DETAILED ACTION
This action is written in response to the remarks an amendments dated 31 March 2026. This action is made final. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
In view of the Applicant’s arguments—as well as the latest claim amendments—the Examiner withdraws all outstanding rejections under §101.
The Applicants argue that the previous art of record does not anticipate or render obvious the claims as currently amended. The Examiner provides updated prior art rejections below necessitated by the current amendments.
Subject Matter Eligibility
In determining whether the claims are subject matter eligible, the examiner has considered and applied the guidance from MPEP § 2106. The examiner finds that the combination of steps performed in the recited method (and related devices, etc.) cannot be practically performed as a mental process.
Claim Rejections - 35 USC § 112(b) - Indefiniteness
The following is a quotation of the second paragraph of 35 U.S.C. 112:
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 3, 14 and 19 are rejected under 35 U.S.C. 112(b), as being indefinite for failing to particularly point out and distinctly claim the subject matter which applicant regards as the invention.
Claim 3 recites “wherein generating one or more device-related temporal predictions comprises processing at least a portion of the obtained data using one or more decision tree-based ensemble machine learning algorithms. However, independent claim 1—from which claim 3 depends—has been amended to recite that this ML algorithm is a neural network-based regressor. Claim 3 is therefore ambiguous because it specifies two separate categories for the same model. (Neural networks (NNs) and decision trees are distinct models that can both be used for regression task.)
For the foregoing reason, the identified claim limitation is ambiguous, and consequently a person of ordinary skill would not be able to understand the scope of the claim with reasonable certainty. Therefore the claim is indefinite. Claims 14 and 19 are rejected for the same reason.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103(a) which forms the basis for all obviousness rejections set forth in this Office action:
(a) A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
The following are the references relied upon in the rejections below:
Cella (US 2019/0339686 A1)
Sardenberg (US 2023/0068328 A1)
Specht (Specht, Donald F. "A general regression neural network." IEEE transactions on neural networks 2.6 (1991): 568-576.)
Claims 1, 3-6, 9-12, 14-17 and 19-23 are rejected under 35 U.S.C. 103 as being unpatentable over Sardenberg and Specht.
Regarding claims 1, 12 and 17, Sardenberg discloses a computer-implemented method (and a non-transitory processor-readable storage medium and apparatus) comprising:
obtaining data pertaining to one or more aspects of at least one device-related repair task;
The examiner interprets ‘device’ (and ‘device-related’) according to its broadest reasonable interpretation in view of its plain meaning.1 The applicant does not define this term.
[0006] “receive, from a machine at a customer jobsite of a customer, sensor data from one or more machine sensors, wherein the sensor data indicates a machine failure; retrieve, from a network-accessible dealer site of a dealer associated with the customer, informational data about current repair technician backlog, shop availability backlog, and parts availability;”
generating one or more device-related temporal predictions associated with the at least one device-related repair task by processing at least a portion of the obtained data using at least one artificial neural network-based [model] …
[0006] “input the received sensor data and the retrieved informational data to a machine learning model for computing an estimated time to when the failed machine will be repaired;”
performing one or more automated actions based at least in part on at least a portion of the one or more device-related temporal predictions;
[0029] “In an embodiment, at step 218, the machine learning model is configured to monitor individual machine sensor information and is further configured to automatically provide a fix for a repair.” (Emphasis added.)
[0030] “In another embodiment, at step 218, the machine learning model is configured to monitor individual machine sensor information and is further configured to automatically notify the customer associated with the machine of proposed fixes (e.g., general maintenance issues) of other components of the machine. As an example, the machine learning model determines that the oil is due to be changed in one month. The machine learning model can be configured to send out a message with the corresponding information about the impending due date for the oil change to the dealer (for purposes of notifying the customer) or to the customer directly.”
Claim 6: “automatically causing the machine to be repaired in accordance with the mandatory repair.“
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.
[0047] “one or more processors”.
Specht discloses the following further limitation which Sardenberg does not disclose:
generating one or more device-related temporal predictions … using at least one artificial neural network-based regressor comprising at least one input layer, multiple hidden layers each comprising at least one activation function, and at least one output layer, wherein the at least one artificial neural network-based regressor is trained using one or more supervised learning techniques and multi-dimensional historical data pertaining …
P. 573, fig. 1 (reproduced below).
PNG
media_image1.png
616
478
media_image1.png
Greyscale
The examiner notes that every node in a neural network (NN) has its own activation function, ie this feature is inherent in the NN depicted above.
P. 573, first col., “The activation function normally used is the exponential, although all of the alternate activation functions shown in [6, table 1] also could be used here.”
P. 568, sec. II, “The regression of a dependent variable, Y, on an independent variable, X, is the computation of the most probable value of Y for each value of X based on a finite number of possibly noisy measurements of X and the associated values of Y. The variables X and Y are usually vectors.”
