DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The information disclosure statements submitted on 9/19/2023, 5/9/2025, and 12/1/2025 were in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Claim Objections
Claims 18 and 20 are objected to because of the following informalities:
In claim 18 lines 10-11, and claim 20 lines 10-11 “the discrete Markov Chain model are in” should read “the discrete Markov Chain model is in”.
Appropriate correction is required.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention in each of these claims is directed to the abstract idea judicial exception without significantly more.
Representative independent claim 1, substantially representative of independent claims 18 and 20, recites:
“A method of performing a prognostic health analysis for an asset, the method comprising:
performing a plurality of independent stochastic simulations using transition probabilities of a discrete Markov Chain model, wherein the discrete Markov Chain model has a state space that comprises a set of asset health states and wherein each of the plurality of independent stochastic simulations simulates a future evolution in the state space of the discrete Markov Chain model over a prognostic horizon;
computing a prognostic asset health state evolution over the prognostic horizon from the plurality of independent stochastic simulations, wherein computing the prognostic asset health state evolution comprises computing a time-dependent scalar function based on probabilities that the discrete Markov Chain model is in a particular state at a particular time as determined from each of the plurality of independent stochastic simulations;
generating output based on the computed prognostic asset health state evolution; and
automatically performing an action relating to the asset based on the computed prognostic asset health state evolution.”
The claim limitations considered to fall within in the abstract idea are highlighted in bold font above and the remaining features are “additional elements.”
Step 1 of the subject matter eligibility analysis entails determining whether the claimed subject matter falls within one of the four statutory categories of patentable subject matter identified by 35 U.S.C. 101: process, machine, manufacture, or composition of matter. Claims 1 and 18 recite a method and claim 20 recites an article of manufacture and therefore all fall within a statutory category.
Step 2A, Prong One of the analysis entails determining whether the claim recites a judicial exception such as an abstract idea. Under a broadest reasonable interpretation, the highlighted portions of claim 1 fall within the abstract idea judicial exception. Specifically, under the 2019 Revised Patent Subject Matter Eligibility Guidance, the highlighted subject matter falls within the mental processes category (including an observation, evaluation, judgment, opinion) and the mathematical concepts category (mathematical relationships, mathematical formulas or equations, mathematical calculations). MPEP § 2106.04(a)(2).
The recited function “performing a prognostic health analysis for an asset,” falls within the mental processes exception because it may be performed as mental processes (evaluation, judgement, opinion).
The recited functions:
“performing a plurality of independent stochastic simulations using transition probabilities of a discrete Markov Chain model, wherein the discrete Markov Chain model has a state space that comprises a set of asset health states and wherein each of the plurality of independent stochastic simulations simulates a future evolution in the state space of the discrete Markov Chain model over a prognostic horizon,” and
“computing a prognostic asset health state evolution over the prognostic horizon from the plurality of independent stochastic simulations, wherein computing the prognostic asset health state evolution comprises computing a time-dependent scalar function based on probabilities that the discrete Markov Chain model is in a particular state at a particular time as determined from each of the plurality of independent stochastic simulations,”
are determined by the Examiner as falling within the mathematical relationships sub-category of mathematical concepts (MPEP 2106.04(a)(2)). Performing a plurality of independent stochastic simulations using transition probabilities of a discrete Markov Chain model falls within the mathematical relationships sub-category because Markov Chain modelling is fundamentally characterized by mathematical calculations/relations (statistical/probabilistic calculations). Determining a prognostic asset health state evolution over the prognostic horizon from the plurality of independent stochastic simulations by use of a time-dependent scalar function based on probabilities that the discrete Markov Chain model is in a particular state at a particular time is also fundamentally characterized by mathematical calculations/relations as disclosed by Applicant’s specification (e.g., [0192]) and therefore also constitutes mathematical relationships.
Similar limitations comprising mathematical relationships type mathematical concepts are recited by dependent claims 5 and 12.
In claim 5, computing confidence or variance information for the prognostic asset health state evolution as a function of time over the prognostic horizon from the plurality of independent stochastic simulations, wherein the output is further generated based on the confidence or variance information is implemented as mathematical (statistical) relationships of numeric values.
