DETAILED ACTION
Notice of 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 .
Status of the Application
Claims 1-9 and 11-21 were pending and were rejected in the previous office action. Claims 1, 5, 7, 11-12, 14-18, and 20 were amended. Claims 9, 13, and 21 were canceled. New claims 22 and 23 were added.
Claims 1-8, 11-12, 14-20, and 22-23 remain pending and are examined in this office action.
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 6/4/2026 has been entered.
Response to Arguments
Claim Objections:
Claim 21 was previously objected to. The previous objection is withdrawn as claim 21 is cancelled.
35 USC § 101:
Applicant’s arguments regarding the § 101 rejection of claims 1-9 and 11-21 (pgs. 11-14, remarks filed 6/4/2026) have been fully considered but are not persuasive. Claims 9, 13, and 21 are canceled.
Applicant argues that the amended features for training a plurality of machine learning models on a plurality of variables relating to historical service capacity, iteratively testing the machine learning models to identify a most important subset of variables (highest correlation with capacity statuses of repair facilities), and retraining the machine learning models to execute with the subset of variables, improves the performance of machine-learning models and thereby overcomes the previous § 101 rejection at Step 2A Prong Two/Step 2B. Ex Parte Desjardins is cited for support. (See pgs. 11-14, remarks).
However, the examiner respectfully disagrees that the claims, as currently recited, reflect the argued improvement disclosed in the specification. The current claims recite limitations to “train a plurality of machine-learning models to correlate historical service data to a respective probability that any of the plurality of repair facilities is over-capacity, the historical service data including a plurality of variables including at least geographic location of the respective repair facility, vehicle damage attributes, and repair services attributes, wherein the training includes: iteratively testing the plurality of machine-learning models to identify a subset of the plurality of variables of the historical service data having a highest correlation with respective known capacity statuses of the plurality of repair facilities; and retraining the plurality of machine-learning models to execute with the subset of variables” – which indicates that the machine-learning models are iteratively trained based on more relevant variables. However, see Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025), showing “[P]atents that do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied, are patent ineligible under § 101.” In that case, similar to here, “[t]he requirements that the machine learning model be ‘iteratively trained’ or dynamically adjusted in the Machine Learning Training patents do not represent a technological improvement” because “[i]terative training using selected training material and dynamic adjustments based on real-time changes are incident to the very nature of machine learning.” Id. at 1212. Therefore, the Recentive Analytics decision makes clear that iterative training of machine learning models to improve the output alone is incident to the very nature of machine learning (i.e. represents the use of machine learning in its ordinary capacity) and does not provide a technological improvement.
While the examiner generally agrees that concepts from the specification potentially amount to an improvement to the performance of the machine learning model (e.g. maintaining or improving performance of the machine learning model while reducing the size of the training dataset and the associated computational resources), the claims do not specifically exclude the variables that are determined not to have an impact on the performance of the machine learning model, and thus, read on simply iteratively training on an additional subset of data.
Suggestion to overcome the current § 101 rejection: In order to overcome the current § 101 rejection, the examiner suggests including a step prior to the “retraining” step which specifies a limitation along the lines of “removing variables that do not affect the predictions of the plurality of machine-learning models, while only retaining the subset of variables having the highest correlation with respective known capacity statuses of the plurality of repair facilities.” Such an amendment would reflect the improvement related to how the model(s) are trained as disclosed in the specification and argued by applicant in the remarks, distinguish the claims from mere general retraining on additional data (as in Recentive), and follow sufficiently similar fact patterns identified in Ex Parte Desjardins.
35 USC § 103:
Applicant’s arguments regarding the previous § 103 rejections of claims 1-9 and 11-21 (pgs. 14-15, remarks filed 6/4/2026) have been fully considered and are persuasive.
The amendments to independent claims 1, 18 and 20 overcome the previous and newly cited prior art for the reasons discussed in detail below. The previous § 103 rejections are withdrawn.