At the time of filing, it would have been obvious to a person of ordinary skill to apply the NN architecture and techniques disclosed by Specht to the Sardenberg system. The recited components of a NN are probably inherent in the Sardenberg system (and indeed in every NN); however, they are taught explicitly in Specht. The Examiner notes that the regression NN disclosed by Specht is applicable for any regression problem, eg repair task time estimations, as suggested by the passage below:
Abstract: “The algorithmic form can be used for any regression problem in which an assumption of linearity is not justified.”
Regarding independent claim 12, Sardenberg further discloses a non-transitory processor-readable storage medium (Claim 15: “non-transitory computer-readable media”.).
Regarding independent claim 17, Sardenberg further discloses at least one processing device comprising a processor coupled to a memory (Claim 8, “one or more memory devices”, ‘processor’).
Regarding claims 3, 14 and 19, Sardenberg discloses the further limitation wherein generating one or more device-related temporal predictions comprises processing at least a portion of the obtained data using one or more decision tree-based ensemble machine learning algorithms.
[0005] decision tree algorithm.
[0021] “random forest model”. The Examiner notes that a random forest is an ensemble of decision trees.
Regarding claims 4, 15 and 20, Sardenberg discloses the further limitation wherein performing one or more automated actions comprises automatically provisioning one or more resources in accordance with at least one of the one or more device-related temporal predictions.
[0029] “The machine learning model can be configured to send out a message with the corresponding information about the machine, the part to be replaced, and the replacement part, to an appropriate processor, which is configured to order the replacement part to be delivered to the dealer.”
Regarding claims 5 and 16, Sardenberg discloses the further limitation wherein performing one or more automated actions comprises automatically generating and outputting, to one or more entities associated with the at least one device-related repair task, one or more communications pertaining to the at least a portion of the one or more device-related temporal predictions.
[0029] “The machine learning model can be configured to send out a message with the corresponding information about the machine, the part to be replaced, and the replacement part, to an appropriate processor, which is configured to order the replacement part to be delivered to the dealer.”
Regarding claim 6, Sardenberg discloses the further limitation wherein performing one or more automated actions comprises automatically training at least a portion of the one or more artificial intelligence techniques based at least in part on feedback related to the at least a portion of the one or more device-related temporal predictions.
P. 568, sec. II, describing the (supervised) training of the NN based on labeled training data: “The regression of a dependent variable, Y, on an independent variable, X, is the computation of the most probable value of Y for each value of X based on a finite number of possibly noisy measurements of X and the associated values of Y. The variables X and Y are usually vectors.”
Regarding claim 9, Sardenberg discloses the further limitation wherein generating one or more device-related temporal predictions comprises generating at least one prediction for one or more support service implementation timelines associated with the at least one device-related repair task.
[0031] “As an example regarding parts availability, the machine learning model is configured to answer: is the part for repair available at the dealer or at a distribution center? Is the part on back-order? What would be the emergency order deliver time?“
Regarding claim 10, Sardenberg discloses the further limitation wherein obtaining data pertaining to one or more aspects of at least one device-related repair task comprises obtaining data pertaining to one or more of user information, device information, device part information, repair-related location information, device-related location information, repair task type, one or more temporal parameters associated with the at least one device-related repair task, and logistics provider information.
[0006] “receive, from a machine at a customer jobsite of a customer, sensor data from one or more machine sensors, wherein the sensor data indicates a machine failure; retrieve, from a network-accessible dealer site of a dealer associated with the customer, informational data about current repair technician backlog, shop availability backlog, and parts availability;”
Regarding claim 11, Sardenberg discloses the further limitation comprising:
training at least a portion of the one or more artificial intelligence techniques using multi-dimensional historical logistics-related data associated with one or more device-related repair tasks.
P. 568, sec. II, “The regression of a dependent variable, Y, on an independent variable, X, is the computation of the most probable value of Y for each value of X based on a finite number of possibly noisy measurements of X and the associated values of Y. The variables X and Y are usually vectors.”
Regarding claim 21, Sardenberg discloses the further limitation wherein performing one or more automated actions comprises automatically generating and outputting, to one or more entities associated with the at least one device-related repair task, one or more communications pertaining to the at least a portion of the one or more device-related temporal predictions.
[0018] “As an example, the machine learning model with the decision tree algorithm outputs, e.g., via an online notification or report, a recommendation for the customer to rent one of the available machines at the dealer, along with the estimate financial impact of the customer renting such machine. At the dealership site, one of the dealers receives the notification, reviews the notification or as an assigned dealer review the notification. Upon approving the notification, the dealer approving the notification sends, e.g., via email or via text, the notification to the customer for approval by the customer.”
