DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Notice to Applicant
A Notice of Abandonment was prematurely issued on 02/11/2026. A Petition to withdraw the abandonment was submitted on 02/17/2026 with a decision to grant the withdrawal on 06/17/2026.
Status of the Claims
The Amendment filed on 10/30/2025 has been entered. Claims 1-3 and 5-21 are pending in the instant patent application. Claims 1-2, 5-12, 14-18 and 20 are amended. Claim 4 is canceled. This Final Office Action is in response to the claims filed.
Response to Claim Amendments
Applicant’s amendments to the claims are insufficient to overcome the 35 U.S.C. §101 rejections. The rejections remain pending and are updated and addressed below in light of the amendments and per guidelines for 101 analysis (PEG 2019).
Applicant’s amendments to the claims are insufficient to overcome the 35 U.S.C. §103 rejections. The rejections remain pending and are updated and addressed below in light of the amendments and newly cited art.
Response to 35 U.S.C. §101 Arguments
Applicant’s arguments regarding 35 U.S.C. §101 rejection of the claims have been fully considered, but are not persuasive.
Regarding Applicant’s arguments that the claims as currently amended do not recite abstract ideas, Examiner respectfully disagrees. Applicant asserts that the claims are similar in patent eligibility to Enfish, McRO and CardioNet, Examiner respectfully disagrees. In McRO, "As part of its analysis, the McRO court examined the specification, which described the claimed invention as improving computer animation through the use of specific rules, rather than human artists, to set morph weights (relating to facial expressions as an animated character speaks) and transition parameters between phonemes (relating to sounds made when speaking). As explained in the specification, human artists did not use the claimed rules, and instead relied on subjective determinations to set the morph weights and manipulate the animated face to match pronounced phonemes. The McRO court thus relied on the specification's explanation of how the claimed rules enabled the automation of specific animation tasks that previously could not be automated when determining that the claims were directed to improvements in computer animation instead of an abstract idea. The McRO court indicated that it was the incorporation of the particular claimed rules in computer animation that "improved [the] existing technological process", unlike cases such as Alice and the present claims, where a computer was merely used as a tool to perform an existing process." (Please see USPTO November 2, 2016 Memo, Recent Subject Matter Eligibility Decisions). Therefore, unlike McRo claims and similar to Alice, the current claims themselves are not a set of rules that enable the automation of the specific tasks in a computer in order to either improve an existing technology. In Enfish, the Court found the claims to be 101 eligible because the claims claimed a specific asserted improvement in computer capabilities (i.e, the self-referential table for computer database). To make their determination of whether the claims are directed to an improvement in existing computer technology, the court looked to the teachings of the specification. The court identified the specification's teachings that the claimed invention achieves other benefits over conventional databases, such as increased flexibility, faster search times, and smaller memory requirements. On the other hand, here, the current claims are not directed to improving a computer or technological process. The computing devices in this case, do not improve the functioning of the computer itself or another technical field. These claimed features are generic components of the computer itself. The claims do not claim anything specific or that differentiates the limitations claimed from limitation of a generic computer. Therefore, any improvements or increased performance/efficiency claimed by the Applicant is an inherent quality of the linking of the abstract idea to these generic computer components. Regarding CardioNet, there was a specific and clear improvement to the technology or cardiac monitoring. Examiner asserts that there is not an improvement to the technology within the amended claim language. Regarding Bascom, the current claims are distinguishable from the Bascom ruling. In Bascom, the claims solved an internet centric problem of filtering internet content. The internet as computer technology as well as the computers are a necessary component of the claims. In other words, it is impossible to perform the claims without the use of the internet because no analog version of the problem existed prior to the internet. Therefore, Bascom is distinguishable from the current claims because current claims are merely linking the combination of the conventional and routine computer components to the abstract idea in order to solve a business problem, while in Bascom, the abstract idea is necessarily rooted in the combination of the conventional and routine computer components. Regarding DDR, in DDR Holdings, the court found the claims to be patent eligible because the claims recites the solution of a hybrid webpage that co-displays the look and feel of the first website with the desired content from the second website. The court found such solution to be rooted in computer technology because there was no other way to accomplish such solution. The computer was an essential part of performing the solution. Although the current claims recite functions performed by a computer system and machine learning model, such functions are not found to be significantly more when recited in their generic manner
Examiner further finds that the limitations presented in the claim language are not indicative of integrating the abstract idea into a practical application. For example, the use of machine learning is merely being used as a tool to carry out the abstract idea with no improvement to the machine learning aspects or to the technology/technical field.
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.
Regarding Claims 1-3 and 5-11, they are directed to a system, however the claims are directed to a judicial exception without significantly more. Claims 1-3 and 5-11 are directed to the abstract idea of integrating sustainability solutions for an enterprise.
Performing the Step 2A Prong 1 analysis while referring specifically to independent Claim 1, claim 1 recites accesses and parses business model data and optimization configuration files of an enterprise system to identify and extract one or more objective functions of the enterprise system, wherein the one or more objective functions define first defining relationships between one or more processes of the enterprise system and one or more business objectives of the enterprise system; generates one or more sustainability cost functions corresponding to the one or more processes using one or more first models that estimate sustainability costs based on the one or more processes, wherein the one or more sustainability cost functions define second relationships between the one or more processes and the sustainability costs, and wherein the sustainability costs comprise quantities of gas emissions; combines the one or more objective functions with the one or more sustainability cost functions to generate a multi- objective optimization function for the enterprise system, wherein the multi-objective optimization function comprises a first objective corresponding to maximizing the one or more business objective and a second objective corresponding to minimizing the one or more sustainability costs; applies a solver to the multi- objective optimization function to determine that formulates potential changes to the one or more processes that balance the first objective and the second objective.
These claim limitations fall within the Mental Processes grouping of abstract ideas for they are concepts that can be practically performed in the human mind and/or with pen/paper (including an observation, evaluation, judgment, opinion). Furthermore, the courts have found claims requiring a generic computer or nominally reciting a generic computer may still recite a mental process even though the claim limitations are not performed entirely in the human mind (see MPEP 2106.04(a)(2)(III)(C)). In addition, the claim limitations fall within the Certain Methods of Organizing Human Activity due to the business relations taking place and Mathematical Concepts due to the mathematical relationships taking place.
Furthermore, the recitation of one or more first trained machine learning processes does not take the claim out of the abstract idea groupings.
Accordingly, the claim recites an abstract idea and dependent claim 3 further recite the abstract idea.
Regarding Step 2A Prong 2 analysis, the judicial exception is not integrated into a practical application. In particular the claim recites the elements of a memory, a processor, an extraction component, a sustainability analysis component, an optimization formulation component, solver application component, optimization solver, and one or more first trained machine learning processes. The memory, a processor, an extraction component, optimization solver, a sustainability analysis component, an optimization formulation component, solver application component, and one or more first trained machine learning processes are merely generic computing devices and do not integrate the judicial exception into a practical application.