Claim Objections
Claims 1, 14, 16, 18, and 20 are objected to because of the following informalities:
Claim 1 (and similar claims 18 and 20) recite “determine the one or more repair facilities most likely to have availability…” but lacks antecedent basis for “the one or more repair facilities most likely to have availability.” While this issue does not render the claim unclear, the examiner suggests amending this limitation to specify “determine [[the]] one or more repair facilities most likely to have availability” or a similar amendment to avoid any potential confusion as to what “the one or more repair facilities most likely to have availability” is referring to.
Claims 14 and 16 refer to “the user computing device,” whereas claim 1 previously recited “the user computer device.” Therefore, this appears to be a typo that is referring to “the user computer device.”
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-8, 11-12, 14-20, and 22-23 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e. an abstract idea) without significantly more.
Step 1:
Claims 1-8, 11-12, and 14-17 recite “A computer system…” (i.e. a machine); claims 18-19 and 22-23 recite “A computer-implemented method…” (i.e. a process); and claim 20 recites “At least one non-transitory computer-readable storage media having computer-executable instructions embodied thereon…” (i.e. an article of manufacture). These claims fall under one of the four categories of statutory subject matter and as a result, pass Step 1 of the subject matter eligibility test. However, “Determining that a claim falls within one of the four enumerated categories of patentable subject matter recited in 35 U.S.C. 101 (i.e., process, machine, manufacture, or composition of matter) in Step 1 does not end the eligibility analysis, because claims directed to nothing more than abstract ideas (such as a mathematical formula or equation), natural phenomena, and laws of nature are not eligible for patent protection.” See MPEP 2106.04. Accordingly, the examiner continues the subject matter eligibility analysis below.
Step 2A Prong One:
Independent claims 1, 18 and 20 recite limitations (additional elements omitted) for monitoring of a workload for a plurality of repair facilities (claims 1/18) and for detecting and acting upon operator reliance to vehicle alerts (claim 20), including limitations to:
generate initial workload rankings for a plurality of repair facilities…
identify a subset of the plurality of variables of the historical service data having a highest correlation with respective known capacity statuses of the plurality of repair facilities;
collect, from the plurality of repair facilities, a plurality of current service data for the plurality of repair facilities;
receive…a query requesting service for repairing a user vehicle provided by one or more of the repair facilities, the query including a geographic location of a user computer device;
using the subset of variables of the current service data and content of the query as inputs…using an output…adjust the initial workload rankings for the plurality of repair facilities based upon the probability, for each of the plurality of repair facilities, that the respective repair facility is over-capacity;
determine the one or more repair facilities most likely to have availability to provide the service for repairing the user vehicle based upon the adjusted workload rankings of the plurality of repair facilities; and
display…a list of the one or more determined repair facilities
The limitations of independent claims 1, 18 and 20 above are determined to recite an abstract idea (ranking a plurality of repair facilities, receiving a repair request to repair a vehicle, adjusting the ranking based upon current service data and a probability of each repair facility being over-capacity, and determining and displaying one or more available repair facilities to service the repair request) for the reasons discussed in the following continued Step 2A Prong One analysis. Note that “An abstract idea can generally be described at different levels of abstraction.” Apple, Inc. v. Ameranth, Inc., 842 F.3d 1229, 1240-41 (Fed. Cir. 2016).
As described in MPEP 2106.04(a)(2)(II), claim limitations which recite commercial or legal interactions (including agreements in the form of contracts, legal obligations, advertising, marketing or sales activities or behaviors, and business relations) or managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) fall into the “certain methods of organizing human activity” category of judicial exceptions. Therefore, since the processes described by the limitations above amount to a commercial interaction and managing interactions between people (i.e. ranking a plurality of repair facilities, receiving a repair request to repair a vehicle, adjusting the ranking based upon current service data and a probability of each repair facility being over-capacity, and determining and displaying one or more available repair facilities to service the repair request), the claims fall into the “certain methods of organizing human activity” grouping of abstract ideas.
