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 .
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(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.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1, 8, 15 recite the limitation "causing the trained neural network to generate a predicted sensor value dataset for a second set of sensors in the second signaling channel;". There is insufficient antecedent basis for this limitation in the claim.
Any claim not specifically addressed, above, is being rejected as incorporating the deficiencies of a claim upon which it depends.
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.
Claim(s) 1, 3, 5-8, 10, 12-15, 17, 19-20 is/are rejected under 35 U.S.C. 102a1 as being anticipated by Trihn et al (US 20200379454 A1)
1. A system for machine-learning (ML)-assisted failover of sensor signaling channels for industrial machines ([0002]-[0003]), the system comprising:
a set of condition monitoring sensors (Fig. 1-154, [0040] 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);
at least one processor ([0035]);
at least one memory ([0035]); and
one or more non-transitory, computer-readable storage media storing instructions ([0035]), which, when executed by the at least one processor, cause the system to:
using a first training dataset, train a neural network to generate a set of sensor value predictions, ([0044] The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152…[0082] The predictive maintenance server 110 may receive a set of scoring sensor data of a piece of equipment 150 and use the set as input data 910. The predictive maintenance server 110 may provide the input data 910 to a neural network 900 to generate an output sample distribution 980 that can be used to determine the likelihood of observing the input data 910 based on a probability density 990...[0071] The machine learning model may be trained by a training data set that includes historical measurements of sensor data X.sub.1, X.sub.2, . . . , X.sub.i−1, X.sub.i+1, . . . , X.sub.n. …The machine learning model may use a first subset of training sensor data that excludes the historical vital measurements to train the model to predict the historical measurement. ... 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.), wherein the first training dataset relates to the set of condition monitoring sensors and includes a set of input values associatively linked to signaling channel identifiers ([0040] 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, … In one embodiment, each sensor 154 may have a unique identifier. [0042] A piece of equipment 150 may include one or more sensors 154. Various sensors 154 of a piece of equipment 150 may perform different measurements. For example, a first sensor 154 may measure the temperature of a first component, a second sensor 154 may measure the temperature of a second component, and a third sensor 154 may measure the current of a third component. An output of a sensor 154 may be referred to as a sensor channel or simply a channel. [0090] …FIG. 10 is a flowchart depicting an example process of determining anomaly of a piece of equipment, according to an embodiment. The predictive maintenance server 110 may receive 1010 a set of scoring sensor data from a piece of equipment 150. The set of scoring sensor data may include data from one or more sensor channels. The predictive maintenance server 110 may input 1020 the set of scoring sensor data into a neural network that includes one or more bottleneck layers such as the neural network 900.... Examiner Note: since each sensor has sensor channels, sensor identifier reads on “signaling channel identifiers”));
using a first set of sensors in the set of condition monitoring sensors, generate an input sensor value dataset (0044] The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152…[0082] The predictive maintenance server 110 may receive a set of scoring sensor data of a piece of equipment 150 and use the set as input data 910. The predictive maintenance server 110 may provide the input data 910 to a neural network 900 to generate an output sample distribution 980 that can be used to determine the likelihood of observing the input data 910 based on a probability density 990...[0070] The predictive maintenance server 110 may select a subset of sensor channels as vitals of the equipment 150. A vital, which may also be referred to as a target sensor, may be a sensor whose measurements are going to be predicted by one or more PPP sub-models and compared to the actual measurements to determine the anomaly of the equipment 150. The PPP model may use one or more vitals (sensor channels) to predict anomaly. For a given vital, the predictive maintenance server 110 may divide 520 the set of scoring sensor data into a first subset of scoring sensor data and a second subset of scoring sensor data that includes actual measurements of the vital. The actual measurements of a vital may be referred to as measured values of a target sensor. A series of measurements of a particular vital may be denoted X.sub.i while measurements of other sensor channels may be denoted as X.sub.1, X.sub.2, . . . , X.sub.i−1, X.sub.i+1, . . . , X.sub.n. The first subset of scoring sensor data may include the measurements X.sub.1, X.sub.2, . . . , X.sub.i−1, X.sub.i+1, . . . , X.sub.n but excludes the vital that is being predicted, while the second subset of scoring sensor data may include measurements of the vital such as X.sub.i. The predictive maintenance server 110 may input 530 the first subset of scoring sensor data to a machine learning model to predict the values of the vital measurements. The machine learning model may be one of the PPP sub-models. [0082] The predictive maintenance server 110 may receive a set of scoring sensor data of a piece of equipment 150 and use the set as input data 910. The predictive maintenance server 110 may provide the input data 910 to a neural network 900 to generate an output sample distribution 980 that can be used to determine the likelihood of observing the input data 910 based on a probability density 990...), wherein the first set of sensors is associated with a first signaling channel ([0090] …FIG. 10 is a flowchart depicting an example process of determining anomaly of a piece of equipment, according to an embodiment. The predictive maintenance server 110 may receive 1010 a set of scoring sensor data from a piece of equipment 150. The set of scoring sensor data may include data from one or more sensor channels. The predictive maintenance server 110 may input 1020 the set of scoring sensor data into a neural network that includes one or more bottleneck layers such as the neural network 900.... Examiner Note: the one or more sensor channel data reads on “first signaling channel”);
