DETAILED ACTION
Status of Claims
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 for Application #18/128,151, filed on 11/25/2025, has been entered. The following is a NON-FINAL OFFICE ACTION in response to the request for continued examination.
Claims 1-20 are now pending and have been examined.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The rationale for this finding is explained below.
Per Step 1 of the analysis, the claims are analyzed to determine if they are directed to statutory subject matter. Claim 1 claims an apparatus comprising one or more processors coupled to one or more memories. Therefore, the claim is interpreted as an apparatus. An apparatus is a statutory category for patentability. Claim 16 claims a system comprising one or more processors coupled to one or more memories. Therefore, the claim is interpreted as an apparatus. An apparatus is a statutory category for patentability.
Per Step 2A, Prong 1 of the analysis, the examiner must now determine if the claims recite an abstract idea. In the instant case, the independent claims recite an abstract idea. Specifically, the independent claims 1 and 16 recite “receiving vehicle information including operating conditions from each vehicle in a fleet and historical vehicle information associated with each vehicle, develop one or more vehicle models, select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a difference between an estimated remaining life and an expected remaining life of at least one target component in the fleet, determine a fleet failure probability regarding a likelihood of a component failure in the fleet, the fleet failure probability having a predefined threshold based on the vehicle model and fleet information including at least fleet usage, fleet vehicle types, and fleet engine types, determine a predictive maintenance schedule based on the fleet failure probability, and in response to an operational parameter exceeding the predefined threshold for at least one target component in the fleet, schedule maintenance for the at least one target component.” Therefore, the claims recite an abstract idea, namely a mental process. A human operator with access to the vehicle information and operating conditions, historical vehicle information, fleet usage, and fleet vehicle types could analyze the data, develop a vehicle model such as a flow chart or hierarchy tree, and make a judgment by determining a fleet failure probability and a predictive maintenance schedule. The computer only automates the abstract idea. The limitations previously added by amendment do not change the analysis. The vehicle information being for each vehicle of a fleet can still be analyzed as part of a mental process. The fleet failure probability can still be calculated using basic math and calculations as part of a mental process even if compared to a threshold value. The scheduling of maintenance in response to an operational parameter exceeding a predefined threshold can be done as part of a mental process in which the data is analyzed, and a decision to schedule maintenance is determined based on the analysis. The actual “scheduling,” absent further detail, is not considered to be any kind of technical process but could simply be a decision or a verbal or manual communication. The developing of one or more models in and of itself, absent further detail, is also considered a mathematical formula, which is also an abstract idea. The limitations currently added by amendment do not change the analysis. Algorithms and mathematical models can be used by a human operator as part of a mental process when doing analysis of the data. A mathematical formula such as an algorithm can be selected from among several algorithms based on its past ability to best predict the outcome. The models being run by a machine learning engine only automate the mental process using a computer and that is addressed below. Therefore, the claims still recite a mental process.
Per Step 2A, Prong 2 of the analysis, the examiner must now determine if the claims integrate the abstract idea into a practical application. The additional elements of the claims include an “apparatus,” a “system,” “one or more processors,” and “one or more memories.” However, these recited elements are considered generic recitations of technical elements as they are recited at a high level of generality. These elements are being used as “tools to automate the abstract idea” (see MPEP 2106.05 (f)), and do not integrate the abstract idea into a practical application. They are not recitations of a special purpose computer or transformation (see MPEP 2106.05 (b) and (c)). The additional elements also include “develop one or more vehicle models…using a machine learning engine that receives vehicle information.” However, the use of the machine learning model is recited at a very high level of generality with no detail as to how specifically the model is “used.” Therefore, this additional element is considered the equivalent of “apply it,” or using a computer as a tool to automate the abstract idea (see MPEP 2106.05(f)). Therefore, this additional element does not integrate the abstract idea into a practical application. The claims also include the actual “receiving” of vehicle information.” It is here assumed that the data is received over an electronic network, although this is not recited in the claims. This additional element, absent further detail, is considered “receiving and/or transmission of data over a network,” which is listed in the MPEP 2106.05 (d) (II) (i) as an example of conventional computer functioning- see “receiving or transmitting data over a network” citing TLI Communications, OIP Techs v Amazon.com, buySAFE v Google. Therefore, this additional element does not integrate the abstract idea into a practical application.