As described in MPEP 2106.04(a)(2)(III), “[T]he "mental processes" abstract idea grouping is defined as concepts performed in the human mind, and examples of mental processes include observations, evaluations, judgments, and opinions.” and “If a claim recites a limitation that can practically be performed in the human mind, with or without the use of a physical aid such as pen and paper, the limitation falls within the mental processes grouping, and the claim recites an abstract idea.” The limitations recited by the representative independent claims 1, 18 and 20 above, under the broadest reasonable interpretation and but for the use of generic computer components, cover concepts (e.g. observation, evaluation, judgment, and opinion) that can reasonably be performed in the human mind or by the human mind with the aid of simple tools such as pen and paper. For example, the “collect” and “receive” steps amounts to an observation, while the “generate,” “identify,” “adjust,” and “determine” steps are evaluations, judgments, or opinions. Furthermore, displaying a list of the one or more repair facilities (but for the use of generic computers/computer components, which is analyzed below) is analogous to outputting a list of the one or more repair facilities by a human, e.g. writing down a list using pen and paper. Therefore, as the processes above described by the representative independent claims 1, 18 and 20 can be characterized as mental processes (i.e. observation, evaluation, judgment, and opinion), but for the recitation of generic computer components in the claims, the claims fall under the “mental processes” category of judicial exceptions (i.e. abstract ideas).
As claims 1, 18 and 20 are identified by the examiner as reciting concepts that fall under more than one abstract idea grouping (i.e. “certain methods of organizing human activity” and “mental processes”), the examiner considers the limitations together as a single abstract idea for the purposes of the Step 2A Prong Two and Step 2B analysis, in accordance with MPEP 2106.04(II)(B).
Step 2A Prong Two:
The judicial exception (i.e. abstract idea) recited in claims 1, 18 and 20 is not integrated into a practical application because the claims recite mere instructions to apply the abstract idea (i.e. ranking a plurality of repair facilities, receiving a repair request to repair a vehicle, adjusting the ranking based upon current service data and a probability of each repair facility being over-capacity, and determining and displaying one or more available repair facilities to service the repair request) using generic computers/computer components (i.e. “A computer system…comprising at least one processor in communication with at least one memory device, the computer system in communication with a user computer device associated with a user, the at least one processor is programmed to…,” “a plurality of machine-learning models”/”at least one machine-learning model of the plurality of machine-learning models,” “a user computer device,” and “an interactive user interface” displayed on the computer computing device of claim 1; “computer-implemented method…performed by one or more processors in communication with a memory,” “a plurality of machine-learning models”/“at least one machine-learning model of the plurality of machine-learning models,” “a user computer device,” and “an interactive user interface” displayed on the user computing device of claim 18; and “At least one non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein when executed by a computer system…including one or more processors and a memory, the computer-executable instructions cause the one or more processors to…,”“a plurality of machine-learning models”/”at least one machine-learning model of the plurality of machine-learning models,” “a user computer device,” and “an interactive user interface” displayed on the user computing device of claim 20). See MPEP 2106.05(f), showing “[C]laims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible. Alice Corp.,” and that the use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general-purpose computer or computer components after the fact to an abstract idea does not integrate a judicial exception into a practical application.