With respect to 2B, the claims do not include additional elements amounting to significantly more than the abstract idea. Claims 1-2 and 5-11 includes various elements that are not directed to the abstract idea under 2A. These elements include a memory, a processor, an extraction component, a sustainability analysis component, an optimization formulation component, optimization solver, solver application component, and one or more first trained machine learning processes, a training component, training data generation component, solver selection component, one or more second machine learning processes, a recommendation component, and the generic computing elements described in the Applicant's specification in at least Para 0106-0126. These elements do not amount to more than the abstract idea because it is a generic computer performing generic functions.
Therefore, Claims 1-2 and 5-11, alone or in combination, are not drawn to eligible subject matter as they are directed to abstract ideas without significantly more.
Regarding Claims 12-19, they are directed to a method, however the claims are directed to a judicial exception without significantly more. Claims 12-19 are directed to the abstract idea of integrating sustainability solutions for an enterprise.
Performing the Step 2A Prong 1 analysis while referring specifically to independent Claim 12, claim 12 recites extracting, one or more objective functions of an enterprise system from defined business model data and optimization configuration files for the enterprise system, wherein the one or more objective functions define first relationships between one or more processes of the enterprise system and one or more business objectives of the enterprise system; determining, one or more sustainability costs related to the one or more processes; and generating, one or more sustainability cost functions corresponding to the one or more processes, wherein the one or more sustainability cost functions define second relationships between the one or more existing processes and the sustainability costs, and wherein the sustainability costs comprise quantities of gas emissions; combining, the one or more objective functions with the one or more sustainability cost functions to generate a multi-objective optimization function for the enterprise system, wherein the multi-objective optimization function comprises a first objective corresponding to maximizing the one or more business objectives and a second objective corresponding to minimizing the one or more sustainability costs; and applying, a solver to the multi-objective optimization function to determine potential changes to the one or more processes that balance the first objective and the second objective.
These claim limitations fall within the Mental Processes grouping of abstract ideas for they are concepts that can be practically performed in the human mind and/or with pen/paper (including an observation, evaluation, judgment, opinion). Furthermore, the courts have found claims requiring a generic computer or nominally reciting a generic computer may still recite a mental process even though the claim limitations are not performed entirely in the human mind (see MPEP 2106.04(a)(2)(III)(C)). In addition, the claim limitations fall within the Certain Methods of Organizing Human Activity due to the business relations taking place and Mathematical Concepts due to the mathematical relationships taking place.
Furthermore, the recitation of one or more first machine learning processes does not take the claim out of the abstract idea groupings.
Accordingly, the claim recites an abstract idea and dependent claims 14-15 further recite the abstract idea.
Regarding Step 2A Prong 2 analysis, the judicial exception is not integrated into a practical application. In particular the claim recites the elements of a system, a processor, optimization solver and one or more first machine learning processes. The system, optimization solver, processor and one or more first machine learning processes are merely generic computing devices and do not integrate the judicial exception into a practical application.
With respect to 2B, the claims do not include additional elements amounting to significantly more than the abstract idea. Claims 12-13 and 16-19 includes various elements that are not directed to the abstract idea under 2A. These elements include a system, a processor, optimization solver, one or more first machine learning processes, one or more second machine learning processes, and the generic computing elements described in the Applicant's specification in at least Para 0106-0126. These elements do not amount to more than the abstract idea because it is a generic computer performing generic functions.
Therefore, Claims 12-13 and 16-19, alone or in combination, are not drawn to eligible subject matter as they are directed to abstract ideas without significantly more.
Regarding Claim 20, it is directed to a computer program product, however the claims are directed to a judicial exception without significantly more. Claim 20 is directed to the abstract idea of integrating sustainability solutions for an enterprise.
Performing the Step 2A Prong 1 analysis while referring specifically to independent Claim 20, claim 20 recites extract one or more objective functions of an enterprise system from defined business model data and optimization configuration files for the enterprise system, wherein the one or more objective functions define first relationships between one or more processes of the enterprise system and one or more business objectives of the enterprise system; determine one or more sustainability costs related to the one or more processes; and generate one or more sustainability cost functions corresponding to the one or more processes, wherein the one or more sustainability cost functions define second relationships between the one or more existing processes and the sustainability costs, and wherein the sustainability costs comprise quantities of gas emissions; combine the one or more objective functions with the one or more sustainability cost functions to generate a multi-objective optimization function for the enterprise system, wherein the multi-objective optimization function comprises a first objective corresponding to maximizing the one or more business objectives and a second objective corresponding to minimizing the one or more sustainability costs; and apply a solver to the multi-objective optimization function to determine potential changes to the one or more processes that balance the first objective and the second objective.
These claim limitations fall within the Mental Processes grouping of abstract ideas for they are concepts that can be practically performed in the human mind and/or with pen/paper (including an observation, evaluation, judgment, opinion). Furthermore, the courts have found claims requiring a generic computer or nominally reciting a generic computer may still recite a mental process even though the claim limitations are not performed entirely in the human mind (see MPEP 2106.04(a)(2)(III)(C)). In addition, the claim limitations fall within the Certain Methods of Organizing Human Activity due to the business relations taking place and Mathematical Concepts due to the mathematical relationships taking place.
Accordingly, the claim recites an abstract idea and claim 21 further recites the abstract idea.
Regarding Step 2A Prong 2 analysis, the judicial exception is not integrated into a practical application. In particular the claim recites the elements of a computer readable storage medium, processor, optimization solver and one or more first machine learning processes. The computer readable storage medium, processor, optimization solver and one or more first machine learning processes are merely generic computing devices and do not integrate the judicial exception into a practical application.
With respect to 2B, the claims do not include additional elements amounting to significantly more than the abstract idea. Claims 20-21 include various elements that are not directed to the abstract idea under 2A. These elements include a computer readable storage medium, optimization solver, processor, one or more first machine learning processes, and the generic computing elements described in the Applicant's specification in at least Para 0106-0126. These elements do not amount to more than the abstract idea because it is a generic computer performing generic functions.
Therefore, Claims 20-21 are not drawn to eligible subject matter as they are directed to abstract ideas without significantly more.
Response to 35 U.S.C. §101 Arguments
Applicant’s amendments to the claims are insufficient to overcome the 35 U.S.C. §103 rejections. The rejections remain pending and are updated and addressed below in light of the amendments and newly cited art. Furthermore, Applicant’s arguments are moot in light of newly amended language.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 12 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) .