using the input sensor value dataset associated with the first signaling channel, cause the trained neural network to generate a predicted sensor value dataset for a second set of sensors in the second signaling channel ([0068] In a piece of normal equipment, the operation of the equipment should include sufficient stability and pattern. A subset of measurements of the sensors of various components of a piece of equipment 150 should predict fairly well another subset of measurements if the equipment is operating normally. Put differently, the measurements of different sensors of a piece of normal equipment 150 may often have a correlation with each other. Likewise, in some cases, when several pieces of equipment 150 are connected or otherwise related to each other, the measurements of the sensors of those pieces of equipment may also have a correlation. According to an embodiment, a PPP model may include a plurality of sub-models that are trained to use a first set of measurements to predict a second set of one or more measurements. The PPP model detects instances or periods when the correlation among different measurements of sensor data breaks down. An overall anomaly score may be generated based on the comparison between the predicted measurements and the actual measurements of sensor data. [0074] The machine learning model may be trained for a particular vital (sensor channel). According to an embodiment, the PPP model may determine the anomaly of a piece of equipment 150 using one or more vitals. For the PPP model that uses more than one vital, multiple machine learning models may be trained. Each machine learning model may be specific to a particular vital. The predictive maintenance server 110 may repeat the process described in steps 520 to 550 for different vitals, as indicated by the arrow 555. For example, for a second vital, a second machine learning model may be used. The predictive maintenance server 110 may divide the set of scoring sensor data into a different way compared to the first vital. For the second vital, the first subset of scoring sensor data may exclude the measurements of the second vital but include the measurements of the first vital. For the second vital, the second subset of scoring sensor data includes the measurements of the second vital. One or more dissimilarity metrics for the second vital may also be generated based on the differences of pairs between a predicted value and an actual value. Since the dissimilarity metrics may be normalized, the dissimilarity metrics can be compared with those corresponding to the first vital. Similar processes may be performed for a third vital, a fourth vital, etc. [0068] In a piece of normal equipment, the operation of the equipment should include sufficient stability and pattern. A subset of measurements of the sensors of various components of a piece of equipment 150 should predict fairly well another subset of measurements if the equipment is operating normally. Put differently, the measurements of different sensors of a piece of normal equipment 150 may often have a correlation with each other. Likewise, in some cases, when several pieces of equipment 150 are connected or otherwise related to each other, the measurements of the sensors of those pieces of equipment may also have a correlation. According to an embodiment, a PPP model may include a plurality of sub-models that are trained to use a first set of measurements to predict a second set of one or more measurements. The PPP model detects instances or periods when the correlation among different measurements of sensor data breaks down. An overall anomaly score may be generated based on the comparison between the predicted measurements and the actual measurements of sensor data.);
using the predicted sensor value dataset, generate a prediction for an operating condition associated with the second signaling channel (Fig. 5-560, [0074] The machine learning model may be trained for a particular vital (sensor channel). According to an embodiment, the PPP model may determine the anomaly of a piece of equipment 150 using one or more vitals. For the PPP model that uses more than one vital, multiple machine learning models may be trained. Each machine learning model may be specific to a particular vital. The predictive maintenance server 110 may repeat the process described in steps 520 to 550 for different vitals, as indicated by the arrow 555. For example, for a second vital, a second machine learning model may be used. The predictive maintenance server 110 may divide the set of scoring sensor data into a different way compared to the first vital. For the second vital, the first subset of scoring sensor data may exclude the measurements of the second vital but include the measurements of the first vital. For the second vital, the second subset of scoring sensor data includes the measurements of the second vital. One or more dissimilarity metrics for the second vital may also be generated based on the differences of pairs between a predicted value and an actual value. Since the dissimilarity metrics may be normalized, the dissimilarity metrics can be compared with those corresponding to the first vital. Similar processes may be performed for a third vital, a fourth vital, etc. [0075] The predictive maintenance