In claim 12, determining the transition probabilities from historical data comprising sensor data for a plurality of assets is implemented via mathematical (statistical) relationships of numeric values.
Step 2A, Prong Two of the analysis entails determining whether the claim includes additional elements that integrate the recited judicial exception into a practical application. “A claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that the claim is more than a drafting effort designed to monopolize the judicial exception” (MPEP § 2106.04(d)).
MPEP § 2106.04(d) sets forth considerations to be applied in Step 2A, Prong Two for determining whether or not a claim integrates a judicial exception into a practical application. Based on the individual and collective limitations of claims 1, 18, and 20 and applying a broadest reasonable interpretation, the most applicable of such considerations appear to include: improvements to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)); applying the judicial exception with, or by use of, a particular machine (MPEP 2106.05(b)); and effecting a transformation or reduction of a particular article to a different state or thing (MPEP 2106.05(c)).
Regarding improvements to the functioning of a computer or other technology such as a signal processing device, none of the “additional elements” such as “generating output based on the computed prognostic asset health state evolution” and “automatically performing an action relating to the asset based on the computed prognostic asset health state evolution,” in any combination appear to integrate the abstract idea in a manner that technologically improves any aspect of a device or system that may be used to implement the highlighted steps or a device for implementing the highlighted steps such as a computer. Neither of these elements appear to relate to the abstract idea in particularized manner that functionally results in the combination of either or both of these elements with the elements falling within the judicial exception constituting an improvement in a technology field. Instead, these elements represent insignificant post-solution activity.
Regarding application of the judicial exception with, or by use of, a particular machine, the additional elements “generating output based on the computed prognostic asset health state evolution,” and “automatically performing an action relating to the asset based on the computed prognostic asset health state evolution” are implemented in a general rather than a particularized manner of implementing and utilizing health state evaluation for an asset and therefore constitute extra solution activity.
Regarding a transformation or reduction of a particular article to a different state or thing, none of claims 1, 5, 12, 18, and 20 includes any such transformation or reduction. Instead, each of the claims as a whole entails applying standard processing techniques (Markov Chain numerical modeling and a scalar function via generic computer processing) to the information to obtain asset health parameter information with the additional elements failing to provide a meaningful integration of the abstract idea in an application that transforms an article to a different state. The additional elements therefore represent extra-solution activity that do not integrate the judicial exception into a practical application. In view of the various considerations encompassed by the Step 2A, Prong Two analysis, none of claims 1, 5, 12, 18, and 20 include additional elements that integrate the recited abstract idea into a practical application.
Therefore, claims 1, 5, 12, 18, and 20 are directed to a judicial exception and require further analysis under Step 2B.
Regarding Step 2B, and as set forth in the Step 2A Prong 2 analysis, the additional elements in claims 1, 18, 22, and 25 constitute extra solution activity and therefore fail to result in the claim as a whole amounting to significantly more than the judicial exception as well as failing to integrate the judicial exception into a practical application. Furthermore, the additional elements appear to be generic and well understood as evidenced by the disclosures of Kirschnick (US 2017/0236064 A1) (disclosing “computing” of a prognostic asset health state in [0100]-[0101], [0105]-[0106] and automatically performing an action relating to the asset based on a malfunction in [0107]), and Mazzaro (US 2015/0106058 A1) (disclosing “automatically performing an action relating to the asset based on a computed health state in [0044]; FIG. 9 blocks 272 and 276, [0074] and [0076]). Therefore, the additional elements are insufficient for the claims to amount to significantly more than the judicial exception.
Independent claims 1, 18, and 20 and dependent claims 5 and 12 are therefore not patent eligible under 101.
Claims 2-4, 6-11, and 13-17 depending from claim 1, and claim 19 depending from claim 18 provide additional features/steps which are part of an expanded algorithm that includes the abstract idea (Step 2A, Prong One). None of dependent claims 2-4, 6-11, 13-17, and 19 recite additional elements that integrate the abstract idea into practical application (Step 2A, Prong Two), and all fail the “significantly more” test under the step 2B for substantially similar reasons as discussed with regards to the independent claims.
For example, claim 2 recites that the “asset” is a power system asset or industrial asset which only characterizes the source/nature of the data and therefore does not represent an additional element.