Regarding claim 22, Sardenberg discloses the further limitation wherein performing one or more automated actions comprises automatically training at least a portion of the at least one artificial neural network-based regressor based at least in part on feedback related to the at least a portion of the one or more device-related temporal predictions.
[0021] neural network models. The examiner notes that every neural network is trained based on feedback regarding the difference between predicted outcomes and target outcomes. This occurs at the node level. Thus this feature is inherent in Sardenberg.
Regarding claim 23, Sardenberg discloses the further limitation wherein obtaining data pertaining to one or more aspects of at least one device-related repair task comprises obtaining data pertaining to one or more of user information, device information, device part information, repair-related location information, device-related location information, repair task type, one or more temporal parameters associated with the at least one device-related repair task, and logistics provider information.
[0021] “the machine learning model also takes as input data from external sources, such as but not limited to weather or seasonal events and holidays, and from data internal to the entity (e.g., Internal CAT data), such as but not limited to machine historical data.” (Emphasis added.)
[0024] “unique machine identifier such as the serial number (S/N)”.
[0017] discussing equipment fleet model information.
Claims 7-8 are rejected under 35 U.S.C. 103 as being unpatentable over Sardenberg/Specht and Cella.
Regarding claim 7, Cella discloses the following further limitation which Sardenberg/Specht does not disclose wherein generating one or more device-related temporal predictions comprises generating at least one prediction for a delivery timeline for at least one of a device and one or more parts thereof to at least one location associated with the at least one device-related repair task.
[0068] “In embodiments, the industrial machine predictive maintenance system further comprises a service and delivery coordination facility that receives and processes information regarding services to perform on the industrial machine based on the at least one of orders and requests for service and parts. In embodiments, the service and delivery coordination facility validates the services to perform on the industrial machine while producing a ledger of service activity and results for the industrial machine.” (Emphasis added.)
[2247] “To facilitate at least semi-automated predictive maintenance, replacement parts, service, and the like may be automatically ordered based on a result of the predictive maintenance facility 5903 indicating that some form of preventive activity is required. An automatic part/service ordering facility 5913 may be connected directly or indirectly to the user interface/control facility 5909 to enable users to approve or adjust an automated order.” (Emphasis added.)
At the time of filing, it would have been obvious to a person of ordinary skill to monitor a plurality of related parcels (as taught by Cella) using the prediction system of Sardenberg/Specht because repair work often necessitates having appropriate replacement components in stock before repair work can be completed (or often before it can begin). See eg Sardenberg at [0006] (reproduced supra, discussing parts availability). Combining these techniques would allow repair facility operators to optimize their workflow.
Regarding claim 8, Sardenberg discloses the further limitation wherein generating one or more device-related temporal predictions comprises generating at least one prediction for a … timeline for at least one of a device … in connection with the at least one device-related repair task.
[0031] “estimated repair timeline”
Sardenberg/Specht do not disclose prediction for a manufacturing processor for a device and a component parcel thereof, ie at least one prediction for a manufacturing timeline for at least one of a device and one or more parts thereof in connection with the at least one device-related repair task. However, Cella discloses this feature.
[0004] “Heavy industrial environments, such as environments for large scale manufacturing (such as manufacturing of aircraft, ships, trucks, automobiles, and large industrial machines), energy production environments (such as oil and gas plants, renewable energy environments, and others), energy extraction environments (such as mining, drilling, and the like), construction environments (such as for construction of large buildings), and others, involve highly complex machines, devices and systems and highly complex workflows, in which operators must account for a host of parameters, metrics, and the like in order to optimize design, development, deployment, and operation of different technologies in order to improve overall results.” (Emphasis added.)
[1154] “In embodiments, a third party (e.g., RMOs, manufacturers) can aggregate data at the component level, equipment level, factory/installation level and provide a statistically valid data set against which to optimize their own systems.”
The obviousness analysis of claim 7 applies equally here.
Additional Relevant Prior Art
The following references were identified by the Examiner as being relevant to the disclosed invention, but are not relied upon in any particular prior art rejection:
Figlin discloses a task repository for managing and implementing pre-defined tasks, with applications to manufacturing and product distribution. (US 2013/0339254 A1)
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Vincent Gonzales whose telephone number is (571) 270-3837. The examiner can normally be reached on Monday-Friday 7 a.m. to 4 p.m. MT. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Miranda Huang, can be reached at (571) 270-7092.
Information regarding the status of an application may be obtained from the USPTO Patent Center.
/Vincent Gonzales/Primary Examiner, Art Unit 2124
1 device, “1. An object designed and manufactured to perform one or more functions.” (The American Heritage Dictionary, https://www.ahdictionary.com/word/search.html?q=device, accessed 19 May 2026.)
Prosecution Insights