Regarding Claim 1, Whitehair teaches the limitations of Claim 1 which state
a memory that stores computer executable components; and a processor that executes the computer executable components stored in the memory (Pancholi: Para 0138);
a sustainability analysis component that generates one or more sustainability cost functions corresponding to the one or more processes (Pancholi: Para 0207 via a carbon emissions cost function) using one or more first trained machine learning models that estimate sustainability costs based on the one or more processes (Pancholi: Para 0247, 0256, 0259 via a vector objective mapping a simulation/process configuration to objective or performance values such as carbon emissions Para 0248, 0250 via generating an approximate objective model from training data using neural network or other learned models), wherein the one or more sustainability cost functions define second relationships between the one or more processes and the sustainability costs (Pancholi: Para 0208, 0209, 0223 via mapping resource consumption decision variables to units and totals of carbon emissions…translation factors relate resource consumption to corresponding emissions…models relate energy/process decision variables to carbon emission objectives and determine the optimized variable values), and wherein the sustainability costs comprise quantities of gas emissions (Pancholi: Para 0210 via carbon emissions are a sustainability metric);
an optimization formulation component that combines the one or more objective functions with the one or more sustainability cost functions to generate a multi- objective optimization function for the enterprise system (Pancholi: Para 0219, 0224, 0236-0237, 0240 via combining a monetary cost function and carbon emissions cost function into one predictive function with weights…user weights are assigned to objectives in a multi-objective cost function…use of multi-objective/pareto optimization).
However, Pancholi does not explicitly disclose the limitation of Claim 1 which state an extraction component that accesses and parses business model data and optimization configuration files of an enterprise system to identify and extract one or more objective functions of the enterprise system, wherein the one or more objective functions define first relationships between one or more processes of the enterprise system and one or more business objectives of the enterprise system.
Whitehair though, with the teachings of Pancholi, teaches of
an extraction component that accesses and parses business model data and optimization configuration files of an enterprise system to identify and extract one or more objective functions of the enterprise system, wherein the one or more objective functions define first relationships between one or more processes of the enterprise system and one or more business objectives of the enterprise system (Whitehair: Para 0038, 0152 via enterprise system/process model containing resources, constraints and objectives translated into a mathematical representation having an objective function; Para 0130, 0309-0315 via parsing enterprise model data structures; Para 0190, 0195 via business process model used to extract relationships and generate equations; Para 0430 via the enterprise model is traversed and converted into a solver matrix; Para 0126, 0207 via solver input syntax is loaded into the solver and enterprise configurations parameters are modeled).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi with the teachings of Whitehair in order to have an extraction component that accesses and parses business model data and optimization configuration files of an enterprise system to identify and extract one or more objective functions of the enterprise system, wherein the one or more objective functions define first relationships between one or more processes of the enterprise system and one or more business objectives of the enterprise system. The motivations behind this being to incorporate the teachings of modeling and/or optimizing systems and/or processes over large enterprises as taught by Whitehead. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
The combination of Pancholi/Whitehair further teaches the limitations of Claim 1 which states
wherein the multi-objective optimization function comprises a first objective corresponding to maximizing the one or more business objective (Whitehair: Para 0121, 0181 via enterprise objective may be maximizing overall profit) and a second objective corresponding to minimizing the one or more sustainability costs (Pancholi: Para 0102, 0130 via Asset allocator may be configured to minimize the economic cost (or maximize the economic value) of operating asset allocation system over the duration of the optimization period).
a solver application component that applies an optimization solver to the multi-objective optimization function (Whitehair: Para 0128, 0182, 0430 via using LP, NLP, MIP and MILP solvers and applying a matrix solver to the generated optimization equation matrix) to determine potential changes to the one or more processes that balance the first objective and the second objective (Whitehair: Para 0181-0182, 0207-0208 via modifying enterprise process model parameters to output an optimized configuration).
Claims 12 and 20 are substantially similar to Claim 1 and are rejected for the same reasons.
Claim(s) 2-3 and 13-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) further in view of Takawale et al. (US 2024/0070680 A1).
Regarding Claim 2, while Pancholi/Whitehead teaches the limitations of Claim 1, it does not explicitly disclose the limitations of Claim 2 which state wherein the computer executable components further comprise a training component that trains the one or more first machine learning models to classify different types of sustainability costs associated with different types of business objective functions and predict measures of influence of the different types of business objective functions on amounts of the different types of sustainability costs and wherein the sustainability analysis component employs the one or more first machine learning models to predict respective types of the one or more sustainability costs related to the one or more processes and respective measures of influence of the one or more objective functions on respective amounts of the respective types of the one or more sustainability costs.
Takawale though, with the teachings of Pancholi/Whitehead, teaches of
wherein the computer executable components further comprise a training component that trains the one or more first machine learning models to classify different types of sustainability costs associated with different types of business objective functions and predict measures of influence of the different types of business objective functions on amounts of the different types of sustainability costs and wherein the sustainability analysis component employs the one or more first machine learning models to predict respective types of the one or more sustainability costs related to the one or more processes and respective measures of influence of the one or more objective functions on respective amounts of the respective types of the one or more sustainability costs (Takawale: Para 0050-0051 via Once all data is collected, segmented, and identified (or classified), a sustainability attribute model training engine 614 may train sustainability attribute models to use the sustainability metrics of a product as input to generate (or calculate) sustainability scores. In some embodiments, supervised machine learning may be applied to train the plurality of sustainability attribute models. Since each sustainability attribute model is specific to a particular sustainability attribute, the sustainability metrics corresponding to the sustainability attribute may be used to train the model corresponding to the sustainability attribute. For example, a sustainability attribute model may be specific to the sustainability attribute of waste. Accordingly, this model can be trained by applying product sustainability metrics related to waste. After training, the model specific to waste can take a product as input and calculate a sustainability score specific to waste. In this way, when a product is input into multiple sustainability attribute models each specific to a different sustainability attribute, multiple sustainability scores can be generated for the same product. These scores may each reflect the sustainability of the input product with respect to each of the attributes corresponding to the sustainability attribute models. It is understood that in applications other than sustainability of products, such as automobiles, telecom, financial services, may be analyzed by the disclosed methods in other embodiments. In such embodiments, attributes other than those related to sustainability may be provided as the input metrics for building attribute models and may be analyzed by the attribute models to help generate a score related to a feature other than sustainability. For example, furniture may be analyzed based on attributes related to comfort to generate a score indicating the level of comfort provided by the furniture. In another example, home locations may be analyzed based on attributes related to family friendly neighborhoods to generate a score indicating how family friendly the neighborhood surrounding the home is. The output of each of the plurality of sustainability attribute models may be used as input by the product sustainability score predictor to generate sustainability scores for products. After building and training the individual sustainability attribute models, the trained sustainability attribute models may be used as components of the product sustainability score predictor. The product sustainability score predictor may generate product sustainability score based on the sustainability scores for each of the sustainability attributes related to a product. For example, in some embodiments, a product sustainability score for a product (e.g., television) may be based on the output of each of the sustainability attribute models for the product (e.g., carbon emissions, water, waste, life span, etc.). To account for a user's preferences related to sustainability, the product sustainability score predictor can be trained to weigh each of the attribute scores for the product according to user preferences. In some embodiments, the user preferences may be provided in the form of explicitly selected user preferences. In some embodiments, the user preferences may be provided additionally or alternatively in the form of historical user purchase information. For example, if a user selects life span as a metric that they prioritize, then the output from an attribute model for life span can be given more weight by the product sustainability score predictor. In some embodiments, the product sustainability score predictor may be a machine learning model. In some embodiments, the product sustainability score predictor may be trained by applying supervised learning).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehead with the teachings of Takawale, in order to have wherein the computer executable components further comprise a training component that trains the one or more first machine learning models to classify different types of sustainability costs associated with different types of business objective functions and predict measures of influence of the different types of business objective functions on amounts of the different types of sustainability costs and wherein the sustainability analysis component employs the one or more first machine learning models to predict respective types of the one or more sustainability costs related to the one or more processes and respective measures of influence of the one or more objective functions on respective amounts of the respective types of the one or more sustainability costs. The motivations behind this being to incorporate the teachings of applying machine learning to generate sustainability scores and curate product recommendations based on the sustainability scores. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Regarding Claim 3, the combination of Pancholi/Whitehead/Takawale teaches the limitations of Claim 3 which state