Per Step 2B of the analysis, the examiner must now determine if the claims include limitations that are “significantly more” than the abstract idea by demonstrating an improvement to another technology or technical field, an improvement to the functioning of the computer itself, or meaningful limitations beyond generally linking the use of an abstract idea to a particular technological environment. The additional elements of the claims include an “apparatus,” a “system,” “one or more processors,” and “one or more memories.” However, these recited elements are considered generic recitations of technical elements as they are recited at a high level of generality. These elements are being used as “tools to automate the abstract idea” (see MPEP 2106.05 (f)), and are not considered significantly more than the abstract idea itself. They are not recitations of a special purpose computer or transformation (see MPEP 2106.05 (b) and (c)). The additional elements also include “develop one or more vehicle models…using a machine learning engine that receives vehicle information.” However, the use of the machine learning model is recited at a very high level of generality with no detail as to how specifically the model is “used.” Therefore, this additional element is considered the equivalent of “apply it,” or using a computer as a tool to automate the abstract idea (see MPEP 2106.05(f)). Therefore, this additional element is not considered significantly more. The claims also include the actual “receiving” of vehicle information.” It is here assumed that the data is received over an electronic network, although this is not recited in the claims. This additional element, absent further detail, is considered “receiving and/or transmission of data over a network,” which is listed in the MPEP 2106.05 (d) (II) (i) as an example of conventional computer functioning- see “receiving or transmitting data over a network” citing TLI Communications, OIP Techs v Amazon.com, buySAFE v Google. Therefore, this additional element is not considered significantly more than the abstract idea itself.
When considered as an ordered combination, the claim is still considered to be directed to an abstract idea as the claims in the ordered combination simply recite the logical steps for receiving the data, developing a vehicle model, determining a fleet failure probability, and determining a predictive maintenance schedule. Therefore, the ordered combination does not lead to a determination of significantly more.
When considering the dependent claims, claims 2 and 3 are considered part of the abstract idea, as determining a remaining useful life of a target component as well as determining a total life prediction can be done as part of the analysis and judgment of the mental process. Claims 4-7 are considered part of the abstract idea, as determining a failure mileage with an associated percentage or receiving a threshold failure rate and determine when the predicted failure rate is predicted to be achieved can be determined or providing a recommendation of maintenance for the target component or generating a report can be done as part of a mental process, and the processor executing instructions from the memory is considered a generic recitation of a technical element as it is recited at a high level of generality. The processor and memory are being used as “tools to automate the abstract idea” (see MPEP 2106.05 (f)), and are not considered significantly more than the abstract idea itself. They are not recitations of a special purpose computer or transformation (see MPEP 2106.05 (b) and (c)). Claim 8 is considered part of the abstract idea as what the report identifies does not change the analysis. Claims 9 and 10 are considered insignificant extra-solution activity, as the type of data received by the model and the type of vehicle information does not change the manner in which the same steps are performed. The data being received by the model, in such as a training of the model or input, is considered the equivalent of “apply it,” or using a computer as a tool to automate the abstract idea (see MPEP 2106.05(f)). Therefore, this additional element is not considered significantly more than the abstract idea itself. Claims 11 and 14-15 are considered part of the abstract idea, as providing a plurality of maintenance intervals and cost per unit of distance associated with each interval or an events per vehicle value associated with each maintenance interval or a percentage of on road repairs avoided or a mean distance between component failures associated with each predictive maintenance interval can be done as part of a mental process. The “application” used to “display” is considered a generic recitation of a technical element as it is recited at a high level of generality. The application is used as a display means and a “tool to automate the abstract idea” (see MPEP 2106.05 (f)), and is not considered significantly more than the abstract idea itself. Claim 12 is considered part of the abstract idea, as the intervals being described as “predefined” does not change the analysis. Claim 13 is considered part of the abstract idea, and the use of the model is considered the equivalent of “apply it,” or using a computer as a tool to automate the abstract idea (see MPEP 2106.05(f)). Therefore, this additional element is not considered significantly more than the abstract idea itself. The other dependent claims mirror those already discussed above.
Therefore, claims 1-20 are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter. See Alice Corporation Pty. Ltd. Vs. CLS Bank International et al., 2014 (please reference link to updated publicly available Alice memo at http://www.uspto.gov/patents/announce/alice_pec_25jun2014.pdf as well as the USPTO January 2019 Updated Patent Eligibility Guidance.)
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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
Claims 1-10, 14, 16-18, and 20 are rejected under 35 USC 103 as being unpatentable over Senzer, et al. WIPO Publication WO 2020/112337 A1 in view of Abari, et al., Pre-Grant Publication No. 2019/0197798 A1 and in further view of Martin, et al., Pre-Grant Publication No. 2021/0118248 A1.
