.DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
2. This a nonfinal rejection in response to claims filed 05/27/2025. Claims 1-15 have been withdraw, and claims 16-20 have been elected without traverse. Claims 16-20 are pending and are examined herein.
Election/Restrictions
3. Applicant’s election without traverse of claims 16-20 in the reply filed on 05/27/2025 is acknowledged.
Claim Objections
4. Claims 16-20 are objected to because of the following informalities:
-Claim 16 recites “the predicted degradations” in lines 7-8. For purposes of clarity, the examiner advises the applicant to use consistent terms when mentioning “predicted degradations.” In line 5 the claims read “receives...predictions about degradations of performance,” which can be tied to “predicted degradations.” The applicant can clarify the claim by either stating “receives...predicted degradations of performance” in line 5, or alternatively amending lines 7-8 to read “the predictions about degradations,” instead. Claims 17-20 are also objected by virtue of their dependency on objected claim 16.
-Claim 20 recites “the system provides output from natural language processing systems “to about” in lines 1-2. The applicant considers this a typo and advises the applicant to correct the error by removing “to” from “to about.” Appropriate correction is required.
Drawings
5. The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they include the following reference character(s) not mentioned in the description:
-Fig. 3, reference numbers “4a” indicating “ERP”, and “4b” indicating a “DB/MDM” are not found in the specification.
-Fig. 5, reference numbers “2a” indicating external data, “2b” indicating Enterprise Resource Planning (ERP), “2c” indicating Database/Master Data, and “5” indicating standard field are not found in the specification.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because:
- reference character “1” has been used to designate both “metric, accountability, loss, calcs” in Fig. 3 and “Example Training Data for ML Model” in Fig. 5.
-reference character “3” has been used to designate both “Metric Def. and Accountability Def.” in Fig. 3 and “Primary System Data” in Fig. 5.
-reference character “4” has been used to designate both “Internal Data” in Fig. 3 and “ML Model” in Fig. 5.
-reference character “6” has been used to designate both “Business Rule Manager” in Fig. 3 and “Local Dictionary” in Fig. 5.
-reference character “7” has been used to designate both “Pre-incident...” in Fig. 3 and “Manual Mappings by User” in Fig. 5.
-reference character “8” has been used to designate both “Post-incident” in Fig. 3 and “ML Model” in Fig. 5.
-reference character “9” has been used to designate both “Action” in Fig. 3 and “Metric Standard Parameters” in Fig. 5.
-reference character “10” has been used to designate both “Watch Def. and Action Def.” in Fig. 3 and “Local Dictionary” in Fig. 5.
-reference character “11” has been used to designate both “suggestive integration” in Fig. 3 and “Database field mapping” in Fig. 5.
Corrected drawing sheets in compliance with 37 CFR 1.121(d), or amendment to the specification to add the reference character(s) in the description in compliance with 37 CFR 1.121(b) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Rejections – 35 USC § 112
6. The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
7. Claim 17 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention for the following reasons: -Claim 17 recites the limitation “such mapping” in line 5. There is insufficient antecedent basis for this limitation in the claim, because there is no previous mention of “mapping.” Resultingly, the scope of the limitation “such mapping promotes preparation,” cannot be reasonable ascertained by one of ordinary skill in the art because it is unclear what the mapping refers to, and what it means for the mapping to “promote preparation of a local dictionary mapping of each data source element to a standard table field.” For purposes of compact prosecution, the limitation is interpreted to include any “preparation of a local dictionary mapping of each data source element to a standard table field.”
-Claim 17 recites the limitation “the present approach” in line 7 has insufficient antecedent basis since there is no previous mention of a “present approach.” Furthermore, one cannot ascertain whether the present approach includes all of the limitations of claim 16, or includes all present approaches in the field.
-Claim 17 recites the limitation “using data source field names and data relevance to assist in suggestive process resulting in identification of corresponding table field” is indefinite because the term “assist in suggestive process” does not provide a standard for ascertaining the requisite degree, and providing one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. For purposes of compact prosecution, the limitation is given the broadest reasonable interpretation in view of the specification of any use of data source field names and their relevance to standard table fields to determine the relevant table field.
Claim Rejections – 35 USC § 101
8. 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.
9. Claims 16-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Is the claim to a process, machine, manufacture or composition of matter?
Claim 16 is directed to a system for autonomously improving a logistics process, comprising: a computer and an application executing thereon. Therefore, the claim is directed to at least one potentially eligible subject matter category, in this case “machine” or “manufacture.” Therefore, the claims are to be further analyzed under the full 2 step process.
Step 2a Prong 1: Is the claim directed to a Judicial Exception(A Law of Nature, a Natural Phenomenon (Product of Nature), or An Abstract Idea?)