wherein the different types of sustainability costs comprise different types of emission classes and wherein the amounts correspond to emission amounts (Pancholi: Para 0227 via A “cost” or “cost savings” can be expressed in terms of any of the control objectives described herein or any other control objectives accounted for by the predictive cost function. For example, a “cost” or “cost savings” may refer to any of a variety of other types of cost such as resource consumption (e.g., expressed in units of energy, water, natural gas, or any other resource), carbon emissions (e.g., expressed in units of carbon), occupant comfort (e.g., expressed in units of comfort), disease transmission risk (e.g., expressed in units of risk or probability), equipment reliability (e.g., expressed in units of reliability or expected failures), or any other control objectives which may be desirable to optimize, either alone or in weighted combination with other control objectives. Similarly, a given energy source may be “less costly” or “more costly” than another energy source if the electric energy supplied from the given energy source causes a lesser impact or greater impact, respectively, on the control objective represented by the “cost” in the predictive cost function. Likewise, a “cost characteristic” of a given energy source or amount of electric energy received from a given energy source may include, for example, a monetary cost or price of the electric energy (e.g., $ per unit of energy or power), an amount of carbon emissions associated with the electric energy (e.g., carbon emissions per unit of energy or power), a marginal operating emissions rate (MOER) associated with the electric energy, carbon credit information (e.g., an amount or cost of carbon credits needed to compensate for the carbon emissions associated with the electric energy), other sustainability metrics or sustainability factors, or any other attribute or characteristic of the electric energy which may be relevant to any of the control objectives associated with the cost function J).
Regarding Claim 13, while the combination of Pancholi/Whitehead teaches the limitations of Claim 12, it does not explicitly disclose the limitations of Claim 13 which state wherein using the one or more first machine learning processes comprises: training, by the system, one or more sustainability models to classify different types of sustainability costs associated with different types of business objective functions and predict measures of influence of the different types of business objective functions on amounts of the different types of sustainability costs; and employing, by the system, the one or more sustainability models to predict respective types of the one or more sustainability costs related to the one or more processes and respective measures of influence of the one or more objective functions on respective amounts of the respective types of the one or more sustainability costs.
Takawale though, with the teachings of Pancholi/Whitehead, teaches of
wherein using the one or more first machine learning processes comprises: training, by the system, one or more sustainability models to classify different types of sustainability costs associated with different types of business objective functions and predict measures of influence of the different types of business objective functions on amounts of the different types of sustainability costs; and employing, by the system, the one or more sustainability models to predict respective types of the one or more sustainability costs related to the one or more processes and respective measures of influence of the one or more objective functions on respective amounts of the respective types of the one or more sustainability costs (Takawale: Para 0050-0051 via Once all data is collected, segmented, and identified (or classified), a sustainability attribute model training engine 614 may train sustainability attribute models to use the sustainability metrics of a product as input to generate (or calculate) sustainability scores. In some embodiments, supervised machine learning may be applied to train the plurality of sustainability attribute models. Since each sustainability attribute model is specific to a particular sustainability attribute, the sustainability metrics corresponding to the sustainability attribute may be used to train the model corresponding to the sustainability attribute. For example, a sustainability attribute model may be specific to the sustainability attribute of waste. Accordingly, this model can be trained by applying product sustainability metrics related to waste. After training, the model specific to waste can take a product as input and calculate a sustainability score specific to waste. In this way, when a product is input into multiple sustainability attribute models each specific to a different sustainability attribute, multiple sustainability scores can be generated for the same product. These scores may each reflect the sustainability of the input product with respect to each of the attributes corresponding to the sustainability attribute models. It is understood that in applications other than sustainability of products, such as automobiles, telecom, financial services, may be analyzed by the disclosed methods in other embodiments. In such embodiments, attributes other than those related to sustainability may be provided as the input metrics for building attribute models and may be analyzed by the attribute models to help generate a score related to a feature other than sustainability. For example, furniture may be analyzed based on attributes related to comfort to generate a score indicating the level of comfort provided by the furniture. In another example, home locations may be analyzed based on attributes related to family friendly neighborhoods to generate a score indicating how family friendly the neighborhood surrounding the home is. The output of each of the plurality of sustainability attribute models may be used as input by the product sustainability score predictor to generate sustainability scores for products. After building and training the individual sustainability attribute models, the trained sustainability attribute models may be used as components of the product sustainability score predictor. The product sustainability score predictor may generate product sustainability score based on the sustainability scores for each of the sustainability attributes related to a product. For example, in some embodiments, a product sustainability score for a product (e.g., television) may be based on the output of each of the sustainability attribute models for the product (e.g., carbon emissions, water, waste, life span, etc.). To account for a user's preferences related to sustainability, the product sustainability score predictor can be trained to weigh each of the attribute scores for the product according to user preferences. In some embodiments, the user preferences may be provided in the form of explicitly selected user preferences. In some embodiments, the user preferences may be provided additionally or alternatively in the form of historical user purchase information. For example, if a user selects life span as a metric that they prioritize, then the output from an attribute model for life span can be given more weight by the product sustainability score predictor. In some embodiments, the product sustainability score predictor may be a machine learning model. In some embodiments, the product sustainability score predictor may be trained by applying supervised learning).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehead with the teachings of Takawale, in order to have wherein using the one or more first machine learning processes comprises: training, by the system, one or more sustainability models to classify different types of sustainability costs associated with different types of business objective functions and predict measures of influence of the different types of business objective functions on amounts of the different types of sustainability costs; and employing, by the system, the one or more sustainability models to predict respective types of the one or more sustainability costs related to the one or more processes and respective measures of influence of the one or more objective functions on respective amounts of the respective types of the one or more sustainability costs. The motivations behind this being to incorporate the teachings of applying machine learning to generate sustainability scores and curate product recommendations based on the sustainability scores. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Regarding Claim 14, Pancholi/Whitehead/Takawale teaches the limitations of Claim 14 which state
wherein the different types of sustainability costs comprise different types of emission classes and wherein the amounts correspond to emission amounts (Pancholi: Para 0227 via A “cost” or “cost savings” can be expressed in terms of any of the control objectives described herein or any other control objectives accounted for by the predictive cost function. For example, a “cost” or “cost savings” may refer to any of a variety of other types of cost such as resource consumption (e.g., expressed in units of energy, water, natural gas, or any other resource), carbon emissions (e.g., expressed in units of carbon), occupant comfort (e.g., expressed in units of comfort), disease transmission risk (e.g., expressed in units of risk or probability), equipment reliability (e.g., expressed in units of reliability or expected failures), or any other control objectives which may be desirable to optimize, either alone or in weighted combination with other control objectives. Similarly, a given energy source may be “less costly” or “more costly” than another energy source if the electric energy supplied from the given energy source causes a lesser impact or greater impact, respectively, on the control objective represented by the “cost” in the predictive cost function. Likewise, a “cost characteristic” of a given energy source or amount of electric energy received from a given energy source may include, for example, a monetary cost or price of the electric energy (e.g., $ per unit of energy or power), an amount of carbon emissions associated with the electric energy (e.g., carbon emissions per unit of energy or power), a marginal operating emissions rate (MOER) associated with the electric energy, carbon credit information (e.g., an amount or cost of carbon credits needed to compensate for the carbon emissions associated with the electric energy), other sustainability metrics or sustainability factors, or any other attribute or characteristic of the electric energy which may be relevant to any of the control objectives associated with the cost function J).
