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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 12/05/25 has been entered.
Information Disclosure Statement
The IDS of 12/05/25 has been considered.
Response to Arguments
Applicant's arguments filed 11/10/25 have been fully considered but they are not persuasive.
The applicant’s key argument is:
PNG
media_image1.png
308
800
media_image1.png
Greyscale
This argument is not persuasive because Xu et al appears to teach the amended limitation, based on broadest reasonable interpretation (BRI). It should be noted that neither the claims, nor the applicant’s specification, appear to define or give details as to what constitutes a “mathematical function” or a “logical ruleset.” Under broadest reasonable interpretation, many of the data features disclosed by Xu et al can be broadly construed to be a mathematical function or a logical ruleset. For example, figure 10 of Xu discloses a “Sort by” dropdown menu, where a user can select an option. Each option can be broadly construed to be a logical ruleset, such as sorting outputs from first to last sheet, as displayed. Xu column 10, lines 50-67 disclose many data elements that can be construed to constitute mathematical functions, such as graphs 485, 490, 495, 500, 505, 510, 515, 520, 525, 530, 535, 540, 545, 550, and 555 in figure 10.
The rejection is maintained.
Drawings
As stated in a previous action, the drawings of 07/17/21 are accepted.
Examiner’s Note - 35 USC § 101
For the reasons discussed previously, claims 1-20 qualify as eligible subject matter under 35 U.S.C. 101.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-11 and 14-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Trinh et al (US PgPub 20200379454) in view of Xu et al (US Pat 11410056) and Lei et al (US PgPub 20210224833)
With respect to claim 1, Trinh et al discloses:
A data processing system for generating predictive maintenance models (abstract; paragraph 0044 states, “The predictive maintenance server 110 may train one or more machine learning models … The predictive maintenance server 110 may also train additional models …” The training of models by the predictive maintenance server suggests the generation of predictive maintenance models …”)
one or more processors (figure 20, reference 2002)
a memory including one or more digital storage devices (figure 20, reference 2006)
a plurality of instructions stored in the memory and executable by the one or more processors (paragraph 0035 states, “Parts of the predictive maintenance server 110 may also include a memory that stores computer code including instructions that may cause the processors to perform certain actions when the instructions are executed, directly or indirectly by the processors.”) to:
receive a historical dataset relating to each system of a plurality of systems of a fleet of aircraft (suggested by paragraphs 0039-0040, which state, “The predictive maintenance system 100 may also include equipment 150 that is machinery such as automobiles, airplanes, and even satellites … 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 … A collection of various sensors 154 may sometimes be referred to as an IoT fleet. A collection of equipment 150 may sometimes be referred to as an equipment fleet. The predictive maintenance may be referred to as fleet management.”), the historical dataset including time-labeled maintenance data and operational data (paragraph 0071 states, “The machine learning model may be trained by a training data set that includes historical measurements of sensor data …”; figure 3, reference 324 discloses “Repair Date and Time” data; paragraph 0006 states, “The reference histogram may represent a first distribution of sensor data during the first time interval.”; paragraph 0041 states, “The data generated by a sensor 154 may be in any suitable format such as a time-series format. For example, a sensor 154 may monitor the temperature of a particular component of a piece of equipment.”)
generate a predictive maintenance model according to a machine learning method (paragraph 0044 states, “The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152. The predictive maintenance server 110 may train one or more machine learning models …”)
With respect to claim 1, Trinh et al differs from the claimed invention in that it does not explicitly disclose:
wherein the maintenance data includes maintenance events of the plurality of systems
calculate values of a plurality of operational data features from series of sensor readings of the operational data
receive, from a user, a first selection comprising:
a first operational data feature of the plurality of operational data features
a first system
a seasonal bias removal selection
modify, based on the seasonal bias removal selection, operational data associated with the first operational data feature and the first system to compensate for a seasonal variation of the first operational data feature
wherein, to modify the operational data, the plurality of instructions are executable by the one or more processors
perform a sinusoidal regression for the first operational data feature based on the seasonal variation of the first operational data feature
subtract sinusoidal regression curve values from data feature values of the first operational data feature
display the operational data on a timeline in a graphical user interface
display maintenance events of the first system overlaid onto the operational data on the timeline
receive, from the user and after displaying the operational data and the maintenance events, a second selection of a second operational data feature of the plurality of operational data features, wherein the second selection comprises a mathematical function or a logical ruleset defining the second operational data feature
subsequently, generate a predictive maintenance model, using the second operational data feature according to a machine learning method
With respect to claim 1, Xu et al discloses:
wherein the maintenance data includes maintenance events of the plurality of systems (column 5, lines 56-63 state, “the data set 355 includes engine performance data for each engine in a training grouping of engines; operational data regarding routine maintenance for each engine … and a listing of removed engines within the training grouping of engines that were removed from service due to an unplanned maintenance event.”)