The claims under the broadest reasonable interpretation in light of the specification are analyzed herein. Representative claim 16 is marked up, isolating the abstract idea from additional elements, wherein the abstract idea is in bold and the additional elements have been italicized as follows:
-A system for autonomously improving a logistics process, comprising: a computer and an application executing thereon that: passes historical and real-time metric, accountability, rule, and action information about a first logistics process to a machine learning model, receives, based at least on the passed information, predictions about degradations of performance associated with the first logistics process from the model, receives suggested applicable actions from the model to resolve the predicted degradations, feeds the predictions and suggested actions to an enterprise resource planning (ERP) system for analysis of business rules based at least on at least the predictions and the suggested actions, and implements a change in business rules based on an instruction received from the ERP system based at least on the analysis.
When evaluating the bolded limitations of the claims under the broadest reasonable interpretation in light of the specification, it is clear that representative claim 16 recites an abstract idea within the category of “certain methods of organizing human activity” outlined in MPEP 2106.05(a)(2). More specifically, the present claims fall under the sub-grouping “commercial or legal interactions,” including agreements in the form of contracts, legal obligations, advertising, marketing or sales activities or behaviors, and business relations. The bolded limitations in the claim merely recite functional limitations for predicting degradations of performance based on historical and real-time data passed into a model. The “commercial interaction” at hand is the generation of a recommendation to resolve the predicted degradations and feed these predictions and suggested actions to an ERP system to analyze and implement changes to business rules. The Enterprise Planning Resource system which, when recited in its plain language, broadly encapsulates any system for enterprise planning, which makes it part of the abstract idea, since enterprise planning is part of “commercial or legal interactions.” The notion that the claims are merely reciting a business relation or sales activities or behaviors are further supported in the specification [0012], “Systems and methods described herein track at least one pre-defined or user-entered metric describing a logistics process based on predefined or user-entered rules and train a machine learning (ML) model to recognize anomalies in the metric’s behavior, predict degradations of the process based at least on the recognized anomalies, and suggest proactive measures to prevent actual degradations or even failures.” Predicting anomalies in metrics to improve logistics processes is merely an abstract idea under “certain methods of organizing human activity” and are therefore to be further analyzed under Prong 2.
Step 2A Prong 2: Does the claim recite additional elements that integrate the judicial exception into a practical application?
Claim 16 recites the following additional elements:
-A system for autonomously improving a logistics process
-computer and an application executing thereon.
-machine learning model
The additional elements listed above, when considered individually and in combination with the claim as a whole, no more than a recitation of the words “apply it” (or an equivalent) or mere instructions to implement an abstract idea or other exception on generic computing components as outlined in MPEP 2106.05(f). In this case, the abstract idea of “generation of a recommendation to resolve the predicted degradations and feed these predictions and suggested actions to an ERP system to analyze and implement changes to business rules” is performed on generic computing components such as computer and an application executing thereon. This is the definition of merely providing instructions to perform the abstract idea on any “computer” as software instructions. Furthermore, limiting the system to autonomously improving a logistics process is also an example of “apply it,” as it merely instructs the abstract idea functions to be done autonomously, in this case on a computer, without reciting an improvement to computers, technological environments or fields of use (see MPEP 2106.05(a)).
Furthermore, the limitation “machine learning model” is merely a general link to a particular technological environment or field of use as outlined in MPEP 2106.05(h). The claims merely limit the model to be a “machine learning” model without meaningfully limiting the use of machine learning in the claims. Since machine learning is merely brought to perform functions that are inherent to its use, for example, predicting information using historical and current data, then it is merely a general link and does not provide an improvement to computers or to the field of machine learning.
Whether considered individually or in combination, the additional elements fail to integrate the abstract idea into a practical application, therefore the claims are directed to an abstract idea.
Step 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
Claim 16 recites the following additional elements:
-A system for autonomously improving a logistics process
-computer and an application executing thereon.
-machine learning model
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, when considered separately and as an ordered combination, they do not add significantly more (also known as an “inventive concept”) to the exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of using generic computing components such as a computer and an application executing thereon, to perform the abstract idea of “generation of a recommendation to resolve the predicted degradations and feed these predictions and suggested actions to an ERP system to analyze and implement changes to business rules” amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. Furthermore, limiting the abstract idea to be performed on an automatically, using machine learning or on an application does not meaningfully limit the claim beyond generally linking the abstract idea to a particular technological environment or field of use. Accordingly, even when viewed as a whole, nothing in the claim adds significantly more (i.e. an inventive concept) to the abstract idea. Thus claim 16 is not patent eligible because the claims are directed to an abstract without significantly more.
Dependent claims 17-20 are also given the full two part analysis both individually and in combination with the claims they depend on herein:
Claim 17 further defines the abstract idea by adding the steps of “the system further predicts table headers of data sources comprising at least one of ERP system, master data, and external data via an ML model trained on a standard or user-defined dictionary of table fields and using data source meta data comprising at least one of database name, table name and field name as features to predict a standard table field corresponding to a data source schema field, such mapping promoting preparation of a local dictionary mapping of each data source element to a standard table field name, the present approach further using data source field names and data relevance to assist in suggestive process resulting in identification of corresponding table field.” These steps merely indicate data processing steps that map unprocessed data to the standard data fields used in the ERP systems. This is a further example of “commercial or legal interactions,” wherein the “commercial interaction” at hand is the generation of a recommendation to resolve the predicted degradations and feed these predictions and suggested actions to an ERP system to analyze and implement changes to business rules. Since it merely recites how the data is to be formatted, whilst still performing the same abstract idea function, it is more of the same abstract idea as claim 16. Furthermore, the additional element of ML model is reintroduced in this claim to perform prediction of table headers. This is still a general link to machine learning, since it merely uses generic machine learning to perform predictions without meaningfully limiting the use of machine learning on the claims. Therefore, even when considered individually or as a whole the additional elements do integrate the abstract idea into a practical application or provide significantly more. Therefore, claim 17 is patent ineligible.