Similarly, claim 3 only characterizes the nature of the output data in terms of being either an RUL or PoF, and therefore also fails to include a meaningful additional element.
Claim 4 recites that computing the prognostic asset health state evolution comprises computing a remaining useful life, which may be performed via mental processes (e.g., evaluation of data relating to equipment health and judgement in determining remaining life) and therefore falls within the mental processes judicial exception.
Claims 6-8 further characterizes the nature/content of the output information and therefore includes no further significant additional elements.
Claim 9 further characterizes the nature of the state space information utilized for/in the Markov modeling and therefore falls within the same judicial exception (mathematical concepts).
Claim 10 recites that the health state evolution calculation includes “computing, for a plurality of times within the prognostic horizon, a probability distribution in the state space,” which falls within the mathematical relations sub-category of the mathematical concepts exception because generating a probability distribution is fundamentally characterized by mathematical calculations/relations (e.g., summation as described in Applicant’s specification). Claims 10 further recites “mapping the probability distribution to a scalar,” which falls within the mathematical concepts exception for the same reasons as “computing a time-dependent scalar function based on probabilities” in claim 1.
Claim 11 recites that the state health evolution is “obtained as a time evolution of the scalar,” which falls within the mathematical concepts exception for the same reasons as “computing a time-dependent scalar function based on probabilities” in claim 1.
Claim 13 recites that determining the transition probabilities comprises “computing the time-dependent scalar function from sensor data for the plurality of assets,” “identifying transitions within the state space of the discrete Markov Chain model based on the time-dependent scalar function,” and “computing the transition probabilities within the state space of the discrete Markov Chain model.” Computing the time-dependent scalar function from sensor data for the plurality of assets falls within the mathematical concepts exception because as disclosed in Applicant’s specification such computation of the function entails mathematical processing of sensor data in the Markov Chain modeling to determine probabilities and then processes that data via a scalar function (e.g., summation). Identifying transitions within the state space of the discrete Markov Chain model based on the time-dependent scalar function, and computing the transition probabilities within the state space of the discrete Markov Chain model each/both fall within the mathematical concepts exception because each is performed as part of the statistical/probabilistic processing of the Markov modeling.
Claim 14 characterizes the independent stochastic simulations as being Markov Chain Monte Carlo simulations, which are numerical processing algorithms that are fundamentally characterizes by mathematical calculations/relations and therefore fall within the mathematical concepts exception.
Claim 15 recites “receiving sensor measurement data captured during operation of the asset,” which represents high-level data collection using ordinary means and therefore constitutes extra solution activity that neither integrates the judicial exception into a practical application nor results in the claim as a whole amounting to significantly more than the judicial exception. Claim 15 further recites “updating the prognostic asset health state evolution based on the received sensor measurement data,” which is an extension of the “computing a prognostic asset health state evolution” step in claim 1 and therefore falls within the same judicial exception.
Claim 16 recites that the plurality of simulations comprise simulations for different ambient or operating scenarios, which is an extension of the “performing a plurality of independent stochastic simulations” step in claim 1 and therefore falls within the same judicial exception.
Claim 17 further characterizes the “asset” as being a power transformer, a distributed energy resource, DER, unit, or a power generator, which only characterizes the source/nature of the data and therefore does not represent an additional element.
Claim 19 (and the alternative element in claim 17) recites that the prognostic horizon is at least one year, which is a modeling/distribution parameter such that this element is an extension of the “computing a prognostic asset health state evolution” element in claim 18 and falls within the same judicial exception.
Dependent claims 2-4, 6-11, 13-17, and 19 therefore also constitute ineligible subject matter under 101.
Claim Rejections - 35 USC § 103
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 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 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 1-4, 10-12, and 15-17 are rejected under 35 U.S.C. 103 as being unpatentable over Kirschnick (US 2017/0236064 A1) as provided by Applicant, in view of Mazzaro (US 2015/0106058 A1) as provided by Applicant, and in further view of Tashman (US 2020/0265331 A1).