calculate values of a plurality of operational data features from series of sensor readings of the operational data (column 2, lines 43-46 state, “The network of physical sensors is placed in specific locations to detect an exhaust gas temperature, vibration, speed, oil pressure, and fuel flow for each aircraft engine.”; column 9, lines 13-16 state, “the following ratios are calculated and defined as accuracy measures …”; see various parameters of figures 11-12)
receive, from a user, a first selection (figures 11-12; column 9, lines 35-40 state, “When the system 10 receives a selection command associated with the link 420, via either voice, touch, eye tracking, etc., from a user of the system 10 …”) comprising:
a first operational data feature of the plurality of operational data features (figures 11-12 show operational data; and the selectable elements (column 9, lines 27-34) represent operational data features)
a first system (column 9, line 35 states, “When the system 10 …”)
a seasonal bias removal selection (obvious in view of combination; column 17, lines 30-45 of Xu state, “identifying the listing of engines comprises comparing the engine-specific information for each engine to one or more of: a predetermined threshold … seasonal trends; and cyclical non-seasonal trends … operational data regarding routine maintenance for each engine … and a listing of removed engines …”; Primary reference Trinh et al also teaches selection – paragraph 0062 states, “The predictive maintenance server 110 may also select some of the data fields (e.g., certain sensor channels) but not all of the data fields for training.” Allowing a user to select (and thereby choose or remove selection) among different data points, including seasonal data considerations, would be obvious to one of ordinary skill in the art.)
modify, based on the seasonal bias removal selection, operational data associated with the first operational data feature and the first system to compensate for a seasonal variation of the first operational data feature (column 6, lines 48-67 state, “Cyclical patterns, seasonality, and other factors are considered when analyzing the data … the exhaust gas temperature can be a cyclical pattern based on engine washes. One example of seasonal patterns is oil consumption. Using these adjustments or considerations in data, the prediction module 225 can detect when an engine parameter is outside of an acceptable range or even detect a sensor error …”; see also column 17, lines 30-45; column 19, lines 4-17 state, “the predetermined threshold varies based on seasonality and ambient operating temperature of each engine in the first plurality of engines … the method also includes moving, within the predetermined period of time, an aircraft that includes an engine within the listing of engines to a location that includes a maintenance facility. In one embodiment, the method also includes identifying, using the module, a maintenance action needed to prevent the unplanned maintenance event …”)
display the operational data on a timeline in a graphical user interface (figures 11-2; column 9, lines 35-40 state, “When the system 10 receives a selection command associated with the link 420, via either voice, touch, eye tracking, etc., from a user of the system 10, the GUI formatting module 230 formats and displays …”; Xu figure 5; column 2, lines 10-11 state, “FIG. 5 is a graphical illustration of a timeline …”; column 4, lines 36-37 state, “Generally, a timeline depicting the life of the engine …”)
display maintenance events of the first system overlaid onto the operational data on the timeline (figures 11-12; column 9, line 65 – column 10, line 14 state, “As can be seen in column 450a … the engine has a 72% likelihood that the associated engine will be removed from service due to an unpalnned maintenance event within the period of time … In some embodiments, a date is also listed in the prediction data, and the date is associated with the next scheduled maintenance visit for the engine 470.”)
receive, from the user and after displaying the operational data and the maintenance events, a second selection of a second operational data feature of the plurality of operational data features (figures 10-13; column 10, line 50 – column 11, line 15 discloses additional operational data features), wherein the second selection comprises a mathematical function or a logical ruleset defining the second operational data feature (Neither the claims, nor the applicant’s specification, appear to define or give details as to what constitutes a “mathematical function” or a “logical ruleset.” Under broadest reasonable interpretation, many of the data features disclosed by Xu et al can be broadly construed to be a mathematical function or a logical ruleset. For example, figure 10 of Xu discloses a “Sort by” dropdown menu, where a user can select an option. Each option can be broadly construed to be a logical ruleset, such as sorting outputs from first to last sheet, as displayed. Xu column 10, lines 50-67 disclose many data elements that can be construed to constitute mathematical functions, such as graphs 485, 490, 495, 500, 505, 510, 515, 520, 525, 530, 535, 540, 545, 550, and 555 in figures 13A-13B.)
subsequently, generate a predictive maintenance model, using the second operational data feature according to a machine learning method (obvious in view of combination; Trinh teaches generating a predictive maintenance model using sensed abnormality data, and Xu et al provides more specific operational data features to help determine the abnormalies. Column 11, lines 8-15 state, “a portion of the graphs … compare data received by the sensors associated with the engine 470 and its paired engine 481. As such, the deterioration of performance of the engine 470 is comparable (visually) with an engine that is performing within normal ranges …” It would be obvious to generate a predictive maintenance model off more specific data.)
With respect to claim 1, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Xu et al into the invention of Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of detecting and tracking performance and potential risks using a network of sensor in a field of use, like with aircraft.
With respect to claim 1, Lei et al discloses:
wherein, to modify the operational data, the plurality of instructions are executable by the one or more processors to:
perform a sinusoidal regression for the first operational data feature based on the seasonal variation of the first operational data feature (paragraph 0058 states, “The possible methods can be any known methods of estimating seasonality … (3) sinusoidal model; and (4) Regression with Fourier terms.”; paragraph 0073 states, “The regression with Fourier terms method, similar to using the sinusoidal model, adds Fourier terms into regression models to utilize sine and cosine terms in order to simulate seasonality. However, the seasonality of such a regression would be represented as the sum of sine or cosine terms …” The examiner broadly interprets the disclosure of regression using sine terms to represent the claimed “sinusoidal regression. Based on the teachings of Lei et al, as well as in additional art, such as that cited below as pertinent prior art, it would appear that using sinusoidal regression for processing seasonal data is well-known and well-established in the art.)
subtract sinusoidal regression curve values from data feature values of the first operational data feature (This limitation would be mathematically obvious to one of ordinary skill in the art. If sinusoidal regression is to be used to represent seasonality, subtracting such values would be obvious to one of ordinary skill in the art that wishes to remove or exclude consideration of that data.)