Regarding Claim 15, Pancholi/Whitehead/Takawale teaches the limitations of Claim 15 which state
wherein the generating comprises: modeling the one or more objective functions as one or more cost functions that formulate the one or more processes as a function of financial costs attributed to the one or more processes; and adapting the one or more cost functions to reformulate the one or more processes as a function of the one or more sustainability costs based on the financial costs, the respective types of the one or more sustainability costs and the respective measures of influence (Pancholi: Para 0210, 0219-0220, 0237 via Although carbon emissions is provided as one example of a sustainability metric that can be accounted for in the cost function J, it is contemplated that the carbon emissions term can be replaced with any other sustainability control objective or sustainability metric (e.g., water usage, global warming potential, non-carbon pollution, etc.), or other sustainability control objectives or sustainability metrics can be added as additional terms in the cost function J in addition to the carbon emissions term. In sustainability cost functions that account for other sustainability metrics, the β variables can include coefficients that translate the values of the decision variables into units of the corresponding sustainability control objective (e.g., amount of water consumption per unit of resource consumption or resource production, amount of global warming potential per unit of resource consumption or resource production, amount of non-carbon pollution per unit of resource consumption or resource production)… it is contemplated that the cost function J may include multiple terms that account for multiple different control objectives within a single cost function. For example, the monetary cost function and carbon emissions cost function shown above can be combined to generate a single predictive cost function J that accounts for both monetary cost and carbon emissions… weights that are used to assign the relative importance of the monetary cost defined by the first term and the carbon emissions cost defined by the second term in the overall cost function J and the remaining variables are the same as described with reference to the monetary cost function and carbon emissions cost function above. It is contemplated that any of the cost functions J described throughout the present disclosure or any of the disclosures incorporated by reference herein can be combined (e.g., by adding them together in a weighted summation and/or subtracting one or more cost functions from one or more other cost functions) to account for any combination of control objectives within a single cost function. In addition to assigning relative importance to various control objectives, the weights w.sub.1 and w.sub.2 may function as unit conversion factors (e.g., cost per dollar, cost per unit of carbon emissions, etc.) to translate different units associated with different control objectives into a common “cost” unit that is optimized when performing the optimization process… planning tool 600 is configured to perform an optimization process using the cost function J to drive the cost defined by the cost function J toward an optimal value (e.g., a minimum or maximum value) subject to a set of constraints. The set of constraints may include equations and/or inequalities that define relationships between variables used in the optimization process. Some of the variables that appear in the set of constraints may be provided as inputs to the optimization process and may be maintained at fixed values when performing the optimization process. Other variables that appear in the set of constraints may have time-varying values and thus may be set to different predetermined values at different time steps k. For example, the time-varying predicted loads custom-character.sub.k to be served at each time step k may be determined prior to performing the optimization process and may have different values at different time steps k. The values of such variables may be set to the predetermined values for each time step k during the optimization process).
Claim(s) 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) in view of Takawale et al. (US 2024/0070680 A1) further in view of Menon et al. (US 2024/0403893 A1).
Regarding Claim 5, while Pancholi/Whitehair/Takawale teaches the limitations of Claim 2, it does not explicitly disclose the limitations of Claim 5 which state wherein the computer executable components further comprise a training data generation component that generates a training dataset that maps the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions on the amounts of the different types of sustainability costs, and wherein the training comprises training the one or more sustainability models using the training dataset.
Menon though, with the teachings of Pancholi/Whitehair/Takawale, teaches of
wherein the computer executable components further comprise a training data generation component that generates a training dataset that maps the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions on the amounts of the different types of sustainability costs, and wherein the training comprises training the one or more sustainability models using the training dataset (Menon: Para 0214-0215 via the sustainability platform system 72 may receive a selection of one or more of the action plans 90 presented at block 362. When a user selects a particular action plan 90 to implement, information related to the reasons in which the action plan 90 was selected can be requested and saved by the sustainability platform system 72. The collected information may include specifying which sustainability parameter was the most important to the user, any quantitative data that provided reasoning for the selection (such as plan x was chosen because the cost was 10% lower than the next best plan), general statements to explain their selection, and the like. The collected information may be provided as training data to the AI model to better predict suitable action plans 90 to recommend to the respective user in the future. With this in mind, the AI model could also be trained to learn from the action selections that the user is making prior to optimization. That is, common patterns across different action variables could be established and used to preferentially select actions for future optimizations. For example, if plans created to address a particular emission scope in a given geography often include similar subsets of actions (such as installing vapor recovery units across gas processing facilities in the southern United States), then when future plans are created in the same geography to address the same scope or problem, these previously selected actions could be recommended).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehair/Takawale with the teachings of Menon in order to have wherein the computer executable components further comprise a training data generation component that generates a training dataset that maps the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions on the amounts of the different types of sustainability costs, and wherein the training comprises training the one or more sustainability models using the training dataset. The motivations behind this being to incorporate the teachings of providing plans, workflows, and recommendations for improving sustainability parameters across enterprise operations as taught by Menon. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Claim(s) 6 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) in view of Takawale et al. (US 2024/0070680 A1) in view of Menon et al. (US 2024/0403893 A1) further in view of Mahindru et al. (US 2023/0140553 A1) .
Regarding Claim 6, while the combination of Pancholi/Whitehair/Takawale/Menon teaches the limitations of Claim 5, it does not explicitly disclose the limitations of Claim 6 which state wherein the training data generation component generates the training dataset by parsing using one or more automated information parsing processes, open-source documents and sustainability report data associated with different types of business objective functions and extracting structured information correlating the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions to the amounts of the different types of sustainability costs.