As to claim 1, Kirschnick teaches “[a] method of performing a prognostic health analysis for an asset ([0009]; [0105]; Abstract), the method comprising:
performing a plurality of independent stochastic simulations ([0100]-[0101], multiple Markov chains) using transition probabilities of a discrete Markov Chain model ([0100]-[0101]), wherein the discrete Markov Chain model has a state space that comprises a set of asset health states ([0095] parameter state spaces for determining malfunction probabilities) and wherein each of the plurality of independent stochastic simulations simulates a future evolution in the state space of the discrete Markov Chain model over a prognostic horizon ([0094], [0101] model predicts malfunction over time);
computing a prognostic asset health state evolution over the prognostic horizon from the plurality of independent stochastic simulations ([0100]-[0101], [0105]-[0106] malfunction probability determined over time via multiple Markov chains),” wherein computing the prognostic asset health state evolution comprises computing a time-dependent” “function ([0101]-[0102] probability function utilized for generating time-dependent probabilities of a health state (malfunction at a future point in time); [0115] apparatus 10 can determine the probability of the single malfunction Mk at the future point in time given the current value state C(t0) of the parameter for different future points in time, allowing for display of the probability over a future time horizon) based on probabilities that the discrete Markov Chain model is in a particular state at a particular time ([0101] probability formula indicates probably of malfunction M as a function of time) as determined from each of the plurality of independent stochastic simulations ([0101] probability distribution over time determined by a stochastic model that combines the Markov chains);
“generating output based on the computed prognostic asset health state evolution ([0103] output prepared to show malfunction prediction; FIGS. 6a and 6b); and
automatically performing an action relating to the asset ([0107] disclosing actions performed in response to malfunction).”
Kirschnick teaches performance of protective actions performed in response to malfunctions but does not explicitly teach automatically performing an action relating to the asset “based on the computed prognostic asset health state evolution.”
Mazarro discloses a system/method for reliability operations that includes automatically performing an action relating to an asset based on computed prognostic asset health state information ([0044]; FIG. 9 blocks 272 and 276, [0074] and [0076]).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Mazarro’s teaching of automatically performing an action relating to an asset based on prognostic health state information with the method disclosed by Kirschnick that includes generating a prognostic health state evolution.
The motivation would have been to protect and/or optimize operation of a device/component using information indicative of degraded or potentially degraded performance as suggested by Mazarro.
Regarding “computing a time-dependent scalar function” Kirschnick’s disclosure cited above in [0101]-[0102] relates to a vector function at a given time step ([0115] for multiple time steps) and therefore does not appear to teach that the time-dependent function is a time-dependent “scalar” function.
Tashman discloses a method for predicting equipment failure (Abstract) in which Markov modeling is used for determining equipment states (([0041] and [0045]-[0046] machine learning model for determining states may be Markov model; FIG. 7 depicting Markov-type machine learning model that models state probabilities including degraded states and failure) and in which a scalar function is used to sample the Markov model to generate state evolution based on probabilities that the model is in a particular state at a particular time ([0041] and [0045]-[0046] machine learning model for determining states may be Markov model; FIG. 7 depicting Markov-type machine learning model that models state probabilities including degraded states and failure; FIG. 8 depicting failure probability graph 806 representing a probability over time function (scalar function because single dimension output); [0059]-[0061] failure graph function generated for posterior observation; [0062] same failure probability time distribution may be applied for forward (future) probabilities).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Tashman’s teaching of using a scalar function (simple probability output) as part of computing a prognostic health state evolution based on Markov states to the method taught by Kirschnick as modified by Mazzaro in which a vectorized functions are used for determining time evolution distributions based on multiple independent stochastic simulations such that in combination the method includes computing a time-dependent “scalar” function, in addition and/or as an alternative to the vectorized function taught by Kirschnick, based on probabilities that the discrete Markov Chain model is in a particular state at a particular time as determined from each of the plurality of independent stochastic simulations to be used for computing a prognostic asset health state evolution over time.
Such a combination would amount to selecting a known design option for characterizing asset health probabilities over a time horizon using a scalar function in addition and/or as an alternative to the vector function processing specifically disclosed by Kirschnick to achieve predictable results.
As to claim 2, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, wherein the asset is a power system asset or an industrial asset (Kirschnick: [0090]).