With respect to claim 1, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Lei et al into the invention of Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of mathematically considering seasonality, either from a standpoint of including or excluding the data.
With respect to claim 2, Trinh et al, as modified, discloses:
wherein the timeline is a time axis of a scatterplot, and the operational data associated with the first operational data feature and the first system are displayed as points of the scatterplot (Trinh et al paragraph 0107 states, “The predictive maintenance server 110 retrieves normal sensor data (X_normal) from a second time interval occurring before the first time interval (e.g., the previous 7 days before X_abnormal as X_normal). For each key dimension, predictive maintenance server 110 generates a visual representation of the anomalous sensor data compared with normal sensor data, for example, histogram difference of each key dimension or a pairwise scatterplot of the difference of values from the two histograms.”)
With respect to claim 3, Trinh et al, as modified, discloses:
wherein each point of the scatterplot indicates a respective value calculated from a series of sensor readings recorded on a flight (suggested by teachings of Trinh et al paragraph 0039 and 0107; paragraph 0039 states, “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.”; paragraph 0107 discloses scatterplot. Furthermore, Xu et al discloses fleet of aircraft (figure 1, reference 15; column 1, lines 19-22))
With respect to claim 4, Trinh et al, as modified, discloses:
wherein the plurality of systems have a shared plurality of operational phases, and the plurality of instructions are further executable by the one or more processors (Xu et al column 6, lines 63-67 state, “In some embodiments, the prediction module 225 also accounts for the performance of the engines during different phases (e.g., take off and cruise) when evaluating engine performance.”) to:
divide the operational data into a plurality of phase-dependent data subsets corresponding to phases of the shared plurality of operational phases (obvious in view of Xu et al column 6, lines 63-67; dividing is obvious when Xu et al is already “accounting” for different phases in the context of considering many different types of data (figures 10-12))
With respect to claim 5, Trinh et al, as modified, discloses:
wherein the first selection includes an operational phase, and the displayed operational data is in a phase-dependent data subset corresponding to the operational phase (obvious in view of Xu et al column 6, lines 63-67 and figures 10-12)
With respect to claim 6, Trinh et al, as modified, discloses:
wherein the shared plurality of operational phases are phases of flight (obvious in view of combination; see Xu et al figure 1, reference 15; column 1, lines 19-22; column 6, lines 64-67)
With respect to claim 7, Trinh et al, as modified, discloses:
wherein the shared plurality of operational phases include one or more of taxi-out, takeoff, climb, cruise, descent, landing, and taxi-in (obvious in view of combination; see Xu et al column 6, lines 63-67)
With respect to claim 8, Trinh et al, as modified, discloses:
wherein the operational data of the historical dataset includes a plurality of series of sensor readings, each series of sensor readings having been recorded during a flight of an aircraft of the fleet of aircraft (see figures 13A-13B of Xu et al – “Flight Date”) and the plurality of instructions are further executable by the one or more processors to (obvious in view of the processor teachings of Trinh et al, as discussed in claim 1 above))
receive a selection of a flight (obvious in view of selection teachings of Xu et al – column 9, lines 35-40)
display the series of sensor readings recorded during the selected flight, on a separate timeline (figures 13A-13B)
With respect to claim 9, Trinh et al, as modified, discloses:
wherein the first operational data feature is an aggregate data feature, calculated according to an aggregating statistical function (Trinh et al paragraph 0046 states, “the predictive maintenance server 110 may use raw data that has not been aggregated. In other cases, the data processing engine may aggregate some of the raw data.”; paragraph 0099 states, “The predictive maintenance server 110 determines 1520 a combined anomaly score for the time interval T as the aggregate value (e.g., sum) of the score values for all sensors for the time interval T …”; paragraph 0108 states, “The predictive maintenance server 110 globally aggregates the models and scores …”)
With respect to claim 10, Trinh et al, as modified, discloses:
wherein displaying the operational data associated with the first operational data feature includes displaying a calculated value of the aggregating statistical function for each of a plurality of series of sensor readings (obvious in view of combination; Trinh et al discloses aggregating statistical function, as discussed in claim 9 above. Xu et al discloses displaying operational data, as shown in figures 10-13)
With respect to claim 11, Trinh et al, as modified, discloses:
wherein receiving the first selection of the first operational data feature includes receiving a selection of a sensor data parameter, a position, and the aggregating statistical function, and the aggregate data feature is calculated from operational data associated with the sensor data parameter and the position according to the selected aggregating statistical function (obvious in view of combination; Trinh et al discloses aggregating statistical function; Xu et al discloses selection of sensor data parameter with respect to a GUI that displays a wide variety of data about a fleet of aircraft)
With respect to claim 14, Trinh et al, as modified, discloses:
wherein the first selection includes a plurality of systems, and the plurality of instructions are further executable by the one or more processors to: display operational data associated with each system of the plurality of systems on the timeline, the operational data associated with each system of the plurality of systems being visually distinct from operational data associated with the other systems of the plurality of systems (Display of operational data is discussed above. Both Trinh et al and Xu et al disclose display of data. See, in particular figures 10-13 of Xu et al.)