Mahindru though, with the teachings of Pancholi/Whitehair/Takawale/Menon teaches of
wherein the training data generation component generates the training dataset by parsing using one or more automated information parsing processes, open-source documents and sustainability report data associated with different types of business objective functions and extracting structured information correlating the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions to the amounts of the different types of sustainability costs (Mahindru: Para 0080-0081 via the mapping illustrated in FIG. 3 is based on detected change/trend in a measure of answer quality provided by the chatbot AI, as determined by the detection component 130 based on the measure of answer quality falling below a defined threshold using the policy schema 140. Based on this detected trend, the correlation component 132 can employ the data model 200 to identify the potential AI model KPIs that have been defined to correlate to a drop or change in the answer quality, which in this case includes “feedback per session” and “first time fix” rate. The correlation component 132 can further employ the data model 200 to identify candidate technical issues associated with these AI model KPIs, which in this case include “data coverage,” “missing data solutions,” and “feedback learning.” The remediation component 134 can further employ the data model 200 to identify candidate remediation solutions that have been previously mapped/correlated to these candidate technical issues, which in this case include improving/adjusting the “knowledge extraction” functions of the chatbot, adjusting the “AI model KB/search coverage,” and “adjusting the learning coefficient.”… the business KPIs, the AI model KPIs, the candidate technical issues, the candidate remediation solutions, and the mappings defined between these graph nodes will vary based on the enterprise system 102 and the task or tasks that the AI model 104 is adapted to perform. In accordance with one or more embodiments as applied to the chatbot AI implementation, the candidate remediation solutions can be categorized into offline actions and online actions, wherein the offline actions include actions related to offline AI model training and content coverage, and wherein online actions include actions that can be automatically executed/applied in the runtime environment of the AI model 104 and the enterprise system 102. For instance, some example offline actions can include, but are not limited to: adding missing content to the training data set and the AI model content in the AI model content knowledge base 106 (e.g., correlated to unavailability of chatbot answers/responses in the AI model content knowledge base 106), augmenting the diversity of the training data set (e.g., correlated to skewed results), augmenting the coverage for entities in the AI model content in the AI model content knowledge base 106 (e.g., correlated to low entity match scores from user utterances), and adjusting the AI model preprocessing and feature extraction protocols (e.g., including optimizing utterance analysis, entity symptom extraction, entity linking, entity-action linking, procedure extraction, and answer extraction. Some example, online actions can include, but are not limited to: adjusting one or more configurations/parameters of the AI model (e.g., weights, coefficients, parameters, activation functions, decision boundaries, decision tree hierarchy, etc.), and performing user feedback learning by generating instant feedback per chat session, and/or by generating consolidated feedback scores (e.g., over a time period, per product/sub-product, per customer, per user, per symptom, etc.)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehair/Takawale/Menon with the teachings of Mahindru in order to have wherein the generating the training dataset comprises parsing, via the training data generation component using one or more automated information parsing processes, open-source documents and sustainability report data associated with different types of business objective functions and extracting structured information correlating the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions to the amounts of the different types of sustainability costs. The motivations behind this being to incorporate the teachings of monitor and evaluate the effects of an artificial intelligence (AI) model on enterprise performance metrics. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Regarding Claim 16 while the combination of Pancholi/Whitehair/Takawale/Menon teaches the limitations of Claim 16 which states wherein using the one or more first machine learning processes further comprises: generating, by the system, a training dataset that maps the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions on the amounts of the different types of sustainability costs, wherein the training comprises training the one or more sustainability models using the training dataset (Menon: Para 0214-0215 via the sustainability platform system 72 may receive a selection of one or more of the action plans 90 presented at block 362. When a user selects a particular action plan 90 to implement, information related to the reasons in which the action plan 90 was selected can be requested and saved by the sustainability platform system 72. The collected information may include specifying which sustainability parameter was the most important to the user, any quantitative data that provided reasoning for the selection (such as plan x was chosen because the cost was 10% lower than the next best plan), general statements to explain their selection, and the like. The collected information may be provided as training data to the AI model to better predict suitable action plans 90 to recommend to the respective user in the future… With this in mind, the AI model could also be trained to learn from the action selections that the user is making prior to optimization. That is, common patterns across different action variables could be established and used to preferentially select actions for future optimizations. For example, if plans created to address a particular emission scope in a given geography often include similar subsets of actions (such as installing vapor recovery units across gas processing facilities in the southern United States), then when future plans are created in the same geography to address the same scope or problem, these previously selected actions could be recommended).
It does not explicitly disclose the limitation of Claim 16 which states wherein the generating the training dataset comprises: parsing, by the system using one or more automated information parsing processes, open-source documents and sustainability report data associated with different types of business objective functions and extracting structured information correlating the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions to the amounts of the different types of sustainability costs.
Mahindru though, with the teachings of Menon/Pancholi, teaches of
wherein the generating the training dataset comprises: parsing, by the system using one or more automated information parsing processes, open-source documents and sustainability report data associated with different types of business objective functions and extracting structured information correlating the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions to the amounts of the different types of sustainability costs (Mahindru: Para 0080-0081 via via the mapping illustrated in FIG. 3 is based on detected change/trend in a measure of answer quality provided by the chatbot AI, as determined by the detection component 130 based on the measure of answer quality falling below a defined threshold using the policy schema 140. Based on this detected trend, the correlation component 132 can employ the data model 200 to identify the potential AI model KPIs that have been defined to correlate to a drop or change in the answer quality, which in this case includes “feedback per session” and “first time fix” rate. The correlation component 132 can further employ the data model 200 to identify candidate technical issues associated with these AI model KPIs, which in this case include “data coverage,” “missing data solutions,” and “feedback learning.” The remediation component 134 can further employ the data model 200 to identify candidate remediation solutions that have been previously mapped/correlated to these candidate technical issues, which in this case include improving/adjusting the “knowledge extraction” functions of the chatbot, adjusting the “AI model KB/search coverage,” and “adjusting the learning coefficient.”… the business KPIs, the AI model KPIs, the candidate technical issues, the candidate remediation solutions, and the mappings defined between these graph nodes will vary based on the enterprise system 102 and the task or tasks that the AI model 104 is adapted to perform. In accordance with one or more embodiments as applied to the chatbot AI implementation, the candidate remediation solutions can be categorized into offline actions and online actions, wherein the offline actions include actions related to offline AI model training and content coverage, and wherein online actions include actions that can be automatically executed/applied in the runtime environment of the AI model 104 and the enterprise system 102. For instance, some example offline actions can include, but are not limited to: adding missing content to the training data set and the AI model content in the AI model content knowledge base 106 (e.g., correlated to unavailability of chatbot answers/responses in the AI model content knowledge base 106), augmenting the diversity of the training data set (e.g., correlated to skewed results), augmenting the coverage for entities in the AI model content in the AI model content knowledge base 106 (e.g., correlated to low entity match scores from user utterances), and adjusting the AI model preprocessing and feature extraction protocols (e.g., including optimizing utterance analysis, entity symptom extraction, entity linking, entity-action linking, procedure extraction, and answer extraction. Some example, online actions can include, but are not limited to: adjusting one or more configurations/parameters of the AI model (e.g., weights, coefficients, parameters, activation functions, decision boundaries, decision tree hierarchy, etc.), and performing user feedback learning by generating instant feedback per chat session, and/or by generating consolidated feedback scores (e.g., over a time period, per product/sub-product, per customer, per user, per symptom, etc.)).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Menon/Pancholi with the teachings of Mahindru in order to have wherein the generating the training dataset comprises: parsing, by the system using one or more automated information parsing processes, open-source documents and sustainability report data associated with different types of business objective functions and extracting structured information correlating the different types of sustainability costs associated with the different types of business objective functions and the measures of influence of the different types of business objective functions to the amounts of the different types of sustainability costs. The motivations behind this being to incorporate the teachings of monitor and evaluate the effects of an artificial intelligence (AI) model on enterprise performance metrics. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Claim(s) 7, 8, 10, 11, 17-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) in view of Saxena et al. (US 2022/0180146 A1).