server 110 may generate 560 an anomaly score for the equipment 150 based on the dissimilarity metrics associated with one or more vitals. The anomaly score may be the overall anomaly score for the equipment 150. The score may be generated periodically (e.g., daily). The anomaly score may be an aggregated result of the dissimilarity metrics. In one embodiment, the anomaly score may be the highest value of the different dissimilarity metrics. In another embodiment, the anomaly score may be the average of the dissimilarity metrics. In yet another embodiment, the anomaly score may be correlated with the dissimilarity metrics. For example, the predictive maintenance server 110 may have a function that maps the dissimilarity metrics to an overall anomaly score. The predictive maintenance server 110 may generate 570 an alert based on the anomaly score. For example, an alert of maintenance may be triggered when the overall anomaly score is higher than a threshold value. For example, the threshold value may be a value that is multiple times (e.g., 3x) of the normal range of anomaly scores. The normal value of the anomaly score may be close to 1 because the anomaly score is often normalized. Another threshold value or a dynamic threshold value may be used.); and
generate a notification comprising an identification of the operating condition (Fig. 5-570, [0075] The predictive maintenance server 110 may generate 560 an anomaly score for the equipment 150 based on the dissimilarity metrics associated with one or more vitals. The anomaly score may be the overall anomaly score for the equipment 150. The score may be generated periodically (e.g., daily). The anomaly score may be an aggregated result of the dissimilarity metrics. In one embodiment, the anomaly score may be the highest value of the different dissimilarity metrics. In another embodiment, the anomaly score may be the average of the dissimilarity metrics. In yet another embodiment, the anomaly score may be correlated with the dissimilarity metrics. For example, the predictive maintenance server 110 may have a function that maps the dissimilarity metrics to an overall anomaly score. The predictive maintenance server 110 may generate 570 an alert based on the anomaly score. For example, an alert of maintenance may be triggered when the overall anomaly score is higher than a threshold value. For example, the threshold value may be a value that is multiple times (e.g., 3x) of the normal range of anomaly scores. The normal value of the anomaly score may be close to 1 because the anomaly score is often normalized. Another threshold value or a dynamic threshold value may be used.).
3. The system of claim 1, wherein the instructions further cause the system to: using the second signaling channel, generate an actual sensor value dataset; determine whether a similarity value between data in the actual sensor value dataset and the predicted sensor value dataset is under a predetermined threshold; based on the determination that the similarity value is under the predetermined threshold, generate a second training dataset using the actual sensor value dataset; and using the second training dataset, further train the trained neural network (Fig. 5-560, [0074] The machine learning model may be trained for a particular vital (sensor channel). According to an embodiment, the PPP model may determine the anomaly of a piece of equipment 150 using one or more vitals. For the PPP model that uses more than one vital, multiple machine learning models may be trained. Each machine learning model may be specific to a particular vital. The predictive maintenance server 110 may repeat the process described in steps 520 to 550 for different vitals, as indicated by the arrow 555. For example, for a second vital, a second machine learning model may be used. The predictive maintenance server 110 may divide the set of scoring sensor data into a different way compared to the first vital. For the second vital, the first subset of scoring sensor data may exclude the measurements of the second vital but include the measurements of the first vital. For the second vital, the second subset of scoring sensor data includes the measurements of the second vital. One or more dissimilarity metrics for the second vital may also be generated based on the differences of pairs between a predicted value and an actual value. Since the dissimilarity metrics may be normalized, the dissimilarity metrics can be compared with those corresponding to the first vital. Similar processes may be performed for a third vital, a fourth vital, etc. [0075] The predictive maintenance server 110 may generate 560 an anomaly score for the equipment 150 based on the dissimilarity metrics associated with one or more vitals. The anomaly score may be the overall anomaly score for the equipment 150. The score may be generated periodically (e.g., daily). The anomaly score may be an aggregated result of the dissimilarity metrics. In one embodiment, the anomaly score may be the highest value of the different dissimilarity metrics. In another embodiment, the anomaly score may be the average of the dissimilarity metrics. In yet another embodiment, the anomaly score may be correlated with the dissimilarity metrics. For example, the predictive maintenance server 110 may have a function that maps the dissimilarity metrics to an overall anomaly score. The predictive maintenance server 110 may generate 570 an alert based on the anomaly score. For example, an alert of maintenance may be triggered when the overall anomaly score is higher than a threshold value. For example, the threshold value may be a value that is multiple times (e.g., 3x) of the normal range of anomaly scores. The normal value of the anomaly score may be close to 1 because the anomaly score is often normalized. Another threshold value or a dynamic threshold value may be used. [0108] … The predictive maintenance server 110 performs incremental training/scoring on subset S to get an updated model and scores for the selected subset).