While the claims recite the execution of a plurality of machine-learning models trained on a plurality of variables for historical service data for the plurality of repair facilities to provide output used to predict a probability that respective repair facilities are over service capacity and adjust workload rankings of the plurality of repair facilities (claims 1, 18 and 20), there is no indication in the claims or the specification that the claimed invention improves machine-learning technology itself, but instead recite generic machine learning models to receive various input data and generate an output (adjust workload rankings). The claims also recite limitations to “train a plurality of machine-learning models to correlate historical service data to a respective probability that any of the plurality of repair facilities is over-capacity, the historical service data including a plurality of variables including at least geographic location of the respective repair facility, vehicle damage attributes, and repair services attributes, wherein the training includes: iteratively testing the plurality of machine-learning models to identify a subset of the plurality of variables of the historical service data having a highest correlation with respective known capacity statuses of the plurality of repair facilities; and retraining the plurality of machine-learning models to execute with the subset of variables” – which indicates that the machine-learning models are iteratively trained and tested based on a subset of the most relevant variables. However, see Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025), showing “[P]atents that do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied, are patent ineligible under § 101.” In that case, similar to here, “[t]he requirements that the machine learning model be ‘iteratively trained’ or dynamically adjusted in the Machine Learning Training patents do not represent a technological improvement” because “[i]terative training using selected training material and dynamic adjustments based on real-time changes are incident to the very nature of machine learning.” Id. at 1212. The Recentive Analytics decision makes clear that iterative training of machine learning models to improve the output alone is incident to the very nature of machine learning (i.e. represents the use of machine learning in its ordinary capacity) and does not provide a technological improvement. Thus, the steps for training, testing, retraining, and “executing” the machine learning models as claimed represents the use of existing machine learning technology and generic computer implementation (generic machine learning models) as a tool to apply the abstract idea.
In addition, the use of the user computer device to provide a query and display information on an interactive user interface (i.e. display one or more determined repair facilities) also merely adds generic computer implementation to output information, rather than anything that integrates the abstract idea into a practical application. The additional elements recited in the claims do not represent an improvement to machine learning technology and/or the recited machine learning models, improve the functioning of any computers, user devices, or user interfaces, or otherwise improve any other technology. Therefore, because the claims, considered as a whole, do not recite anything that integrates the abstract idea into a practical application, the claims are directed to an abstract idea.
Step 2B:
Claims 1, 18 and 20 do not include additional elements, whether considered alone or as an ordered combination, that are sufficient to amount to significantly more than the judicial exception (i.e. abstract idea) because as mentioned above, the claims recite mere instructions to apply the abstract idea (i.e. ranking a plurality of repair facilities, receiving a repair request to repair a vehicle, adjusting the ranking based upon current service data and a probability of each repair facility being over-capacity, and determining and displaying one or more available repair facilities to service the repair request) using generic computers/computer components (i.e. “A computer system…comprising at least one processor in communication with at least one memory device, the computer system in communication with a user computer device associated with a user, the at least one processor is programmed to…,” “a plurality of machine-learning models”/“at least one machine-learning model of the plurality of machine-learning models,” “a user computer device,” and “an interactive user interface” displayed on the computer computing device of claim 1; “computer-implemented method…performed by one or more processors in communication with a memory,” “a plurality of machine-learning models”/”at least one machine-learning model of the plurality of machine-learning models,” “a user computer device,” and “an interactive user interface” displayed on the user computing device of claim 18; and “At least one non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein when executed by a computer system…including one or more processors and a memory, the computer-executable instructions cause the one or more processors to…,”“a plurality of machine-learning models”/”at least one machine-learning model of the plurality of machine-learning models,” “a user computer device,” and “an interactive user interface” displayed on the user computing device of claim 20).
As mentioned above, the recited steps for training, testing, retraining (i.e. iteratively training the machine learning models), and execution of a plurality of machine-learning models trained on a plurality of variables for historical service data for the plurality of repair facilities to provide output used to predict a probability that respective repair facilities are over service capacity and adjust workload rankings of the plurality of repair facilities (claims 1, 18 and 20) does not improves machine-learning technology itself, but instead recites the use of machine learning technology in its ordinary capacity and using the machine learning models to apply the abstract idea. See Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025), showing “[P]atents that do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied, are patent ineligible under § 101.” In that case, similar to here, “[t]he requirements that the machine learning model be ‘iteratively trained’ or dynamically adjusted in the Machine Learning Training patents do not represent a technological improvement” because “[i]terative training using selected training material and dynamic adjustments based on real-time changes are incident to the very nature of machine learning.” Id. at 1212. Thus, the steps for training, testing, retraining, and “executing” the machine learning models as claimed does not provide a technological improvement and represents the use of existing machine learning technology and generic computer implementation (generic machine learning models) as a tool to apply the abstract idea.