As to claim 3, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, wherein generating the output comprises generating an output related to a remaining useful life (RUL) (Kirschnick: [0058] and [0115]) or a probability of failure (PoF).”
As to claim 4, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, wherein computing the prognostic asset health state evolution comprises computing a remaining useful life (Kirschnick: [0058] and [0115]).”
As to claim 10, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, wherein computing the prognostic asset health state evolution comprises computing, for a plurality of times within the prognostic horizon, a probability distribution in the state space and mapping the probability distribution to a scalar (Kirschnick: [0011]-[0015] and [0101]-[0102], determining and mapping probability distribution functions to determine vector malfunction probabilities. As combined with Tashman per the grounds for rejecting claim 1, the probability distribution values are mapped to a scalar).”
As to claim 11, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 10, wherein the prognostic asset health state evolution is obtained as a time evolution of the scalar (Kirschnick: [0058] and [0114]-[0115] malfunction probability determined for multiple points in time. As combined with Tashman per the grounds for rejecting claim 1, the state evolution is obtained as a time evolution of a scalar).”
As to claim 12, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, further comprising determining the transition probabilities from historical data comprising sensor data for a plurality of assets (Kirschnick: FIG. 7 steps s24-s27, [0129]-[0130], parameters for determining transition matrix obtained from historical component data; [0038] and [0104] parameters may be obtain via sensors).”
As to claim 15, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, further comprising:
receiving sensor measurement data captured during operation of the asset (Kirschnick: [0038] and [0104] parameters may be obtained via sensors); and
updating the prognostic asset health state evolution based on the received sensor measurement data (Kirschnick: [0096] transition matrix is updated periodically on the basis of new parameter values).
As to claim 16, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, wherein the plurality of simulations comprise simulations for different ambient or operating scenarios (Kirschnick: [0050]-[0052], [0066]-[0068] transition matrices correspond via parameters to multiple “single malfunctions”; [0092]).”
As to claim 17, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, wherein:
the asset is a power transformer, a distributed energy resource, DER, unit, or a power generator (Kirschnick: [0090]); or
the prognostic horizon is 1 year or more.
Claims 5-8 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Kirschnick in view of Mazzaro and Tashman as applied to claims 1 and 12 above, and further in view of Tobon-Mejia, D.A. “Hidden Markov Models for failure diagnostic and prognostic,” PROGNOSTICS AND SYSTEM HEALTH MANAGEMENT CONFERENCE (PHM-SHENZHEN), 2011, IEEE, 24 May 2011, pages 1-8, as provided by Applicant.
As to claim 5, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1, further comprising computing” “information for the prognostic asset health state evolution as a function of time over the PG-P190777USO1-36-prognostic horizon from the plurality of independent stochastic simulations (Kirschnick: [0094], [0101] model predicts malfunction over time).”
None of Kirschnick, Mazarro, and Tashman appear to expressly teach computing “confidence or variance” information or “wherein the output is further generated based on the confidence or variance information.”
Tobon-Mejia teaches computing “confidence or variance” information for the prognostic asset health state evolution (page 1, I. Introduction paragraph beginning with “In this paper, a failure prognostic method”) and “wherein the output is further generated based on the confidence or variance information (page 1, I. Introduction paragraph beginning with “In this paper, a failure prognostic method” confidence value is calculated and used with RUL in the decision process).”
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have combined Tobon-Mejia’s teaching of computing confidence information for the health state evolution to system and method disclosed by Kirschnick as modified by Mazarro and Tashman because computing such confidence information could be utilized as a threshold such as to determine whether corrective action is required and/or could be utilized to adjust the health state evolution methodology to increase likelihood of a correct prediction of asset health.
As to claim 6, the combination of Kirschnick, Mazarro, Tashman, and Tobon-Mejia teaches “[t]he method of claim 5, wherein the output is further generated based on the confidence information (Tobon-Mejia: page 1, I. Introduction paragraph beginning with “In this paper, a failure prognostic method” confidence value is calculated and used with RUL in the decision process) and wherein the confidence information comprises a future evolution of a confidence interval over the prognostic horizon (Tobon-Mejia: page 5, C. The MoG-HMM based method, paragraph beginning with “e) Estimation of the RUL”).