With respect to claim 15, Trinh et al, as modified, discloses:
wherein the plurality of instructions are further executable by the one or more processors to: visually indicate a division of the displayed operational data into multiple time periods according to a selected time relationship to the displayed maintenance events (see figures 10-13 of Xu et al)
With respect to claim 16, Trinh et al, as modified, discloses:
wherein the multiple time periods include a first set of periods of time more than a first threshold time prior to each of the displayed maintenance events, and a second set of periods of time less than a second threshold time prior to each of the displayed maintenance events, and the plurality of instructions are further executable by the one or more processors to: receive a selection of the first and second threshold times (obvious in view of combination; Xu et al column 4, line 54 – column 5, line 4 states, “Moreover, the module 55 predicts the remaining life of an engine 35 based on: the trend of the sensor data associated with that engine relative to sensor data associated with all sensors in the fleet 15, the trend of the sensor data associated with that engine relative to predetermined thresholds … the threshold must be exceeded a number of times within a predetermined period …” Column 4, lines 26-31 of Xu et al state, “the engine sensor module 55 can rearrange the order of engines planned for heavy or light maintenance based on the at-risk engine listing and/or the predicted time range at which the at-risk engine is expected to have a repair issue that warrants an unexpected removal.” The claimed limitation is obvious in view of this time range, especially when taking into account the various thresholds considered by Xu et al.)
With respect to claim 17, Trinh et al, as modified, discloses:
receive a selection of two groups of series of sensor readings (Trinh et al paragraphs 0042-0043 and Xu et al column 2, lines 43-46)
display values of the first operational data feature in one or more of a density plot, a box plot, and a scatter plot including a linear regression (obvious in view of combination; Trinh et al discloses both scatter plot (paragraph 0107) and linear regression (paragraph 0048). Trinh et al does not explicitly disclose including the linear regression with the scatter plot, but it would be obvious to one of ordinary skill in the art to display the data that is been processed through the various machine learning techniques, which may include linear regression techniques. Furthermore, Xu et al discloses a wide variety of different plots, including displaying data in an array of boxes, such as in figures 10-12)
With respect to claim 18, Trinh et al, as modified, discloses:
wherein the first selection includes a plurality of operational data features, and the plurality of instructions are further executable by the one or more processors (as discussed above with respect to claim 1)
visually indicate a division of the displayed operational data into multiple time periods according to a selected time relationship to the displayed maintenance events, the multiple time periods including a first set of periods of time more than a first threshold time prior to each of the displayed maintenance events, and a second set of periods of time less than a second threshold time prior to each of the displayed maintenance events (see rejection of claims 15-16 above)
wherein a first group of the two groups of series of sensor readings is selected from the first set of time periods, and a second group of the two groups of series of sensor readings is selected from the second set of time periods (obvious for reasons discussed in the rejection of claims 15-16 above)
With respect to claim 19, Trinh et al discloses:
A computer implemented method of generating a predictive maintenance model (abstract; paragraph 0004 states, “a computer-implemented predictive maintenance method is described.”; paragraph 0044 states, “The predictive maintenance server 110 may train one or more machine learning models … The predictive maintenance server 110 may also train additional models …” The training of models by the predictive maintenance server suggests the generation of predictive maintenance models …”)
receiving a historical dataset relating to each system of a plurality of systems of a fleet of aircraft (suggested by paragraphs 0039-0040, which state, “The predictive maintenance system 100 may also include equipment 150 that is machinery such as automobiles, airplanes, and even satellites … 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 … A collection of various sensors 154 may sometimes be referred to as an IoT fleet. A collection of equipment 150 may sometimes be referred to as an equipment fleet. The predictive maintenance may be referred to as fleet management.”), the historical dataset including time-labeled maintenance data and operational data (paragraph 0071 states, “The machine learning model may be trained by a training data set that includes historical measurements of sensor data …”; figure 3, reference 324 discloses “Repair Date and Time” data; paragraph 0006 states, “The reference histogram may represent a first distribution of sensor data during the first time interval.”; paragraph 0041 states, “The data generated by a sensor 154 may be in any suitable format such as a time-series format. For example, a sensor 154 may monitor the temperature of a particular component of a piece of equipment.”), and the plurality of systems have a shared plurality of operational phases (Xu et al column 6, lines 63-67 state, “In some embodiments, the prediction module 225 also accounts for the performance of the engines during different phases (e.g., take off and cruise) when evaluating engine performance.”)