In addition, the use of the user computer device to provide a query and display information on an interactive user interface (i.e. display one or more determined repair facilities) also merely adds generic computer implementation to output information, rather than anything that integrates the abstract idea into a practical application.
The additional elements recited in the claims do not improve the functioning of any computers, or otherwise provide an improvement to any other technology, and considering the additional elements as an ordered combination does not add significantly more than the abstract idea. Therefore, claims 1, 18 and 20 recite an abstract idea without significantly more.
Dependent Claims 2-8, 11-12, 14-17, 19, and 22-23:
Claims 2-8, 11-12, 14-17, 19, and 22-23 are directed to the same abstract idea as independent claims 1 and 18 above without significantly more, as they do not recite anything that integrates the abstract idea into a practical application or amounts to significantly more than the abstract idea.
Claims 2, 14, and 19 recite the following limitations, which further describe the abstract idea and recite mere instructions to apply the abstract idea using generic computers/generic computer implementation: “wherein the at least one processor is further programmed to receive the plurality of current service data from a plurality of computer devices associated with the plurality of repair facilities” (claim 2); “the at least one processor is further programmed to generate a user interface to provide vehicle repair information based on the geographic location of the user computing device” (claim 14 – which uses a generic user interface to display an output based on location information); and “receiving the plurality of current service data from a plurality of computer devices associated with the plurality of repair facilities” (claim 19).
Claims 3-6, 8, 11, 17, and 22 recite the following limitations, which do not add any additional elements beyond those already addressed above (e.g. at least one processor and the one or more machine learning models), but merely further describe the abstract idea being carried out above by reciting limitations for: “wherein the plurality of current service data includes a current workload of repairing a plurality of vehicles…” (claim 3); “automatically remove a repair facility from a list…” (claim 4); “remove a repair facility from the one or more determined repair facilities…” (claim 5 and similar claim 22); “receive a plurality of performance data from the plurality of repair facilities…” (claim 6); “determine a condition of the user vehicle to be repaired; and re-rank the plurality of repair facilities…” (claim 8); “reduce a workload ranking for a repair facility…” (claim 11); and “rank the plurality of repair facilities in the corresponding geographic regions” (claim 17).
Claims 6, 7, 11-12, 16, and 23 recite limitations to: “retrain the plurality of machine-learning models based upon the plurality of performance data” (claim 6); “collect a plurality of updated current service data…and iterative re-execute the plurality of machine-learning models…to iteratively generate updated workload rankings…” (claim 7 and similar claim 23); “when the one or more machine-learning models determines that the repair facility is over capacity” (claim 11); “wherein iterative testing of the plurality of machine-learning models results in identifying a variable of average time to completion as highly correlated with the probability that the respective repair facility is over-capacity, and wherein the re-training comprises re-training the plurality of machine-learning models to output a high probability that a respective repair facility is over capacity when the respective average time to completion is greater than or equal to 45 days” (claim 12); and “select the at least one machine-learning to execute based on the geographic location of the user computing device” (claim 16). These limitations do not add anything that indicates an improvement to machine-learning models or machine-learning technology, but instead describe limitations that are consistent with the ordinary functioning of basic machine learning models (training, re-training, testing, updating, executing, and selecting the machine learning models to receive an input and generate an output) and use generic machine learning and generic computer components to apply the abstract idea.