Regarding claim 1, Senzer teaches:
An apparatus, comprising: one or more processing circuits comprising one or more memory devices coupled to
one or more processors, the one or more memory devices configured to store instructions thereon that, when executed by the one or more processors, cause the one or more processors to (see Figures 1 and 2)
receive vehicle information including operating conditions from each vehicle in a vehicle fleet and historical vehicle information associated with each vehicle (see Figure 2 #240, [0030], and [0059]; see also [0025], [0034] in which multiple vehicles 103a-103n in a fleet of vehicles is taught, [0037], [0059], [0062], and [0083]-[0084] all which make clear that the vehicle information and operating conditions can be for all of multiple vehicles in a vehicle fleet)
develop one or more vehicle models using a machine learning engine that receives the vehicle information (see at least [0066]-[0074] and [0079]-[0083], especially [0072 and [0075]-[0076] in which multiple ML algorithms/models are developed and used)
determine a fleet failure probability regarding a likelihood of component failure in the fleet, the fleet failure probability having a predefined threshold based on the vehicle model and fleet information including at least fleet usage, fleet vehicle types (see at least [0023]-[0027], [0079], and [0094]; see also [0081]-[0082] in which the operational-life predictor predicts operational life left before the failure of a component passes above a predefined threshold by which failure is likely to occur- the examiner notes that it has already been established above that these vehicles are optionally each vehicle in a fleet of vehicles; see also [0094]-[0096] in which a notification threshold is used in which when the vehicle failure probability is above a predefined threshold the fleet manager is notified)
determine a predictive maintenance schedule based on the fleet failure probability (see at least [0021]-[0025] and [0079]-[0082])
in response to an operational parameter exceeding the predefined threshold for at least one target component in the fleet, recommend scheduling maintenance for the at least one target component (see [0006] and [0023] in which a maintenance suggestion notification based on an estimated operational life helps the recipient schedule maintenance; see [0025] in which a remedy for imminent tire failure is to replace the tires; see [0043] in which maintenance monitoring for the fleet of vehicles leads to maintenance recommendations for the fleet manager; see [0078]-[0083] in which operational state fleet failure probability above a threshold leads to notification of the fleet manager with maintenance recommendation for scheduling; see [0085] in which maintenance scheduling recommendations are made based on the failure probability being above a certain level)
Senzer, however, does not appear to specify:
schedule maintenance for the at least one target component
Abari teaches:
schedule maintenance for the at least one target component (see at least Abstract, [0015], and [0018]-[0022] in which vehicles in the fleet are scheduled for maintenance based on various factors including usage, scheduling, time until risk of breakdown, etc.)
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine Abari with Senzer because Senzer already teaches the system notifying a fleet manager of threshold failure probabilities for components and includes recommended maintenance schedules, but does not teach actually scheduling the maintenance, although it does mention the suggestions being so that the scheduling can be done and mentions replacing tires as a result of a determination, and actually scheduling the maintenance would ensure that it is done in a timely fashion, minimizing the possibility of fleet failure.
Senzer and Abari, however, does not appear to specify:
the vehicle model and fleet information including fleet engine types
The examiner, however, takes Official Notice that it is old and well known in the maintenance arts to make a determination of failure probability based on many different factors such as usage, similar vehicle types, maintenance history, and other factors already taught by Senzer, and using engine type. Companies such as Hitachi, Caterpillar, Toyota, Honda, and others have done so for at least one year prior to the effective filing date of the application. Therefore, using the fleet engine type as taught by Abari to modify Senzer would allow for yet another factor in the determination, potentially leading to an even better determination.
Senzer and Abari, however, does not appear to specify:
select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a difference between an estimated remaining life and an expected remaining life of at least one target component in the fleet
Martin teaches:
select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a highest precision value for an expected remaining life of at least one target component in the fleet (see Abstract, [0018], [0048]-[0051], and [0080]-[0085] in which multiple “Remaining Useful Life” ML models are developed for vehicle components, and a model is selected based on an operational parameter that provides for a highest precision value of expected remaining life, which is a measure of prediction error)
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine Martin with Senzer and Abari because Senzer already teaches the system notifying a fleet manager of threshold failure probabilities for components and includes recommended maintenance schedules, but does not teach selecting a best model from among models to determine the expected remaining life and likelihood of failure, and doing so would allow for a “best fit” approach for the particular situation, resulting in the best estimations and better decisions for predictive maintenance.