As to claim 7, the combination of Kirschnick, Mazarro, Tashman, and Tobon-Mejia teaches “[t]he method of claim 5, wherein the output is further generated based on the variance information (Tobon-Mejia: page 3, III. Hidden Markov Models, paragraph beginning with “In Equation (1), O is the observation vector…”, covariance matrix and standard deviation) and wherein the variance information comprises a future evolution of a variance over the prognostic horizon (Tobon-Mejia: page 4, IV. Failure Prognostic and Diagnostic Methods Based on HMMS, paragraph beginning with “Finally, by assuming that”).”
As to claim 8, the combination of Kirschnick, Mazarro, Tashman, and Tobon-Mejia teaches “[t]he method of claim 5, wherein the confidence or variance information comprises a time evolution of a lower boundary and a time evolution of an upper boundary (Tobon-Mejia: page 7, Figure 11 showing time evolution of upper RUL and lower RUL), the lower boundary being associated with a first set of transition probabilities and the upper boundary being associated with a second set of transition probabilities different from the first set of transition probabilities (Tobon-Mejia: page 7, Figure 11 upper and lower RULs having different failure times (dependent axis) reflecting different transition probabilities).”
As to claim 13, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 12,” but does not expressly disclose “wherein determining the transition probabilities comprises:
computing a time-dependent scalar function from sensor data for the plurality of assets,
identifying transitions within the state space of the discrete Markov Chain model based on the time-dependent scalar function, and
computing the transition probabilities based on the transitions within the state space of the discrete Markov Chain model.”
Tobon-Mejia teaches “wherein determining the transition probabilities comprises:
computing a time-dependent scalar function from sensor data for the plurality of assets (Abstract and page 2, II. Failure Prognostic: Definition and Taxonomy, paragraph beginning with “2) Data-driven prognostic” sensor data used to extract features used to build the model configured to determine states; page 3, A. The HMMs case, paragraph beginning with “In a second step,” Viterbi algorithm used to determine state sequences over time),
identifying transitions within the state space of the discrete Markov Chain model based on the time-dependent scalar function (page 3, A. The HMMs case, paragraph beginning with “In a second step,” Viterbi algorithm used to determine state sequences over time; page 3, B. The HSMMs case, paragraph beginning with “1) Learning phase: similarly,” model parameters determined using Baum-Welch algorithm and Viterbi algorithm), and
computing the transition probabilities based on the transitions within the state space of the discrete Markov Chain model (page 3, B. The HSMMs case, paragraph beginning with “1) Learning phase: similarly,” model parameters (e.g., transition probabilities) determined using Baum-Welch algorithm and Viterbi algorithm).”
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Tobon-Mejia’s teaching of determining the transition probabilities by computing a time-dependent scalar function from sensor data for the plurality of assets, identifying transitions within the state space of the discrete Markov Chain model based on the time-dependent scalar function, and computing the transition probabilities based on the transitions within the state space of the discrete Markov Chain model to the method disclosed by Kirschnick as modified by Tashman and Mazarro.
As disclosed by Tobon-Mejia, Viterbi and Baum-Welch algorithms are standard algorithms used for training Markov models including determining transition probabilities and therefore such a combination would amount to selecting a known design option for training Markov models including determining transition probabilities to achieve predicable results.
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Kirschnick in view of Mazzaro and Tashman as applied to claim 1, and in further view of Zhang, Xiaodong “An Integrated Approach to Bearing Fault Diagnostics and Prognostics,” PROCEEDINGS OF AMERICAN CONTROL CONFERENCE, IEEE, 8 June 2005, pages 2750-2755, as provided by Applicant.
As to claim 9, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1,” and teaches a state space having multiple states, but does not explicitly teach “wherein the state space comprises:
at least one state in which operation of the asset is not adversely affected by a failure;
at least one state in which operation of the asset is adversely affected by a failure, but the asset continues to operate; and
a state in which the asset is inoperative due to a failure.”