Claim 15 also further describes the field of use of the machine-learning models (“wherein each model of the plurality of machine-learning models represents a different geographic region, and wherein each machine-learning model is configured to model a respective plurality of repair facilities in the corresponding geographic region” of claim 15. Thus, the limitations of claim 15 at best generally link the performance of the abstract idea to a particular technical environment, but do not add anything that integrates the abstract idea into a practical application or adds significantly more.
Therefore, claims 1-8, 11-12, 14-20, and 22-23 are ineligible under § 101.
Novelty/Non-Obviousness
Claims 1-8, 11-12, 14-20, and 22-23 are novel and nonobvious over the closest prior art for the following reasons:
With respect to independent claim 1,
US 20180082379 A1 (Kelsh) teaches: A computer system for real-time monitoring of a workload for a plurality of repair facilities (Kelsh: ¶ 0030, ¶ 0196-0198 and Fig. 10 showing computing system implementing methods described herein, including one or more processors in communication with memory and executing instructions stored in memory; ¶ 0039-0041, ¶ 0139-0142 showing “Find a Repair Shop” functionality and receiving information on repair shop availability),
the computer system comprising at least one processor in communication with at least one memory device (Kelsh: ¶ 0030, ¶ 0196-0198 memory storing instructions executed by processor),
the computer system in communication with a user computer device associated with a user (Kelsh: Fig. 1, ¶ 0022, ¶ 0033 showing user device 120 in communication with pre-FNOL system computer including at least repair clearinghouse server), the at least one processor is programmed to:
generate workload rankings for a plurality of repair facilities in communication with the computer system (Kelsh: ¶ 0141 showing “each of the repair shops may list availability based on the repair colors (e.g., green, yellow, and red)…” and ¶ 0142 and ¶ 0182 showing “generate a sortable list…The sortable list may display each of the repair shops within the geographical location associated with the mobile device 120 of the user…The user of mobile device 120 may be able to sort the repair shop data comprised within the sortable list based on factors such as highest repair ranking, nearest location, repair color availability, and the like”; also see ¶ 0039, ¶ 0139-0141, ¶ 0189 showing repair shops provide availability data, e.g. ¶ 0141 “Additionally, each of the repair shops may list availability based on the repair colors (e.g., green, yellow, and red). For example, a repair shop proximate to the user's geographical location may have ten slots per month available for green repairs, seven slots per month available for yellow repairs, and two slots per month available for red repairs…For example, if a vehicle determined as a green repair schedules a repair at a repair shop with no green repairs underway and/or scheduled, the forecasted repair completion time may be one day. Conversely, if a vehicle determined as a green repair schedules a repair at a repair shop with no green repairs underway but six green repairs scheduled ahead of the vehicle, the forecasted repair completion time may be six days” and see ¶ 0039 showing “repair clearinghouse database 174 may store data associated with repair shops, car rental agencies, and tow truck companies such as profiles, user ratings, availability schedules, and the like”)
collect, from the plurality of repair facilities, a plurality of current service data for the plurality of repair facilities (Kelsh: ¶ 0141 “each of the repair shops may list availability based on the repair colors (e.g., green, yellow, and red)…a repair shop… may have ten slots per month available for green repairs, seven slots per month available for yellow repairs, and two slots per month available for red repairs. The forecasted repair completion time for each of the respective repair colors may vary within the approximate ranges based on the number of vehicles associated with the specific color currently undergoing repair and/or scheduled to be repaired. For example, if a vehicle determined as a green repair schedules a repair at a repair shop with no green repairs underway and/or scheduled, the forecasted repair completion time may be one day. Conversely, if a vehicle determined as a green repair schedules a repair at a repair shop with no green repairs underway but six green repairs scheduled ahead of the vehicle, the forecasted repair completion time may be six days”, and ¶ 0039 “repair clearinghouse database 174 may store data associated with repair shops, car rental agencies, and tow truck companies such as profiles, user ratings, availability schedules, and the like”);
receive, from a user computing device, a query requesting service for repairing a user vehicle provided by one or more of the repair facilities (Kelsh: ¶ 0181-0182 showing user selects, using their mobile device, a “Find a Repair Shop” option, or ¶ 0187-0188 showing user selects “Yes” to request assistance in identifying a repair shop to repair the damages to their vehicle).