Senzer, Abari, and Martin, however, does not appear to specify:
prediction error being based on a difference between an estimated remaining life and an expected remaining life
Martin does however teach selection from among multiple “Remaining Useful Life” ML models developed for vehicle component expected remaining useful life, the selection based on a highest precision value of expected remaining life, which is a measure of prediction error, and any determination of highest precision would inherently include some kind of comparison between what the model is predicting as an expectation and what the estimated actual resulting remaining life is going to be or is.
Therefore, it would be obvious to one of ordinary skill in the art at the time of filing of the application to combine prediction error being based on a difference between an estimated remaining life and an expected remaining life with Senzer, Abari, and Martin because Martin teaches selection from among multiple “Remaining Useful Life” ML models developed for vehicle component expected remaining useful life, the selection based on a highest precision value of expected remaining life, which is a measure of prediction error, and any determination of highest precision would inherently include some kind of comparison between what the model is predicting as an expectation and what the estimated actual resulting remaining life is going to be or is, and comparing between an estimated remaining life and an expected remaining life allows for the better and more accurate model to be used for the particular component.
Regarding claim 2, Senzer, Abari, and Martin teaches:
the apparatus of claim 1
Senzer further teaches:
wherein determining the fleet failure probability includes determining a remaining useful life of a target component (see at least [0022]-[0023] and [0081]-[0082] in which the operational life can be estimated for a vehicle at a given time based on the failure probability and timing of one or more components)
Regarding claim 3, Senzer, Abari, and Martin teaches:
the apparatus of claim 2
Senzer further teaches:
wherein determining the fleet failure probability includes determining a total life prediction based on a current life of the target component and the remaining useful life (see [0022]-[0023] and [0025] and [0081]-[0082])
Regarding claim 4, Senzer, Abari, and Martin teaches:
the apparatus of claim 3
Senzer further teaches:
determine a predicted failure mileage with an associated percentage of likelihood (see at least [0081]-[0082] including such as “for example the operational-life predictor 269 may indicate that a vehicle can be driven for 5,000 more miles before a timing chain has above a 90% likelihood of breaking”)
Regarding claim 5, Senzer, Abari, and Martin teaches:
the apparatus of claim 4
Senzer further teaches:
receive a threshold failure rate (see [0081])
query the predicted failure mileage to determine when the failure rate is predicted to be achieved (see [0082] in which historical failure rate data from other vehicles is used by the model to determine when and at what mileage the failure in the current vehicle is likely to happen, and in which the operational-life predictor 269 can establish a correlation between captured vehicle data and a failure based on the prior historical vehicle data; the examiner notes that while the cited portion of the reference does not use the word “query,” the examiner is unsure how the “predicted failure mileage” itself could be “queried,” and further as known in the art, a model or other computer instructions accessing stored data to make some kind of determination, matching, or correlation can be considered a “query” as known in the art)
Regarding claim 6, Senzer, Abari, and Martin teaches:
the apparatus of claim 5
Senzer further teaches:
wherein the predictive maintenance schedule provides a recommendation of maintenance for the target component based on a determination that the failure rate is predicted to be achieved (see Abstract, [0021]-[0023], [0025], [0080]-[0082], and [0085])
Regarding claim 7, Senzer, Abari, and Martin teaches:
the apparatus of claim 1
Senzer further teaches:
generate a report indicating maintenance scheduled within a predetermined period of time in the future (see [0080]-[0085] in which notifications of suggested maintenance for various vehicles in the fleet are generated and sent, and Figure 3 and [0084]-[0086] in which the displayed vehicle profile includes suggested maintenance schedules; the examiner notes that the sending of a notification or displaying of a maintenance schedule for a vehicle is considered substantially similar to and a type of “report”)
Regarding claim 8, Senzer, Abari, and Martin teaches:
the apparatus of claim 7
Senzer further teaches:
wherein the report identifies each component scheduled for maintenance within the period of time (see Figure 3, [0023]-[0025], and [0085] in which components such as tires or oil filter are included in the notifications and in the profile suggestions for maintenance as well as the period of time within which maintenance is being recommended)
Regarding claim 9, Senzer, Abari, and Martin teaches:
the apparatus of claim 1
Senzer further teaches:
wherein the vehicle model receives data lake information including at least one of reliability information, warranty claims, manufacturer information, repair claim history, weather information, fleet vehicle diagnostic information, trip summaries, routing information, or traffic information (see [0019], [0026], and [0049] which teach traffic and weather information, [0027] which teaches repair history, [0043] which teaches manufacturer information, [0049] which teaches a large variety of vehicle information, and at least such as [0020]-[0025] and [0079]-[0083] which teach ML models using this vehicle information)
Regarding claim 10, Senzer, Abari, and Martin teaches:
the apparatus of claim 1
Senzer further teaches:
wherein the vehicle information is specific to at least one of a vehicle, an engine, or an individual component (see [0019], [0026], [0027], [0043], and [0049] which teach vehicle, fleet, and component specific information) as well as [0020]-[0025] and [0079]-[0083])
includes at least one of: a serial number; a model year; a date put into service; an original equipment manufacturer; a vehicle length; a vehicle usage; a repair history including prior repairs, past fault codes, or replaced parts; ambient history including ambient temperature, ambient pressure, and relative humidity; duty cycle including sample rate or duty cycle, usage rate, engine hours, and odometer readings; or historical diagnostic information (see at least [0027] and [0049] which teaches repair history, [0027] and [0051] which teach odometer readings, see [0084] which teaches a VIN number, [0099] which teaches duty cycle and operation hours)
Regarding claim 14, Senzer, Abari, and Martin teaches:
the apparatus of claim 1
Senzer further teaches:
an application communicably coupled to the one or more processing circuits and structured to provide a plurality of predictive maintenance intervals (see at least [0021]-[0025] and [0079]-[0085])
Senzer, Abari, and Martin, however, does not appear to specify:
display an events per vehicle value associated with each predictive maintenance interval
The examiner takes Official Notice that it is old and well known to display various metrics associated with maintenance or other actions, such as cost savings pr vehicle value for particular actions or events when facilitating a report, GUI, API, or interface for management of vehicles, equipment, fleets, construction sites, and other such industrial and business settings. Companies such as Amazon.com, rental car companies, Uber, Caterpillar, and others have done so well prior to the effective filing date of this application.
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine display an events per vehicle value associated with each predictive maintenance interval with Senzer, Abari, and Martin because Senzer already teaches an application displaying predictive maintenance intervals, and displaying an events per vehicle value would allow the user to understand the cost savings and value of each action taken associated with the vehicle.
Regarding claim 16, Senzer teaches:
A system comprising:
one or more processing circuits comprising one or more memory devices coupled to one or more processors, the one or more memory devices configured to store instructions thereon that, when executed by the one or more processors, cause the one or more processors to: (see Figures 1 and 2)
receive vehicle information and fleet information regarding a fleet including at least fleet usage, fleet vehicle types (see Figure 2 #240, [0023]-[0027], [0030], [0059], [0079], and [0094]; (see Figure 2 #240, [0030], and [0059]; see also [0025], [0034] in which multiple vehicles 103a-103n in a fleet of vehicles is taught, [0037], [0059], [0062], and [0083]-[0084] all which make clear that the vehicle information and operating conditions can be for all of multiple vehicles in a vehicle fleet)
develop one or more analytics models based on the received vehicle information and fleet information (see at least [0066]-[0074] and [0079]-[0083], especially [0072 and [0075]-[0076] in which multiple ML algorithms/models are developed and used)
determine a predicted component life of at least one target component using the analytics model (see at least [0020]-[0023] and [0078]-[0083] in which the operational life can be estimated for a vehicle at a given time based on the failure probability and timing of one or more components)
determine a predictive maintenance interval based on the predicted component life of the at least one target component in the fleet (see at least [0021]-[0025] and [0079]-[0082])
generate a report indicating maintenance scheduled within the predictive maintenance interval (see [0080]-[0085] in which notifications of suggested maintenance for various vehicles in the fleet are generated and sent, and Figure 3 and [0084]-[0086] in which the displayed vehicle profile includes suggested maintenance schedules; the examiner notes that the sending of a notification or displaying of a maintenance schedule for a vehicle is considered substantially similar to and a type of “report”)
in response to an operational parameter for the at least one target component in the fleet exceeding a predefined threshold, recommend to schedule maintenance for the at least one target component (see [0006] and [0023] in which a maintenance suggestion notification based on an estimated operational life helps the recipient schedule maintenance; see [0025] in which a remedy for imminent tire failure is to replace the tires; see [0043] in which maintenance monitoring for the fleet of vehicles leads to maintenance recommendations for the fleet manager; see [0078]-[0083] in which operational state fleet failure probability above a threshold leads to notification of the fleet manager with maintenance recommendation for scheduling; see [0085] in which maintenance scheduling recommendations are made based on the failure probability being above a certain level)
an application communicably coupled to the one or more processing circuits and structured to provide the predictive maintenance interval and display the report (see Figures 2-3, [0047], [0068], [0071], and [0080]-[0086] in which the displayed vehicle profile including suggested maintenance schedules is facilitated through an API or other software application)
Senzer, however, does not appear to specify:
schedule maintenance for the at least one target component
Abari teaches:
schedule maintenance for the at least one target component (see at least Abstract, [0015], and [0018]-[0022] in which vehicles in the fleet are scheduled for maintenance based on various factors including usage, scheduling, time until risk of breakdown, etc.)