Regarding Claim 7, while the combination of Pancholi/Whitehair teaches the limitations of Claim 1, it does not explicitly disclose the limitations of Claim 7 which state wherein the computer executable components comprise: a solver selection component that: extracts problem characteristics of the multi-objective optimization function, and determines, using one or more second machine learning, estimated performance characteristics of different optimization problem solvers as applied to solve the multi-objective optimization function based on the problem characteristics, and selects the optimization solver from the different optimization solvers for solving the multi-objective optimization function based an analysis of the estimated performance characteristics relative to one or more defined selection criteria.
Saxena though, with the teachings of Pancholi/Whitehair teaches of
wherein the computer executable components comprise: a solver selection component that: extracts problem characteristics of the multi-objective optimization function, and determines, using one or more second machine learning, estimated performance characteristics of different optimization problem solvers as applied to solve the multi-objective optimization function based on the problem characteristics, and selects the optimization solver from the different optimization solvers for solving the multi-objective optimization function based an analysis of the estimated performance characteristics relative to one or more defined selection criteria (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehair with the teachings of Saxena in order to have wherein the computer executable components comprise: a solver selection component that: extracts problem characteristics of the multi-objective optimization function, and determines, using one or more second machine learning, estimated performance characteristics of different optimization problem solvers as applied to solve the multi-objective optimization function based on the problem characteristics, and selects the optimization solver from the different optimization solvers for solving the multi-objective optimization function based an analysis of the estimated performance characteristics relative to one or more defined selection criteria. The motivations behind this being to incorporate the teachings of performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Regarding Claim 8, while the combination of Pancholi/Whitehair teaches the limitations of Claim 1, it does not explicitly disclose the limitations of Claim 8 which state wherein the solver application component further determines resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives.
Saxena though, with the teachings of Pancholi/Whitehair teaches of
wherein the solver application component further determines resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehair with the teachings of Saxena in order to have wherein the solver application component further determines resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives. The motivations behind this being to incorporate the teachings of performing multi-objective automated machine learning. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Regarding Claim 10, Pancholi/Whitehair/Saxena teaches the limitations of Claim 10 which state
wherein the computer executable components further comprise a training component that trains the one or more second machine learning models to predict performance characteristics of the different optimization problem solvers as applied to solve different types of optimizations problems based on known problem characteristics of the different types of optimization problems, known solver characteristics of the different types of optimization solvers, and known performance characteristics the different optimization problem solvers as applied to solve different types of optimizations problems, and wherein the solver analysis component employs the one or more solver assessment models to determine the estimated performance characteristics of the multi-objective optimization function based on the known solver characteristics and the problem characteristics of the multi-objective optimization function (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines).
Regarding Claim 11, Pancholi/Whitehair/Saxena teaches the limitations of Claim 11 which state
wherein the computer executable components further comprise a training data generation component that applies the different optimization problem solvers to solve the different types of optimizations problems to generate the known performance characteristics the different optimization problem solvers; extracts, the known problem characteristics of the different types of optimization problems based on analysis of the different types of optimization problems; and extracts, the known solver characteristics of the different types of optimization solvers based on analysis of the different types of optimization solvers (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines).
Regarding Claim 17, while Pancholi/Whitehead teaches the limitations of Claim 12, it does not explicitly disclose the limitations of Claim 17 which state extracting, by the system, problem characteristics of the multi-objective optimization function; determining, by the system using one or more second machine learning processes, estimated performance characteristics of different optimization problem solvers as applied to solve the multi-objective optimization function based on the problem characteristics, and selecting, by the system, the optimization solver from the different optimization solvers for solving the multi-objective optimization function based an analysis of the estimated performance characteristics relative to one or more defined selection criteria.
Saxena though, with the teachings of Pancholi/Whitehead, teaches of
extracting, by the system, problem characteristics of the multi-objective optimization function; determining, by the system using one or more second machine learning processes, estimated performance characteristics of different optimization problem solvers as applied to solve the multi-objective optimization function based on the problem characteristics, and selecting, by the system, the optimization solver from the different optimization solvers for solving the multi-objective optimization function based an analysis of the estimated performance characteristics relative to one or more defined selection criteria (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehead with the teachings of Saxena in order to have extracting, by the system, problem characteristics of the multi-objective optimization function; determining, by the system using one or more second machine learning processes, estimated performance characteristics of different optimization problem solvers as applied to solve the multi-objective optimization function based on the problem characteristics, and selecting, by the system, the optimization solver from the different optimization solvers for solving the multi-objective optimization function based an analysis of the estimated performance characteristics relative to one or more defined selection criteria. The motivations behind this being to incorporate the teachings of performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Regarding Claim 18, Pancholi/Whitehair/Saxena teaches the limitations of Claim 18 which state
generates, by the system, resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines).