5. The system of claim 1, wherein the instructions further cause the system to: modify metadata associated with at least one of the first signaling channel and the second signaling channel to label a particular channel as a source of predicted signaling data for another particular channel ([0109] FIGS. 19A-19C illustrate user interfaces for displaying anomalies, according to an embodiment. FIG. 19A shows a list of anomalies in a tabular form at store/equipment level. The table shows fields including store name 1900, a channel id 1902 (identifying a sensor), a device type group 1904 (metadata describing device), duration 1904 (time interval associated with the anomaly), status 1908 indicating whether the anomaly is on-going, an average risk score 1910, the last risk score 1912 that was determined for the anomaly, and diagnosis status 1914. [0059] The trained anomaly detection models 360 may be supplemented with labeled training data 340. The anomaly detection models 360 may be adjusted based on the labeled training data 340 to improve the performance of anomaly detection. [0096] In another embodiment, the predictive maintenance server 110 determines measurements of the degree of anomaly for each sensor channel. The predictive maintenance server 110 uses the measurements of the degree of anomaly for helping an operator to diagnose an equipment problem, e.g., to identify failing components of the equipment. To determine the root cause sensors that mainly contribute to the detected anomaly, the predictive maintenance server 110 may calculate the probability density for each sensor and collect most abnormal K sensors among N sensors. The predictive maintenance server 110 may flag the most abnormal K sensors as the root cause sensors.).
6. The system of claim 1, wherein the modified metadata is utilized in training or retraining of the neural network ([0108] … The predictive maintenance server 110 performs incremental training/scoring on subset S to get an updated model and scores for the selected subset. Examiner note: incremental training/scoring implied training and retraining of the model/NN).
7. The system of claim 1, wherein the instructions further cause the system to: cause a computing device to display the notification, wherein the computing device comprises at least one of an on-board computing system, an on-board navigation system, or a mobile computing device communicatively coupled to the one or more industrial machines ([0080]… 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. [0039] The predictive maintenance system 100 may also include equipment 150 that is machinery such as automobiles, airplanes, and even satellites that may not be confined or located in a facility site 140. The equipment 150 that is not hosted within a facility site 140 may directly transmit data and communicate with the predictive maintenance server 110 through the network 130. [0053] The predictive maintenance server 110 may include a front-end interface 280 that may be a user interface for presentation of facility profiles, equipment profiles, maintenance data, sensor data, statistics and plots of various data (e.g., values over time), maintenance and inspection recommendations. The front-end interface 280 may be a graphical user interface (GUI) that displays various information and graphical elements. The front-end interface 280 may take different forms. In one case, the front-end interface 280 may be a software application that can be displayed at an electronic device such as a computer or a smartphone. The software application may be developed by the entity controlling the predictive maintenance server 110 and be downloaded and installed at the electronic device. In another case, the front-end interface 280 may take the form of a webpage interface of the predictive maintenance server 110 that allows operators to view status and manage different pieces of equipment through web browsers. In yet another case, the front-end interface 280 may provide an application program interface (API) for different devices such as servers of different facilities to retrieve data and maintenance information and recommendation. Example GUIs are shown in FIG. 19A through FIG. 19C.)
Claims 8, 10, 12-14 are drawn to claims 1, 3, 5-7 and are rejected for the same reason. Note for limitation ”One or more non-transitory, computer-readable storage media storing instructions, which, when executed by at least one processor, cause a system to perform operations comprising ([0007] In one embodiment, a non-transitory computer readable medium that is configured to store instructions is described. The instructions, when executed by one or more processors, cause the one or more processors to perform a process that includes steps described in the above computer-implemented methods or described in any embodiments of this disclosure. In one embodiment, a system may include one or more processors and a storage medium that is configured to store instructions. The instructions, when executed by one or more processors, cause the one or more processors to perform a process that includes steps described in the above computer-implemented methods or described in any embodiments of this disclosure”).
Claims 15, 17, 19-20 are drawn to claims 1, 3, 5-6 and are rejected for the same reason. Note for limitation ”A computer-implemented method for machine-learning (ML)-assisted failover of sensor signaling channels for industrial machines, the method comprising“([0007] In one embodiment, a non-transitory computer readable medium that is configured to store instructions is described. The instructions, when executed by one or more processors, cause the one or more processors to perform a process that includes steps described in the above computer-implemented methods or described in any embodiments of this disclosure. In one embodiment, a system may include one or more processors and a storage medium that is configured to store instructions. The instructions, when executed by one or more processors, cause the one or more processors to perform a process that includes steps described in the above computer-implemented methods or described in any embodiments of this disclosure”).