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine Abari with Senzer because Senzer already teaches the system notifying a fleet manager of threshold failure probabilities for components and includes recommended maintenance schedules, but does not teach actually scheduling the maintenance, although it does mention the suggestions being so that the scheduling can be done and mentions replacing tires as a result of a determination, and actually scheduling the maintenance would ensure that it is done in a timely fashion, minimizing the possibility of fleet failure.
Senzer and Abari, however, does not appear to specify:
the vehicle information and fleet information including fleet engine types
The examiner, however, takes Official Notice that it is old and well known in the maintenance arts to make a determination of failure probability based on many different factors such as usage, similar vehicle types, maintenance history, and other factors already taught by Senzer, and using engine type. Companies such as Hitachi, Caterpillar, Toyota, Honda, and others have done so for at least one year prior to the effective filing date of the application. Therefore, using the fleet engine type as taught by Abari to modify Senzer would allow for yet another factor in the determination, potentially leading to an even better determination.
Senzer and Abari, however, does not appear to specify:
select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a difference between an estimated remaining life and an expected remaining life of at least one target component in the fleet
Martin teaches:
select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a highest precision value for an expected remaining life of at least one target component in the fleet (see Abstract, [0018], [0048]-[0051], and [0080]-[0085] in which multiple “Remaining Useful Life” ML models are developed for vehicle components, and a model is selected based on an operational parameter that provides for a highest precision value of expected remaining life, which is a measure of prediction error)
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine Martin with Senzer and Abari because Senzer already teaches the system notifying a fleet manager of threshold failure probabilities for components and includes recommended maintenance schedules, but does not teach selecting a best model from among models to determine the expected remaining life and likelihood of failure, and doing so would allow for a “best fit” approach for the particular situation, resulting in the best estimations and better decisions for predictive maintenance.
Senzer, Abari, and Martin, however, does not appear to specify:
prediction error being based on a difference between an estimated remaining life and an expected remaining life
Martin does however teach selection from among multiple “Remaining Useful Life” ML models developed for vehicle component expected remaining useful life, the selection based on a highest precision value of expected remaining life, which is a measure of prediction error, and any determination of highest precision would inherently include some kind of comparison between what the model is predicting as an expectation and what the estimated actual resulting remaining life is going to be or is.
Therefore, it would be obvious to one of ordinary skill in the art at the time of filing of the application to combine prediction error being based on a difference between an estimated remaining life and an expected remaining life with Senzer, Abari, and Martin because Martin teaches selection from among multiple “Remaining Useful Life” ML models developed for vehicle component expected remaining useful life, the selection based on a highest precision value of expected remaining life, which is a measure of prediction error, and any determination of highest precision would inherently include some kind of comparison between what the model is predicting as an expectation and what the estimated actual resulting remaining life is going to be or is, and comparing between an estimated remaining life and an expected remaining life allows for the better and more accurate model to be used for the particular component.