Kelsh further teaches determining a plurality of service providers including repair shops in response to a user query and using current service data from the repair shops indicating availability to generate a ranking of the repair shops/service providers as per above (Kelsh: ¶ 0039, ¶ 0139-0142, ¶ 0182, ¶ 0189), wherein the list may be sorted, i.e. re-ranked according to various criteria such as availability for a repair color/repair type availability (Kelsh: ¶ 0142, ¶ 0182), and further teaches considering historical service data including user ratings (Kelsh: ¶ 0039);
determine the one or more repair facilities having availability based upon the adjusted workload rankings of the plurality of repair facilities to provide the service for repairing the user vehicle (Kelsh: ¶ 0182, ¶ 0188 showing repair clearinghouse server generates a sortable list displaying each of the repair shops within the geographical location associated with the mobile device 120 of the user, starting with the nearest repair shop to the geographical location associated with the mobile device 120 and terminating with the repair shop furthest from the geographical location associated with the mobile device 120. The user of mobile device 120 may be able to sort the repair shop data comprised within the sortable list based on factors such as highest repair ranking, nearest location, repair color availability, and the like; wherein as per above, the repair color availability was determined based on the workload of each repair shop); and
cause the user computer device to display an interactive user interface including a list of the one or more determined repair facilities (Kelsh: ¶ 0142, ¶ 0182, ¶ 0188 showing the display of the list of repair shops is rendered on the user’s mobile device; also see ¶ 0139-0142 showing interactive user interface elements to find a repair shop).
US 20240037007 A1 (Zavesky) teaches a system for allocating resources to perform a task according to predicted capacity, and teaches based on receiving a task request (Zavesky: ¶ 0019 “resource management equipment 150 receiving task request 105”), executing a plurality of machine learning models (Zavesky: ¶ 0045-0046 showing capacity prediction component can include a plurality of machine learning models, including an ensemble model) trained on a plurality of historical production data for a plurality of resources (Zavesky: ¶ 0047 “initial and subsequent training of ANN 575 can be based on collected production data stored in historical data store 525”; also see ¶ 0039, ¶ 0048, ¶ 0052 past activity or past results), in order to predict current service capacity of a plurality of worker resources available to perform the requested task (Zavesky: ¶ 0028-0029, ¶ 0040, ¶ 0051-0054 showing predicting capacities of worker resources to perform tasks; see ¶ 0047 showing the capacity prediction may be carried out by capacity prediction component as per above). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have included the processes for use of machine learning models for predicting capacities associated with resources to be allocated to received tasks of Zavesky in the repair shop identification system of Kelsh with a reasonable expectation of success of arriving at the claimed invention, with the motivation to “efficiently predict capacities for performing tasks by resources” (Zavesky: ¶ 0017).
US 20200365034 A1 (Barth) teaches re-ranking service providers based on updated performance data, including current capacity and/or likelihood that each service provider will quickly accept a service request (Barth: ¶ 0048 showing “Service provider data can be updated (422). The service provider data can be updated based on the obtained performance data. In this way, the performance of the service provider can be used to refine the generation of eligible service providers and/or the ranking of service providers as described herein. This re-ranking can be automatically performed in real-time or near real-time, thereby allowing for service providers to be targeted for service requests based on the current capacity of the service provider and/or the likelihood that the service provider will quickly accept a service request. The capacity of the service provider can also be updated based on the completion of service requests. Timing data associated with the performance of the service request can also be used to predict the future capacity of the service provider based on the particular services required for a service request”; note that as per ¶ 0041-0042 showing the ranking/re-ranking can occur responsive to a service request from a vehicle user, and see ¶ 0053, ¶ 0041, ¶ 0043 showing historical performance, historical service times, availability, etc. pertaining to the service providers).