generating a predictive maintenance model according to a machine learning method (paragraph 0044 states, “The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152. The predictive maintenance server 110 may train one or more machine learning models …”)
With respect to claim 19, Trinh et al differs from the claimed invention in that it does not explicitly disclose:
wherein the maintenance data includes maintenance events of the plurality of systems and the plurality of systems have a shared plurality of operational phases
calculating values of a plurality of operational data features from series of sensor readings of the operational data
dividing the operational data into a plurality of phase-dependent data subsets corresponding to the phases of the shared plurality of operational phases
receiving, from a user, a first selection comprising:
a first operational data feature of the plurality of operational data features
a first system
a seasonal bias removal selection
modifying, based on the seasonal bias removal selection, operational data associated with the first operational data feature and the first system to compensate for a seasonal variation of the first operational data feature
wherein modifying the operational data comprises:
performing a sinusoidal regression for the first operational data feature based on the seasonal variation of the first operational data feature
subtracting sinusoidal regression curve values from data feature values of the first operational data feature
displaying the operational data in a phase-dependent data subset corresponding to the operational phase of the first selection, on a timeline in a graphical user interface
displaying maintenance events of the first system overlaid onto the operational data on the timeline
receiving, from the user and after displaying the operational data and the maintenance events, a second selection of a second operational data feature of the plurality of operational data features, wherein the second selection comprises a mathematical function or a logical ruleset defining the second operational data feature
subsequently, generating a predictive maintenance model, using the second operational data feature according to a machine learning method
With respect to claim 19, Xu et al discloses:
wherein the maintenance data includes maintenance events of the plurality of systems and the plurality of systems have a shared plurality of operational phases (column 5, lines 56-63 state, “the data set 355 includes engine performance data for each engine in a training grouping of engines; operational data regarding routine maintenance for each engine … and a listing of removed engines within the training grouping of engines that were removed from service due to an unplanned maintenance event.”)
calculating values of a plurality of operational data features from series of sensor readings of the operational data (column 2, lines 43-46 state, “The network of physical sensors is placed in specific locations to detect an exhaust gas temperature, vibration, speed, oil pressure, and fuel flow for each aircraft engine.”; column 9, lines 13-16 state, “the following ratios are calculated and defined as accuracy measures …”; see various parameters of figures 11-12)
dividing the operational data into a plurality of phase-dependent data subsets corresponding to the phases of the shared plurality of operational phases (obvious in view of Xu et al column 6, lines 63-67; dividing is obvious when Xu et al is already “accounting” for different phases in the context of considering many different types of data (figures 10-12))
receiving, from a user, a first selection (figures 11-12; column 9, lines 35-40) comprising:
a first operational data feature of the plurality of operational data features (figures 11-12 show operational data; and the selectable elements (column 9, lines 27-34) represent operational data features)
a first system (column 9, line 35 states, “When the system 10 …”)
a seasonal bias removal selection (obvious for reasons discussed in claim 1 above)
modifying, based on the seasonal bias removal selection, operational data associated with the first operational data feature and the first system to compensate for a seasonal variation of the first operational data feature (column 6, lines 48-67 state, “Cyclical patterns, seasonality, and other factors are considered when analyzing the data … the exhaust gas temperature can be a cyclical pattern based on engine washes. One example of seasonal patterns is oil consumption. Using these adjustments or considerations in data, the prediction module 225 can detect when an engine parameter is outside of an acceptable range or even detect a sensor error …”; see also column 17, lines 30-45; column 19, lines 4-17 state, “the predetermined threshold varies based on seasonality and ambient operating temperature of each engine in the first plurality of engines … the method also includes moving, within the predetermined period of time, an aircraft that includes an engine within the listing of engines to a location that includes a maintenance facility. In one embodiment, the method also includes identifying, using the module, a maintenance action needed to prevent the unplanned maintenance event …”)
displaying the operational data that is in a phase-dependent data subset corresponding to the operational phase of the first selection, on a timeline in a graphical user interface (figures 10-12; column 9, lines 35-40 state, “When the system 10 receives a selection command associated with the link 420, via either voice, touch, eye tracking, etc., from a user of the system 10, the GUI formatting module 230 formats and displays …”; see also column 6, lines 63-67, which states, “In some embodiments, the prediction module 225 also accounts for the performance of the engines during different phases (e.g., take off and cruise) when evaluating engine performance. The claimed limitation is obvious in view of the disclosure of Xu, as applied to Trinh et al.)
displaying maintenance events of the first system overlaid onto the operational data on the timeline (figures 11-12; column 9, line 65 – column 10, line 14 state, “As can be seen in column 450a … the engine has a 72% likelihood that the associated engine will be removed from service due to an unpalnned maintenance event within the period of time … In some embodiments, a date is also listed in the prediction data, and the date is associated with the next scheduled maintenance visit for the engine 470.”)
receiving, from the user and after displaying the operational data and the maintenance events, a second selection of a second operational data feature of the plurality of operational data features (figures 13A-13B; column 10, line 50 – column 11, line 15 discloses additional operational data features), wherein the second selection comprises a mathematical function or a logical ruleset defining the second operational data feature (Neither the claims, nor the applicant’s specification, appear to define or give details as to what constitutes a “mathematical function” or a “logical ruleset.” Under broadest reasonable interpretation, many of the data features disclosed by Xu et al can be broadly construed to be a mathematical function or a logical ruleset. For example, figure 10 of Xu discloses a “Sort by” dropdown menu, where a user can select an option. Each option can be broadly construed to be a logical ruleset, such as sorting outputs from first to last sheet, as displayed. Xu column 10, lines 50-67 disclose many data elements that can be construed to constitute mathematical functions, such as graphs 485, 490, 495, 500, 505, 510, 515, 520, 525, 530, 535, 540, 545, 550, and 555 in figures 13A-13B.)
subsequently, generating a predictive maintenance model, using the second operational data feature according to a machine learning method (obvious in view of combination; Trinh teaches generating a predictive maintenance model using sensed abnormality data, and Xu et al provides more specific operational data features to help determine the abnormalies. Column 11, lines 8-15 state, “a portion of the graphs … compare data received by the sensors associated with the engine 470 and its paired engine 481. As such, the deterioration of performance of the engine 470 is comparable (visually) with an engine that is performing within normal ranges …” It would be obvious to generate a predictive maintenance model off more specific data.)