Zhang teaches “wherein the state space comprises:
at least one state in which operation of the asset is not adversely affected by a failure (page 2752, B. HMM-based Diagnostics/Prognostics, Figure 3 and paragraph beginning with “A block diagram of the proposed,” Initial state “Normal”);
at least one state in which operation of the asset is adversely affected by a failure, but the asset continues to operate (page 2752, B. HMM-based Diagnostics/Prognostics, Figure 3 and paragraph beginning with “A block diagram of the proposed,” Degraded states “nick,” “scratches,” and “more nicks”); and
a state in which the asset is inoperative due to a failure (page 2752, B. HMM-based Diagnostics/Prognostics, Figure 3 and paragraph beginning with “A block diagram of the proposed,” Terminal state “Failure”).”
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Zhang’s teaching of using a state space that uses states including operation of the asset is not adversely affected by a failure, operation of the asset is adversely affected by a failure, but the asset continues to operate, and the asset is inoperative due to a failure to the method disclosed by Kirschnick as modified by Mazarro and Tashman.
The motivation would have been to encompass the full range of asset operability in the malfunction prediction.
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Kirschnick in view of Mazarro and Tashman as applied to claim 1, and in further view of Chen (US 2017/0039479 A1).
As to claim 14, the combination of Kirschnick, Mazarro, and Tashman teaches “[t]he method of claim 1,” and in a similar context as Kirschnick, Mazarro and Tashman, Chen teaches “wherein the plurality of independent stochastic simulations are Markov Chain Monte Carlo simulations ([0030]).”
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have combined the teachings of Chen with the teachings of Kirschnick as modified by Mazarro and Tashman such that the Markov modeling system disclosed by Kirschnick utilizes Markov Chain Monte Carlo processing including comprising a plurality of independent stochastic simulations that are Markov Chain Monte Carlo simulations.
The motivation would have been to leverage the autocorrelation features of Markov Chain Monte Carlo simulations to achieve convergence among multiple chains more efficiently as disclosed by Chen.
Claims 18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kirschnick (US 2017/0236064 A1) as provided by Applicant, in view of Mazzaro (US 2015/0106058 A1) as provided by Applicant.
As to claim 18, Kirschnick teaches “[a] method of performing a prognostic health analysis for an asset ([0009]; [0105]; Abstract), the method comprising:
performing a plurality of independent stochastic simulations ([0100]-[0101], multiple Markov chains) using transition probabilities of a discrete Markov Chain model ([0100]-[0101]), wherein the discrete Markov Chain model has a state space that comprises a set of asset health states ([0095] parameter state spaces for determining malfunction probabilities) and wherein each of the plurality of independent stochastic simulations simulates a future evolution in the state space of the discrete Markov Chain model over a prognostic horizon ([0094], [0101] model predicts malfunction over time);
computing a prognostic asset health state evolution over the prognostic horizon from the plurality of independent stochastic simulations ([0100]-[0101], [0105]-[0106] malfunction probability determined over time via multiple Markov chains), wherein computing the prognostic asset health state evolution comprises computing a function ([0101]-[0102] probability function utilized for generating time-dependent probabilities of a health state (malfunction at a future point in time); [0115] apparatus 10 can determine the probability of the single malfunction Mk at the future point in time given the current value state C(t0) of the parameter for different future points in time, allowing for display of the probability over a future time horizon) based on probabilities that the discrete Markov Chain model are in particular states during the prognostic horizon ([0101] probability formula indicates probably of malfunction M as a function over time) as determined from each of the plurality of independent stochastic simulations ([0101] probability distribution over time determined by a stochastic model that combines the Markov chains);
generating output based on the computed prognostic asset health state evolution ([0103] output prepared to show malfunction prediction; FIGS. 6a and 6b); and
automatically performing an action ([0107] disclosing actions performed in response to malfunction).”
Kirschnick teaches performance of protective actions performed in response to malfunctions but does not explicitly teach automatically performing an action relating to the asset “based on the computed prognostic asset health state evolution.”
Mazarro discloses a system/method for reliability operations that includes automatically performing an action relating to an asset based on computed prognostic asset health state information ([0044]; FIG. 9 blocks 272 and 276, [0074] and [0076]).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Mazarro’s teaching of automatically performing an action relating to an asset based on prognostic health state information with the method disclosed by Kirschnick that includes generating a prognostic health state evolution.