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 (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 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 2, 9, 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Trinh et al (US 20200379454 A1) in view of Dasgupta et al (US 2021312284 A1)
Claims 2, 9, 16: Trihn fails to disclose a determination that the second signaling channel is unavailable. However, Dasgupta disclose conditionally execute the trained neural network to generate the predicted sensor value dataset for the second signaling channel based on a determination that the second signaling channel is unavailable (Fig. 3-323, [0007] In other embodiments an error counter monitors a number of errors produced by the target sensor, the error counter increments the number of errors each time the real-time output value of the target sensor does not match the estimated output value of the target sensor. A notification generator may be included to produce a message when the number of errors reaches or exceeds a pre-determined number of errors. The notification generator is configured to identify the target sensor and suggest a course of corrective action. The course of action may include re-calibrating the target sensor or replace the target sensor. [0027] When the actual sensor value 311 is determined as a match 317 to the estimated value 313, the actual sensor value 311 is deemed reliable and is used 321 as the output of the sensor. If the actual sensor value 311 is determined not to match 319 the estimate 313, then the estimated value is used 323 and the target sensor is flagged as producing an error. For each non-matching value produced by the target sensor, an error count is incremented 325. Once the target sensor produces too many errors as determined by the system operator, a notification or warning may be generated to notify the operator that remediation is necessary for the target sensor. For example, the warning message may indicate that the target sensor requires calibration, or possible replacement.).
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the equipment monitoring of Trinh to incorporate error monitoring of Dasgupta.
Given the advantage of using estimated value when actual value is unavailable due to error/faulty sensor, one having ordinary skill in the art would have been motivated to make this obvious modification.
Claim(s) 4, 11, 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Trinh et al (US 20200379454 A1) in view of Taguchi et al (US 20230140271 A1)
Claims 4, 11, 18: Trihn disclose generate a first vectorized representation of the actual sensor value dataset; generate a second vectorized representation of the predicted sensor value dataset ([0118] Furthermore, the process of clustering sequences of data is significantly more efficient when performed based on the feature vector representation of the sequences as compared to processing of the sequence data itself. This is so because the number of elements in the sequence data can be significantly higher than the number of elements in the feature vector representation of a sequence. Accordingly, a comparison of raw data of two sequences requires significantly more computations than comparison of two feature vector representations. Furthermore, since each sequence can be of different size, comparison of data of two sequences would require additional processing to extract individual features. [0004] In one embodiment, a computer-implemented predictive maintenance process is described. The process may include 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. The process may also include selecting a subset of sensor data, the subset of sensor data comprising data generated from the sensors and excluding the measured values of the target sensor. The process may further include inputting the subset of sensor data into a machine learning model to generate predicted sensor values of the target sensor. The process may further include determining differences between the predicted sensor values of the target sensor and the measured values of the target sensor. The process may further include 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.). However, Trinh fails to disclose cosine similarity.
Taguchi disclose determine a cosine similarity score for the first vectorized representation and the second vectorized representation ([0098] The score calculation unit 1301 receives the first differential value distribution from a first distribution generation unit 107 and the second differential value distribution from a second distribution generation unit 108, and calculates a score regarding a distribution difference. As the score, for example, similarity or dissimilarity between the first differential value distribution and the second differential value distribution may be used. As the similarity, for example, a histogram intersection and a cosine similarity may be used. The dissimilarity may be, for example, Kullback-Leibler divergence or Wasserstein distance, or may be a Euclidean distance, a difference between the maximum values, a difference between the minimum values, a difference between average values, a difference between median values, or a weighted sum of these criteria.).
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the equipment monitoring of Trinh to incorporate cosine similarity of Taguchi.
Given the fact that cosine similarity is just one of the many similarity measure/score one can use, one having ordinary skill in the art would have been motivated to make this obvious modification.
Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Wu et al (US 20170364818 A1) disclose sensor monitoring and predictive maintenance using machine learning. See abstract, [0019], [0020].
Carvalho et al (“A systematic literature review of machine learning methods applied to predictive maintenance” 2019) disclose a review of machine learning methods applied to predictive maintenance. See abstract.
Hector et al (“Predictive maintenance in Industry 4.0: a survey of planning models and machine learning techniques” May 14, 2024) disclose a survey of machine learning techniques for predictive maintenance. See abstract.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LUT WONG whose telephone number is (571)270-1123. The examiner can normally be reached M-F 10am-6pm EST.
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, Abdullah Al Kawsar can be reached at 5712703169. 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.
/LUT WONG/ Primary Examiner, Art Unit 2127
Prosecution Insights