With respect to claim 19, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Xu et al into the invention of Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of detecting and tracking performance and potential risks using a network of sensor in a field of use, like with aircraft.
With respect to claim 19, Lei et al discloses:
wherein modifying the operational data comprises:
perform a sinusoidal regression for the first operational data feature based on the seasonal variation of the first operational data feature (paragraph 0058 states, “The possible methods can be any known methods of estimating seasonality … (3) sinusoidal model; and (4) Regression with Fourier terms.”; paragraph 0073 states, “The regression with Fourier terms method, similar to using the sinusoidal model, adds Fourier terms into regression models to utilize sine and cosine terms in order to simulate seasonality. However, the seasonality of such a regression would be represented as the sum of sine or cosine terms …” The examiner broadly interprets the disclosure of regression using sine terms to represent the claimed “sinusoidal regression. Based on the teachings of Lei et al, as well as in additional art, such as that cited below as pertinent prior art, it would appear that using sinusoidal regression for processing seasonal data is well-known and well-established in the art.)
subtract sinusoidal regression curve values from data feature values of the first operational data feature (This limitation would be mathematically obvious to one of ordinary skill in the art. If sinusoidal regression is to be used to represent seasonality, subtracting such values would be obvious to one of ordinary skill in the art that wishes to remove or exclude consideration of that data.)
With respect to claim 19, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Lei et al into the invention of Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of mathematically considering seasonality, either from a standpoint of including or excluding the data.
With respect to claim 20, Trinh et al discloses:
A computer program product for generating predictive maintenance models (abstract; paragraph 0044 states, “The predictive maintenance server 110 may train one or more machine learning models … The predictive maintenance server 110 may also train additional models …” The training of models by the predictive maintenance server suggests the generation of predictive maintenance models …”; paragraphs 0004-0007 disclose computer-implemented methods.)
a non-transitory computer-readable storage medium having computer-readable program code embodied in the storage medium, the computer-readable program code configured to cause a data processing system to generate a predictive maintenance model, the computer-readable program code (paragraph 0007 states, “In one embodiment, a non-transitory computer readable medium that is configured to store instructions is described.”)
at least one instruction to receive a historical dataset relating to each system of a plurality of systems, the historical dataset including time-labeled maintenance data and operational data (paragraph 0071 states, “The machine learning model may be trained by a training data set that includes historical measurements of sensor data …”; figure 3, reference 324 discloses “Repair Date and Time” data; paragraph 0006 states, “The reference histogram may represent a first distribution of sensor data during the first time interval.”; paragraph 0041 states, “The data generated by a sensor 154 may be in any suitable format such as a time-series format. For example, a sensor 154 may monitor the temperature of a particular component of a piece of equipment.”)
at least one instruction to generate a predictive maintenance model according to a machine learning method (paragraph 0044 states, “The predictive maintenance server 110 receives and analyzes the data transmitted from various sensors 154 and settings 152. The predictive maintenance server 110 may train one or more machine learning models …”)
With respect to claim 20, Trinh et al differs from the claimed invention in that it does not explicitly disclose:
each aircraft of a fleet of aircraft
wherein the maintenance data includes maintenance events of the fleet of aircraft
at least one instruction to calculate values of a plurality of operational data features from a plurality of series of sensor readings of the operational data, wherein each series of sensor readings has been recorded during one flight of an aircraft of the fleet of aircraft
at least one instruction to receive, from a user, a first selection comprising:
a first operational data feature of the plurality of operational data features
a first aircraft
a seasonal bias removal selection
at least one instruction to modify, based on the seasonal bias removal selection, operational data associated with the first operational data feature and the first aircraft to compensate for a seasonal variation of the first operational data feature
wherein, to modify the operational data, the computer-readable program code comprises:
at least one instruction to perform a sinusoidal regression for the first operational data feature based on the seasonal variation of the first operational data feature
at least one instruction to subtract sinusoidal regression curve values from data feature values of the first operational data feature
at least one instruction to display the operational data on a timeline in a graphical user interface, wherein the timeline is a time axis of a scatterplot, and each point of the scatterplot indicates a value calculated from a series of sensor readings of the plurality of series of sensor readings
at least one instruction to display maintenance events of the first aircraft overlaid onto the operational data on the timeline
at least one instruction to receive a selection of one of the points of the scatterplot and display the series of sensor readings on a separate timeline
at least one instruction to receive, from the user and after displaying the operational data and the maintenance events, a second selection of a second operational data feature of the plurality of operational data features, wherein the second selection comprises a mathematical function or a logical ruleset defining the second operational data feature
at least one instruction to subsequently generate a predictive maintenance model, using the second operational data feature according to a machine learning method
With respect to claim 20, Xu et al discloses:
each aircraft of a fleet of aircraft (figure 1, reference 15 shows fleet of aircraft; figure 6 states, “Engine Data For Each Engine In The Fleet”; see also figure 7, references 285, 290, 295)