Regarding claim 17, Senzer, Abari, and Martin teaches:
the system of claim 16
Senzer further teaches:
determining a remaining useful life of the at least one target component using the analytics model (see at least [0020]-[0023] and [0081]-[0082] in which the operational life can be estimated for a vehicle at a given time based on the failure probability and timing of one or more components)
determining a total life prediction based on a current life of the at least one target component and the remaining useful life (see [0022]-[0023] and [0025] and [0081]-[0082])
receiving a threshold failure rate from the application and determining when the threshold failure rate is predicted to be achieved based on the remaining useful life (see at least [0081]-[0082] including such as “for example the operational-life predictor 269 may indicate that a vehicle can be driven for 5,000 more miles before a timing chain has above a 90% likelihood of breaking”; see also [0081] in which a threshold failure rate is received; see also [0082] in which historical failure rate data from other vehicles is used by the model to determine when and at what mileage the failure in the current vehicle is likely to happen, and in which the operational-life predictor 269 can establish a correlation between captured vehicle data and a failure based on the prior historical vehicle data)
Regarding claim 18, Senzer, Abari, and Martin teaches:
the system of claim 17
Senzer further teaches:
wherein the indicated scheduled maintenance in the report is based on a determination that the predefined threshold is predicted to be achieved (see [0020]-[0023], [0078]-[0085], and [0094]-[0096] in which notifications of suggested maintenance for various vehicles in the fleet are generated and sent based on predicted failures of components or vehicles above a threshold, and Figure 3 and [0083]-[0086] and [0094]-[0096] in which the displayed vehicle profile includes suggested maintenance schedules based on predicted failure rates; the examiner notes that the sending of a notification or displaying of a maintenance schedule for a vehicle is considered substantially similar to and a type of “report”)
Regarding claim 20, Senzer teaches:
A method comprising:
receiving vehicle information and fleet information (see Figure 2 #240, [0030], and [0059])
develop one or more vehicle models based on the received vehicle information and fleet information (see at least [0066]-[0074] and [0079]-[0083], especially [0072 and [0075]-[0076] in which multiple ML algorithms/models are developed and used)
determining a total life prediction using the analytics model and based on a current life of at least one target component in a fleet and a remaining useful life of the at least one target component in the fleet (see at least [0020]-[0025] and [0079]-[0086]; see also see also [0025], [0034] in which multiple vehicles 103a-103n in a fleet of vehicles is taught, [0037], [0059], [0062], and [0083]-[0084] all which make clear that the vehicle information and operating conditions can be for all of multiple vehicles in a vehicle fleet)
receiving a fleet failure rate (see at least [0020]-[0023], [0078]-[0085], and [0094]-[0096]; the examiner notes that the “fleet failure rate” based on the claim language could be the failure rate of one component in two or more vehicles, as nothing more than that is required)
determining when the failure rate is predicted to be achieved based on the total life prediction (see at least [0081]-[0082] including such as “for example the operational-life predictor 269 may indicate that a vehicle can be driven for 5,000 more miles before a timing chain has above a 90% likelihood of breaking”; see also [0082] in which historical failure rate data from other vehicles is used by the model to determine when and at what mileage the failure in the current vehicle is likely to happen, and in which the operational-life predictor 269 can establish a correlation between captured vehicle data and a failure based on the prior historical vehicle data; see also (see at least [0023]-[0027], [0079], and [0094]; see also [0081]-[0082] in which the operational-life predictor predicts operational life left before the failure of a component passes above a predefined threshold by which failure is likely to occur- the examiner notes that it has already been established above that these vehicles are optionally each vehicle in a fleet of vehicles; see also [0094]-[0096] in which a notification threshold is used in which when the vehicle failure probability is above a predefined threshold the fleet manager is notified)
determining a predictive maintenance interval based on the total life prediction (see at least [0021]-[0025], [0079]-[0085], and [0094]-[0096]; see especially [0081]-[0082] including such as “for example the operational-life predictor 269 may indicate that a vehicle can be driven for 5,000 more miles before a timing chain has above a 90% likelihood of breaking”; see also [0082] in which historical failure rate data from other vehicles is used by the model to determine when and at what mileage the failure in the current vehicle is likely to happen, and in which the operational-life predictor 269 can establish a correlation between captured vehicle data and a failure based on the prior historical vehicle data)
in response to an operational parameter exceeding a predefined threshold for the at least one target component in the fleet, recommending scheduling maintenance for the at least one target component (see [0006] and [0023] in which a maintenance suggestion notification based on an estimated operational life helps the recipient schedule maintenance; see [0025] in which a remedy for imminent tire failure is to replace the tires; see [0043] in which maintenance monitoring for the fleet of vehicles leads to maintenance recommendations for the fleet manager; see [0078]-[0083] in which operational state fleet failure probability above a threshold leads to notification of the fleet manager with maintenance recommendation for scheduling; see [0085] in which maintenance scheduling recommendations are made based on the failure probability being above a certain level;
generating a report indicating maintenance scheduled within the predictive maintenance interval (see [0080]-[0085] in which notifications of suggested maintenance for various vehicles in the fleet are generated and sent, and Figure 3 and [0084]-[0086] in which the displayed vehicle profile includes suggested maintenance schedules; the examiner notes that the sending of a notification or displaying of a maintenance schedule for a vehicle is considered substantially similar to and a type of “report”)
displaying the report (see Figures 2-3, [0047], [0068], [0071], and [0080]-[0086]
Senzer, however, does not appear to specify:
generating a report indicating when the fleet failure rate is predicted to be achieved based on the total life prediction
Senzer does however teach generating and displaying a report (see Figures 2-3, [0047], [0068], [0071], and [0080]-[0086]) as well as determining when the threshold failure rate is predicted to be achieved based on the total life prediction (see at least [0081]-[0082] including such as “for example the operational-life predictor 269 may indicate that a vehicle can be driven for 5,000 more miles before a timing chain has above a 90% likelihood of breaking”; see also [0082] in which historical failure rate data from other vehicles is used by the model to determine when and at what mileage the failure in the current vehicle is likely to happen, and in which the operational-life predictor 269 can establish a correlation between captured vehicle data and a failure based on the prior historical vehicle data).