However, the previous combination of Kelsh, Zavesky, and Barth do not teach the currently recited limitations of claim 1, to: train a plurality of machine-learning models to correlate historical service data to a respective probability that any of the plurality of repair facilities is over-capacity, the historical service data including a plurality of variables including at least geographic location of the respective repair facility, vehicle damage attributes, and repair services attributes, wherein the training includes: iteratively testing the plurality of machine-learning models to identify a subset of the plurality of variables of the historical service data having a highest correlation with respective known capacity statuses of the plurality of repair facilities; and retraining the plurality of machine-learning models to execute with the subset of variables. Furthermore, they do not explicitly teach adjusting initial workload ranking based on the probability that each of the repair facilities is over-capacity in order to determine repair facilities most likely to have availability to provide the service.
US 20240320766 A1 (Ernst) teaches downranking or removing a repair service provider from a list based upon determining that the service provider is currently unavailable (Ernst: ¶ 0135 “filter the service providers to remove and/or down-weight service providers that are predicted to be unavailable for requests”), which may be based on when a decline rate is above a threshold, i.e. the availability/capacity of the provider is below a threshold amount (Ernst: ¶ 0156 “a service provider with a decline rate over a threshold, such as 50%, 75%, etc. for a particular type of request, can be updated to remove that type of request from the service providers capabilities (e.g., when identifying service providers and/or filtering out service providers at 702)”, also see ¶ 0052).
US 20220207931 A1 (Hinduja) teaches updating the plurality of machine-learning models with the plurality of current service data (Hinduja: ¶ 0166-0167, ¶ 0185-0186 showing new ground truth data collected from the service facilities, which is used to update the trained prediction model; note as above, ¶ 0041 teaches the prediction model may include an ensemble model, i.e. combination of plurality of machine learning models). US 20220164895 A1 to Beckwith et al. (Beckwith) predicts insurance claims reimbursements based upon coverage, expected repairs, and a selected facility for performing the repair (Beckwith: ¶ 0067-0070).
US 20250036944 A1 to Heinzemann et al. (Heinzemann) is the closest prior art not previously cited and teaches optimizing the training of a machine learning model by removing or modifying data from a machine learning training set to obtain a reduced training data set that maintains a threshold level of model performance (Heinzemann: ¶ 0036-0037, ¶ 0043, ¶ 0056-0064, 0067). However, even if Heinzemann teaches conceptually, identifying a reduced set of variables which do not degrade the model performance, it still would not teach the training of the plurality machine learning models on all of the specific data recited in claim 1 and iteratively testing the plurality of machine learning models to identify a specific subset of variables that have a highest correlation with known capacity statuses of repair facilities as claimed.
Therefore, the other closest prior art cited above would not have rendered the limitations above obvious to one of ordinary skill in the art, considered as a whole.
The other previously cited prior art including US 11,238,469 B1 (Talvola), US 20220180293 A1 (Cahalin), US 20210125252 A1 (Biggs), US 20190188743 A1 (Phillips), US 20120317087 A1 (Lymberopoulos), and US 20190087744 A1 (Schiemenz) do not cure the deficiencies above.
Claims 18 and 20 recite similar limitations and are novel/nonobvious for substantially the same reasons as claim 1. Claims 2-8, 11-12, 14-17, 19, and 22-23 depend from claims 1 and 18.
Therefore, the prior art does not teach claims 1-8, 11-12, 14-20, and 22-23.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Hunter Molnar whose telephone number is (571)272-8271. The examiner can normally be reached Monday - Friday, 7:30 - 4:00 EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jeffrey Zimmerman can be reached at (571) 272-4602. 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.
/HUNTER MOLNAR/Examiner, Art Unit 3628
Prosecution Insights