or more refrigeration systems, wherein the historical telemetry data includes measured temperature values and setpoint temperature values (“receiving a set of sensor data generated from sensors associated with equipment, one of the sensors being a target sensor, the set of sensor data comprising measured values of the target sensor,” ¶ 0004; “The machine learning model may be trained by a training data set that includes historical measurements of sensor data,” ¶ 0071) of one or more refrigeration systems, including measured temperature values, historical equipment failures, and setpoint temperature values (“a sensor 154 may monitor the temperature of a particular component of a piece of equipment. For every predetermined period of time (e.g., a second, a few seconds, a minute, an hour, etc.), the sensor 154 generates a data point,” ¶ 0041; “an operator may set a target temperature of a particular component or location of the equipment 150. The sensor readings may be fed back to the equipment 150, as represented by the arrows 158, to maintain the measurements of the components in the proximity of the target measurements specified in the settings. In another case, a setting value may be dynamic. For example, a target temperature of a particular component or location of the equipment 150 may be dynamically changed based on other conditions of the equipment 150. The values of the settings 152 may also be reported as a time series and transmitted to the data store 120 through the controller 160 or directly,” ¶ 0043; “The VAE model may be an unsupervised learning model that is trained based on training data that does not include labels or only includes a small number of labels on whether a piece of equipment is normal or defective or on the repair history of the equipment,” ¶ 0081);
training, using one or more hardware processors a machine learning model using the historical telemetry data to derive an anomaly score, wherein the machine-learning model includes an encoding network and a decoding network, and wherein the machine-learning model is trained based on an evaluation between an input to the encoding network and an output from the decoding network (“The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152. The predictive maintenance server 110 may train one or more machine learning models that assign anomaly scores to a piece of equipment 150. The anomaly scores may include an overall anomaly score and individual anomaly scores each corresponding to a component, a measurement, or an aspect of the equipment 150,” ¶ 0044; “the neural network 900 may be a multi-layer neural network that might include an encoder 920, one or more bottleneck layers 930, and decoder 950,” ¶ 0083; “As another example, to train a model to perform a regression task, during the creation/training phase, one or more data elements of the training data are input to the model being trained, and the model generates output indicating a predicted value of one or more other data elements of the training data. The predicted values of the training data are compared to corresponding actual values of the training data, and the computer modifies the model until the model accurately and reliably (e.g., within some specified criteria) predicts values of the training data,” ¶ 0049);
determining that the anomaly count is greater than a threshold count value; incrementing an anomaly count in response to determining that the anomaly score is greater than a threshold (“When the anomaly scores are determined to be beyond a specific range such as above a predetermined threshold, the predictive maintenance server 110 identifies a particular facility site 140 and a particular piece of equipment 150 and provides an indication that the equipment 150 may need an inspection and possible repair. The predictive maintenance server 110 may also train additional models such as classifiers and regressors that can identify a specific component of the equipment 150 that may need an inspection, repair and/or replacement,” ¶ 0044; “The anomaly score is normalized so that the baseline anomaly score may be 1, which represents the average error in predicting the measurements of a vital using the machine learning model. A high dissimilarity metric value or anomaly score value such as a value within a range of 4 to 5 (even 8 to 9 at some points) indicates the loss of predictive power of other sensor channels to predict the values of the vitals. This indicates a time period that the equipment 150 may operate in an abnormal state. The predictive maintenance server 110 may send an alert to the facility site 140 and/or may display the alert in the front-end interface to indicate to an operator that maintenance may be needed for the equipment 150,” ¶ 0080);
providing an automatically determined indication in response to determining that the anomaly count is greater than a threshold count (“generating an anomaly score for the equipment based on the differences. The process may further include generating, based on anomaly score, an alert for the equipment,” ¶ 0004); and
validating the machine-learning model using the historical equipment failures (“The predictive maintenance server 110 may also process and validate data in a small batch for the scoring data 422 in block 440. If data validation fails, the predictive maintenance server 110 may trigger an alert in block 452 indicating potential problems might have occurred,” ¶ 0064; “After the machine learning model is trained, a series of output values (predicted values) may be generated for a historical period of time (e.g., one month). The series of predicted values of the vital may be compared to the actual values of the vital,” ¶ 0071).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LEO T HINZE whose telephone number is (571)272-2864. The examiner can normally be reached M-Th 9-2.
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, Stephen Meier can be reached on (571)272-2149. 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.
/LEO T HINZE/
Patent Examiner
AU 2853
24 June 2026
/STEPHEN D MEIER/ Supervisory Patent Examiner, Art Unit 2853
Prosecution Insights