The motivation would have been to protect and/or optimize operation of a device/component using information indicative of degraded or potentially degraded performance as suggested by Mazarro.
As to claim 20, Kirschnick teaches “[a] non-transitory computer readable medium with instructions stored thereon ([0017] computer program (instructions) used to implement invention (inherently requires non-transitory computer readable medium (memory); claim 10), wherein, when executed by a processor (claim 10), the instructions enable the processor to:
perform a plurality of independent stochastic simulations ([0100]-[0101], multiple Markov chains) using transition probabilities of a discrete Markov Chain model ([0100]-[0101]), wherein the discrete Markov Chain model has a state space that comprises a set of asset health states ([0095] parameter state spaces for determining malfunction probabilities) and wherein each of the plurality of independent stochastic simulations simulates a future evolution in the state space of the discrete Markov Chain model over a prognostic horizon ([0094], [0101] model predicts malfunction over time);
compute a prognostic asset health state evolution over the prognostic horizon from the plurality of independent stochastic simulations ([0100]-[0101], [0105]-[0106] malfunction probability determined over time via multiple Markov chains), wherein computing the prognostic asset health state evolution comprises computing a function ([0101]-[0102] probability function utilized for generating time-dependent probabilities of a health state (malfunction at a future point in time); [0115] apparatus 10 can determine the probability of the single malfunction Mk at the future point in time given the current value state C(t0) of the parameter for different future points in time, allowing for display of the probability over a future time horizon) based on probabilities that the discrete Markov Chain model are in particular states during the prognostic horizon ([0101] probability formula indicates probably of malfunction M as a function over time) as determined from each of the plurality of independent stochastic simulations ([0101] probability distribution over time determined by a stochastic model that combines the Markov chains);
generating output based on the computed prognostic asset health state evolution ([0103] output prepared to show malfunction prediction; FIGS. 6a and 6b); and
automatically performing an action ([0107] disclosing actions performed in response to malfunction).”
Kirschnick teaches performance of protective actions performed in response to malfunctions but does not explicitly teach automatically performing an action relating to the asset “based on the computed prognostic asset health state evolution.”
Mazarro discloses a system/method for reliability operations that includes automatically performing an action relating to an asset based on computed prognostic asset health state information ([0044]; FIG. 9 blocks 272 and 276, [0074] and [0076]).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Mazarro’s teaching of automatically performing an action relating to an asset based on prognostic health state information with the method disclosed by Kirschnick that includes generating a prognostic health state evolution.
The motivation would have been to protect and/or optimize operation of a device/component using information indicative of degraded or potentially degraded performance as suggested by Mazarro.
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Kirschnick in view of Mazzaro as applied to claim 18 above, and further in view of Lee (US 2021/0065086 A1).
As to claim 19, the combination of Kirschnick and Mazzaro teaches “[t]he method of claim 18,” but each of Kirschnick and Mazzaro are silent regarding length of a prognostic horizon for the monitored equipment and therefore neither expressly teaches “wherein the prognostic horizon is at least one year.”
Setting a prognostic horizon for a range of at least one year for equipment monitoring was known in the art prior to the effective filing date. For example, Lee discloses a method for implementing equipment failure curve analysis (Abstract) that includes predicting equipment failure ([0057], [0073]), and includes setting an equipment monitoring prognostic horizon for at least one year (FIG. 6 depicting GUI 600 set to display failure probabilities over a span of 500 days, [0053]).
It would have been obvious to one of ordinary skill in the art before the effective filing date, to have applied Lee’s teaching of setting an equipment monitoring prognostic horizon for at least one year to the method taught by Kirschnick as modified by Mazzaro, such that in combination the prognostic horizon over which the states of equipment are determined is at least one year.
The motivation would have been to use a time horizon/window having a sufficient timespan so as to ascertain potential degradation issues that may evolve gradually as suggested by Lee.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEW W BACA whose telephone number is (571)272-2507. The examiner can normally be reached Monday - Friday 8:00 am - 5:30 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, Andrew Schechter can be reached at (571) 272-2302. 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.
/MATTHEW W. BACA/Examiner, Art Unit 2857
/ANDREW SCHECHTER/Supervisory Patent Examiner, Art Unit 2857