Therefore, it would be obvious to one of ordinary skill in the art to combine generating a report indicating when the fleet failure rate is predicted to be achieved based on the total life prediction with Senzer because Senzer already generates reports on operational states of vehicles and vehicle components as well as maintenance suggestions and schedules as well as does use a total life prediction and determines threshold failure rate timing for components and vehicles, and displaying the results of these determinations would allow the viewer to understand the calculations and reasonings behind the determination of the maintenance recommendations.
Senzer, however, does not appear to specify:
schedule maintenance for the at least one target component
Abari teaches:
schedule maintenance for the at least one target component (see at least Abstract, [0015], and [0018]-[0022] in which vehicles in the fleet are scheduled for maintenance based on various factors including usage, scheduling, time until risk of breakdown, etc.)
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine Abari with Senzer because Senzer already teaches the system notifying a fleet manager of threshold failure probabilities for components and includes recommended maintenance schedules, but does not teach actually scheduling the maintenance, although it does mention the suggestions being so that the scheduling can be done and mentions replacing tires as a result of a determination, and actually scheduling the maintenance would ensure that it is done in a timely fashion, minimizing the possibility of fleet failure.
Senzer and Abari, however, does not appear to specify:
select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a difference between an estimated remaining life and an expected remaining life of at least one target component in the fleet
Martin teaches:
select one of the vehicle models based on an operational parameter that provides for a least prediction error, the least prediction error being based on a highest precision value for an expected remaining life of at least one target component in the fleet (see Abstract, [0018], [0048]-[0051], and [0080]-[0085] in which multiple “Remaining Useful Life” ML models are developed for vehicle components, and a model is selected based on an operational parameter that provides for a highest precision value of expected remaining life, which is a measure of prediction error)
It would be obvious to one of ordinary skill in the art before the effective date of filing of the application to combine Martin with Senzer and Abari because Senzer already teaches the system notifying a fleet manager of threshold failure probabilities for components and includes recommended maintenance schedules, but does not teach selecting a best model from among models to determine the expected remaining life and likelihood of failure, and doing so would allow for a “best fit” approach for the particular situation, resulting in the best estimations and better decisions for predictive maintenance.
Senzer, Abari, and Martin, however, does not appear to specify:
prediction error being based on a difference between an estimated remaining life and an expected remaining life
Martin does however teach selection from among multiple “Remaining Useful Life” ML models developed for vehicle component expected remaining useful life, the selection based on a highest precision value of expected remaining life, which is a measure of prediction error, and any determination of highest precision would inherently include some kind of comparison between what the model is predicting as an expectation and what the estimated actual resulting remaining life is going to be or is.
Therefore, it would be obvious to one of ordinary skill in the art at the time of filing of the application to combine prediction error being based on a difference between an estimated remaining life and an expected remaining life with Senzer, Abari, and Martin because Martin teaches selection from among multiple “Remaining Useful Life” ML models developed for vehicle component expected remaining useful life, the selection based on a highest precision value of expected remaining life, which is a measure of prediction error, and any determination of highest precision would inherently include some kind of comparison between what the model is predicting as an expectation and what the estimated actual resulting remaining life is going to be or is, and comparing between an estimated remaining life and an expected remaining life allows for the better and more accurate model to be used for the particular component.
Claims 11-13 are rejected under 35 U.S.C. 103 as being unpatentable over Senzer, et al. WIPO Publication WO 2020/112337 A1 in view of Abari, et al., Pre-Grant Publication No. 2019/0197798 A1 and in further view of Martin, et al., Pre-Grant Publication No. 2021/0118248 A1 and in further view of Schaaf, Pre-Grant Publication No. 2017/0270718 A1.
Regarding claim 11, Senzer, Abari, and Martin teaches:
the apparatus of claim 1
Senzer, Abari,