wherein the maintenance data includes maintenance events of the fleet of aircraft (column 5, lines 56-63 state, “the data set 355 includes engine performance data for each engine in a training grouping of engines; operational data regarding routine maintenance for each engine … and a listing of removed engines within the training grouping of engines that were removed from service due to an unplanned maintenance event.”; “fleet of aircraft” disclosed throughout disclosure of Xu et al)
at least one instruction to calculate values of a plurality of operational data features from series of sensor readings of the operational data, wherein each series has been recorded during one flight of an aircraft of the fleet of aircraft (column 2, lines 43-46 state, “The network of physical sensors is placed in specific locations to detect an exhaust gas temperature, vibration, speed, oil pressure, and fuel flow for each aircraft engine.”; column 9, lines 13-16 state, “the following ratios are calculated and defined as accuracy measures …”; see various parameters of figures 11-12)
at least one instruction to receive, from a user, a first selection (see rejection of claim 1 above) comprising:
a first operational data feature of the plurality of operational data features (figures 11-2; column 9, lines 27-34))
a first aircraft (abstract; aircraft further disclosed throughout the disclosure of Xu et al)
a seasonal bias removal selection (obvious for reasons discussed in claim 1 above)
at least one instruction to modify, based on the seasonal bias removal selection, operational data associated with the first operational data feature and the first aircraft to compensate for a seasonal variation of the first operational data feature (column 6, lines 48-67 state, “Cyclical patterns, seasonality, and other factors are considered when analyzing the data … the exhaust gas temperature can be a cyclical pattern based on engine washes. One example of seasonal patterns is oil consumption. Using these adjustments or considerations in data, the prediction module 225 can detect when an engine parameter is outside of an acceptable range or even detect a sensor error …”; see also column 17, lines 30-45; column 19, lines 4-17 state, “the predetermined threshold varies based on seasonality and ambient operating temperature of each engine in the first plurality of engines … the method also includes moving, within the predetermined period of time, an aircraft that includes an engine within the listing of engines to a location that includes a maintenance facility. In one embodiment, the method also includes identifying, using the module, a maintenance action needed to prevent the unplanned maintenance event …”)
at least one instruction to display the operational data on a timeline in a graphical user interface (figures 11-2; column 9, lines 35-40 state, “When the system 10 receives a selection command associated with the link 420, via either voice, touch, eye tracking, etc., from a user of the system 10, the GUI formatting module 230 formats and displays …”), wherein the timeline is a time axis of a scatterplot, and each point of the scatterplot indicates a value calculated from a series of sensor readings of the plurality of series of sensor readings (obvious in view of combination; Trinh et al paragraph 0107 states, “The predictive maintenance server 110 retrieves normal sensor data (X_normal) from a second time interval occurring before the first time interval (e.g., the previous 7 days before X_abnormal as X_normal). For each key dimension, predictive maintenance server 110 generates a visual representation of the anomalous sensor data compared with normal sensor data, for example, histogram difference of each key dimension or a pairwise scatterplot of the difference of values from the two histograms.”)
at least one instruction to display maintenance events of the first aircraft overlaid onto the operational data on the timeline (figures 11-12; column 9, line 65 – column 10, line 14 state, “As can be seen in column 450a … the engine has a 72% likelihood that the associated engine will be removed from service due to an unpalnned maintenance event within the period of time … In some embodiments, a date is also listed in the prediction data, and the date is associated with the next scheduled maintenance visit for the engine 470.”)
at least one instruction to receive a selection of one of the points of the scatterplot and display the series of sensor readings on a separate timeline (obvious in view of combination; see scatterplot teachings of Trinh et al)
at least one instruction to receive, from the user and after displaying the operational data and the maintenance events, a second selection of a second operational data feature of the plurality of operational data features (figures 13A-13B; column 10, line 50 – column 11, line 15 discloses additional operational data features), wherein the second selection comprises a mathematical function or a logical ruleset defining the second operational data feature (Neither the claims, nor the applicant’s specification, appear to define or give details as to what constitutes a “mathematical function” or a “logical ruleset.” Under broadest reasonable interpretation, many of the data features disclosed by Xu et al can be broadly construed to be a mathematical function or a logical ruleset. For example, figure 10 of Xu discloses a “Sort by” dropdown menu, where a user can select an option. Each option can be broadly construed to be a logical ruleset, such as sorting outputs from first to last sheet, as displayed. Xu column 10, lines 50-67 disclose many data elements that can be construed to constitute mathematical functions, such as graphs 485, 490, 495, 500, 505, 510, 515, 520, 525, 530, 535, 540, 545, 550, and 555 in figures 13A-13B.)
at least one instruction to subsequently generate a predictive maintenance model, using the second operational data feature according to a machine learning method (obvious in view of combination; Trinh teaches generating a predictive maintenance model using sensed abnormality data, and Xu et al provides more specific operational data features to help determine the abnormalies. Column 11, lines 8-15 state, “a portion of the graphs … compare data received by the sensors associated with the engine 470 and its paired engine 481. As such, the deterioration of performance of the engine 470 is comparable (visually) with an engine that is performing within normal ranges …” It would be obvious to generate a predictive maintenance model off more specific data.)
With respect to claim 20, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Xu et al into the invention of Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of detecting and tracking performance and potential risks using a network of sensor in a field of use, like with aircraft.