Regarding Claim 19, the combination of Pancholi/Whitehair/Saxena teaches the limitations of Claim 19 which state
wherein using the one or more second machine learning processes comprises: training, by the system, one or more solver assessment models to predict performance characteristics of the different optimization problem solvers as applied to solve different types of optimizations problems based on known problem characteristics of the different types of optimization problems, known solver characteristics of the different types of optimization solvers, and known performance characteristics the different optimization problem solvers as applied to solve different types of optimizations problems; and employing, by the system, the one or more solver assessment models to determine the estimated performance characteristics of the multi-objective optimization function based on the known solver characteristics and the problem characteristics of the multi-objective optimization function (Saxena: Para 0083-0087 via a multi-objective automated machine learning system 400 (hereon referred to as the MOAML system 400) configured to use performing multi-objective automated machine learning, and, more specifically, to identifying a plurality of machine learning pipelines as Pareto-optimal solutions to optimize a plurality of objectives… Referring to FIG. 5, a flowchart is provided illustrating a process 500 for performing multi-objective automated machine learning to optimize a plurality of objectives. Also referring to FIG. 4, in one or more embodiments, input data 502 intended to be used as input to one or more machine learning (ML) models 432 for solving problems, resolving queries presented to them, or generating predictions of particular outcomes is input to, i.e., ingested by the multi-objective joint optimization engine 504 (shown as 420 in FIG. 4). The multi-objective joint optimization engine 504 is referred to hereon as the engine 504, and is discussed in detail further in this disclosure… In addition, in some embodiments, a set of objectives 508 to be attained, sometimes referred to as objectives 508 of interest (shown as 440 in FIG. 4), are input into the engine 504 through the GUI 506. Typically, the objectives 508 will be to either minimize or maximize the respective outcomes. In some embodiments, custom objectives such as robustness and fairness measures of a result are used. Also, in some embodiments, domain-specific custom objectives may be used. Moreover, in some embodiments, one or more opaque box objectives may be used. In general, a transparent box is an objective where the functional or analytical form of the objective function f(x) is known, or in some instances, does not exist. For example, if a function is defined requiring a 2-dimensional input vector similar to F((x.sub.1, x.sub.2))=[x.sub.1.sup.2+sqrt(x.sub.2)], then the functional form of this function is known. However, for opaque box objectives, such functional formulations are not known. This is typical in instances of ML problems where the objective functions f(x) do not have any functional form. In such instances, the respective ML model 432 is initially trained on some training dataset, then the objective function f(x) is evaluated, e.g., and without limitation, classification accuracy, by making predictions using the ML model 432 and then determining the accuracy… user-selectable model optimization parameters 510 (shown as 450 in FIG. 4) are also input to the engine 504, where such model optimization parameters 510 may be user-selected constraints, including, without limitation, computational or processing time. Further, one or more data transformers and ML models are collected 512 (where the ML models and the transformers are labeled 432 and 434, respectively, in FIG. 4) and a ML pipeline search space 436 is built 514 through populating the ML pipeline search space 436 with the textual names of the ML models 432 and transformers 434 that can be used for creating the ML pipelines. The ML models 432 and transformers 434 are components that will be selected to define a plurality of ML pipelines (shown in FIG. 4 as 460). Specifically, in some embodiments, the collection operation 512 includes choosing a subset of transformers 434 and ML models 432 from a pre-defined collection (library) of known transformers and models that can be used for multi-objective optimization problem and the building operation 514 is configured to populate to the search space 436 using the transformers 434 and ML models 432 from this subset. In addition, ML pipeline components include a plurality of user-selected hyperparameters 515 (shown as 470 in FIG. 4) to further define the respective ML pipelines)
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) in view of Menon et al. (US 2024/0403893 A1).
Regarding Claim 9, while the combination of Pancholi/Whitehead teaches the limitations of Claim 1, it does not explicitly disclose the limitation of Laim 9 which states wherein the computer executable components further comprise: a recommendation component that selects one or more changes of the potential changes based on the impact data associated with the one or more changes satisfying a sustainability criterion and generates and provides an entity associated with the enterprise system recommendation data recommending performance of the one or more changes.
Menon though, with the teaching of Pancholi/Whitehead teaches of
wherein the computer executable components further comprise: a recommendation component that selects one or more changes of the potential changes based on the impact data associated with the one or more changes satisfying a sustainability criterion and generates and provides an entity associated with the enterprise system recommendation data recommending performance of the one or more changes (Menon: Para 0132-0133 via the sustainability platform system 72 may receive the recommendations or generated action plans from the respective engineering workflow systems 78. In some embodiments, the recommendations or action plans may include operational changes (e.g., equipment operation schedule change to operate at certain times when renewable sources of energy are available) for devices in the facility operations, the production operations, or both. In addition, the recommendations or action plans may include equipment changes that may involve replacing equipment with more efficient equipment, adding equipment that may not be previously present (e.g., carbon capture), identifying business partners to purchase carbon credits or exchange services, and the like…the sustainability platform system 72 may send commands to devices within the enterprise to implement the action plan 90. As such, IoT devices 44 may adjust respective operations of other devices to implement the recommended actions provided in the action plan 90. In some embodiments, the action plans provided by the various engineering workflow systems 78 may be evaluated by the sustainability platform system 72 to determine whether each of them can be implemented with one another. Further, the action plans may be evaluated with respect to budgetary constraints and other constraints. The sustainability platform system 72 may select a combination of the provided action plans to use to generate commands based on the combination that suits the interests and constraints of the enterprise. These decisions may be made based on an optimization algorithm perform by the sustainability platform system 72, user input received by the sustainability platform system 72, or the like).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehair with the teachings of Menon in order to have wherein the computer executable components further comprise: a recommendation component that selects one or more changes of the potential changes based on the impact data associated with the one or more changes satisfying a sustainability criterion and generates and provides an entity associated with the enterprise system recommendation data recommending performance of the one or more changes. The motivations behind this being to incorporate the teachings of providing plans, workflows, and recommendations for improving sustainability parameters across enterprise operations as taught by Menon. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Claim(s) 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Pancholi et al. (US 2022/0284519 A1) in view of Whitehair et al. (US 2013/0124265 A1) further in view of Nielsen et al. (US 2023/0067850 A1).
Regarding Claim 21, while Pancholi/Whitehead teaches the limitations of Claim 20 it does not explicitly disclose the limitation of Claim 21 which states determines resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives; selects one or more changes of the potential changes based on the impact data associated with the one or more changes satisfying a sustainability criterion; and generate and provide an entity associated with the enterprise system recommendation data recommending performance of the one or more changes.
Nielsen though, with the teachings of Pancholi/Whitehead, teaches of
determines resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives (Nielsen: Para 0012, 0021, 0025, 0028, 0034-0035 via evaluating alternative equipment/energy parameter scenarios and display their respective carbon emission and cost outcome. Thus resulting impact data for potential operational changes);
selects one or more changes of the potential changes based on the impact data associated with the one or more changes satisfying a sustainability criterion (Nielsen: Para 0012 via generating recommendations using budget, sustainability and carbon constraints…Para 0043-0047 via selecting recommendations based on metrics and target comparisons); and
generate and provide an entity associated with the enterprise system recommendation data recommending performance of the one or more changes (Nielsen: Para 0025, 0034-0035, 0036, 0043 via providing the operator with a recommended scenario and alternative outcomes. Presenting a recommended adjustment to meet the target and showing the recommendation on a user interface).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Pancholi/Whitehead with the teachings of Neilsen in order to have determines resulting impact data indicating how the potential changes impact the one or more sustainability costs and the one or more business objectives; selects one or more changes of the potential changes based on the impact data associated with the one or more changes satisfying a sustainability criterion; and generate and provide an entity associated with the enterprise system recommendation data recommending performance of the one or more changes. Furthermore, in addition to being in the same CPC class, the teachings, suggestions, and motivations in this prior art would have led one of ordinary skill to modify the prior art reference or combine prior art reference teachings to arrive at the claimed invention.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TYRONE E SINGLETARY whose telephone number is (571)272-1684. The examiner can normally be reached 9 - 5:30.
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, Beth Boswell can be reached at 571-272-6737. 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.
/T.E.S./ Examiner, Art Unit 3625
/BETH V BOSWELL/ Supervisory Patent Examiner, Art Unit 3625
Prosecution Insights