With respect to claim 20, Lei et al discloses:
wherein, to modify the operational data, the computer-readable program code comprises:
at least one instruction to perform a sinusoidal regression for the first operational data feature based on the seasonal variation of the first operational data feature (paragraph 0058 states, “The possible methods can be any known methods of estimating seasonality … (3) sinusoidal model; and (4) Regression with Fourier terms.”; paragraph 0073 states, “The regression with Fourier terms method, similar to using the sinusoidal model, adds Fourier terms into regression models to utilize sine and cosine terms in order to simulate seasonality. However, the seasonality of such a regression would be represented as the sum of sine or cosine terms …” The examiner broadly interprets the disclosure of regression using sine terms to represent the claimed “sinusoidal regression. Based on the teachings of Lei et al, as well as in additional art, such as that cited below as pertinent prior art, it would appear that using sinusoidal regression for processing seasonal data is well-known and well-established in the art.)
at least one instruction to subtract sinusoidal regression curve values from data feature values of the first operational data feature (This limitation would be mathematically obvious to one of ordinary skill in the art. If sinusoidal regression is to be used to represent seasonality, subtracting such values would be obvious to one of ordinary skill in the art that wishes to remove or exclude consideration of that data.)
With respect to claim 20, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of Lei et al into the invention of Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of mathematically considering seasonality, either from a standpoint of including or excluding the data.
Claim(s) 12-13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Trinh et al (US PgPub 20200379454) in view of Xu et al (US Pat 11410056) and Lei et al (US PgPub 20210224833), as applied to claims 1-11 and 14-20 above, and further in view of DiSalvo (US PgPub 20100262901).
With respect to claim 12, Trinh et al, as modified, discloses:
the data processing system of claim 1 (as applied to claim 1 above)
wherein the plurality of instructions are further executable by the one or more processors (inherent to nature of processors; processors execute instructions)
With respect to claim 12, Trinh et al, as modified, differs from the claimed invention in that it does not explicitly disclose:
calculate a best-fit periodic curve for the operational data associated with the first operational data feature and the first system
compare a period of the calculated curve to a year
when the period is within a pre-determined threshold from a year, display the calculated curve on the timeline of the graphical user interface
With respect to claim 12, DiSalvo discloses:
calculate a best-fit periodic curve for the operational data associated with the first operational data feature and the first system (paragraph 0328 discloses charting software that can execute a wide variety of plot styles and methods, including best fit curves; Please note that the abstract of DiSalvo explicitly discloses aircraft routing as one example industry that the invention may apply to.)
compare a period of the calculated curve to a year (obvious in view of combination; Trinh et al anticipates various periods of time. For example, paragraph 0041 states, “For every predetermined period of time (e.g., a second, a few second, a minute, an hour, etc.) …” Trinh et al also discloses historical data periods (paragraph 0071 states, “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). Xu et al also discloses consideration of a period of time, as discussed above. It would be obvious to one of ordinary skill in the art to consider other time periods, such as a year.)
when the period is within a pre-determined threshold from a year, display the calculated curve on the timeline of the graphical user interface (obvious in view of combination; Trinh et al discloses using thresholds and showing displayed data based on threshold comparisons (see abstract; paragraphs 0043-0044, 0052, 0066, 0075, 0091, 0104, and 0110); paragraph 0061 of Trinh et al states, “the measurement data from sensors 154 and the setting data 152 may be synchronized to the same timeline and frequency.” Xu et al also teaches displaying data in relation to a period of time based on certain thresholds, as discussed above.)
With respect to claim 12, it would have been obvious to one having ordinary skill in the art before the effective filing date of the invention to incorporate the teachings of DiSalvo into the invention of modified Trinh et al. The motivation for the skilled artisan in doing so is to gain the benefit of contextualizing and visualizing representations of the data.
With respect to claim 13, Trinh et al, as modified, discloses:
wherein the plurality of instructions are further executable by the one or more processors to: modify the operational data associated with the first operational data feature and the first system to compensate for the calculated curve, prior to displaying the operational data on the timeline of the graphical user interface (obvious in view of combination; Trinh et al discloses various changes and modifications, which one of ordinary skill in the art would understand to apply in a variety of contexts. For example, paragraph 0043 states, “a target temperature of a particular component or location of the equipment 150 may be dynamically changed based on other conditions of the equipment 150.” Paragraph 0066 states, “If the anomaly score 460 determined continues to exceed a threshold value while an alert is generated, the system continues presenting the alert and modifies the severity level such as by increasing the severity level.” Paragraph 0120 states, “The disclosure herein is directed to all such variations and modifications to such elements and methods known to those skilled in the art.” Furthermore, Xu et al discloses displaying a wide variety of operational data on a timeline based on user selection features (figures 10-13).)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Sikka et al (US PgPub 20210081841) discloses visually creating and monitoring machine learning models.
McCormick et al (US PgPub 20130104064) discloses customizable vehicle fleet reporting system.
Makhija et al (US Pat 11669310) discloses codeless development of enterprise application.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LEONARD S LIANG whose telephone number is (571)272-2148. The examiner can normally be reached M-F 10:00 AM - 7 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, ARLEEN M VAZQUEZ can be reached on (571)272-2619. 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.
/LEONARD S LIANG/Examiner, Art Unit 2857 01/16/26