Claim 18 further limits the abstract idea by adding the step, “wherein prior to passing the information to the model the system trains the model to recognize anomalies in the information.” This is more of the same abstract idea, since it is merely reciting data processing steps towards performing the same commercial interaction of claim 16. Even when considering these functions in combination with the previously stated additional element of machine learning it is still a general link to the field of machine learning, without a particular improvement (see MPEP 2106.05(h)). Therefore, even when additional elements are considered individually or the claim as a whole, the claims do not integrate the abstract idea into a practical application or provide significantly more. Therefore, claim 18 is patent ineligible.
Claim 19 further limits the abstract idea by adding the step of “promotes analysis of logistics processes and execution of related business rules in an autonomous manner that does not require changes to existing business processes.” Similarly, performing analysis that does not change the existing business practices is merely an abstract idea. Furthermore, the additional element of this being done in an “autonomous manner” is an example of “apply it” or mere instructions to perform the abstract idea on a computer. Therefore, whether considered individually or as a whole, the additional elements fail to integrate the abstract idea into a practical application or provide significantly more. Therefore, claim 19 is patent ineligible.
Claim 20 further limits the abstract idea by adding the step of “provides output from natural language processing (NLP) systems to about the first logistics process to the model to supplement the metric, accountability, rule, and action information.” This is more of the same abstract idea of “generation of a recommendation to resolve the predicted degradations and feed these predictions and suggested actions to an ERP system to analyze and implement changes to business rules.” It merely adds the additional element of performing the output using “Natural language processing (NLP) systems” after the abstract idea. However, this is an example of generally linking the abstract idea to the technical field of NLP without meaningfully limiting its use on the claims (i.e., does not mention how NLP is specifically used, only generically involves the use of NLP). Therefore, whether considered individually or as a whole, the additional elements fail to integrate the abstract idea into a practical application or provide significantly more. Therefore, claim 20 is patent ineligible.
Claim Rejections – 35 USC § 103
10. 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.
11. 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.
12. Claims 16, and 18-20 are rejected under 35 U.S.C. 103 as being unpatentable over Karthik Rajagopal (US 20240362576 A1) hereinafter Rajagopal, (which has the effective filing date of 04-28-2023, which is the filing date of US Provisional Application #US 63462802) in view of Adeel Najmi (US 20160217406 A1) hereinafter Najmi.
Regarding Claim 16:
Rajagopal discloses systems for identifying delay risks in carrier shipping services and identifying at least on correction. Rajagopal teaches:
- A system for autonomously improving a logistics process, (Rajagopal [0047] In some examples, the supply chain management system 100 can provide real-time supply chain visibility and end-to-end management for shipments... The supply chain management system 100 can perform an automated process as described herein, which can, among other things, expedite the flow of goods through busy routes and/or locations such as ports, significantly reduce detention and demurrage costs, improve shipment and/or supply chain visibility, improve system performance/efficiency, improve customer satisfaction, etc.)
-comprising: a computer and an application executing thereon that: (Rajagopal [0031] The supply chain management system 100 can be part of a computing device or multiple computing devices. In some examples, the supply chain management system 100 can be part of an electronic device (or devices) such as a server, a content management system, a host machine in a network such as a cloud or private network, a computer in a vehicle, a computer node, or any other suitable electronic device(s). In some examples, the supply chain management system 100 can be or can include one or more software services and/or virtual instances hosted on a datacenter or a network environment.)
-passes historical and real-time metric, accountability, rule, and action information about a first logistics process to a machine learning model, (Rajagopal [0026] The predictive tracking management system can include a machine learning (ML) model or other artificial intelligence (AI) network that is capable of predicting the route a load (e.g., a package, parcel, or freight) is going to travel from pickup to delivery and/or throughout a trip from pickup to delivery and issues that may adversely affect the transit of the load. The predictive tracking management system may monitor real-time events ... to determine potential risks of delay and providing information... load restrictions/constraints, transportation vehicle for the load (e.g., truck, trailer, etc.), appointment time, shipment priority, etc.), information about the carrier (e.g., who is transporting the load, what type of vehicle uses the carrier, etc.), the shipper of the load, the day/time in which the load is being transported, information about the source/origin location and/or the destination (e.g., city, state, country, facility, business hours, distance from source/origin to destination, etc.), historical information (e.g., shipping or carrier patterns or histories, load patterns or histories, industry patterns or histories, source and/or destination patterns or histories, historic usage of different lanes, etc.), the day/time the load was picked up by the carrier, news information, weather information, traffic information, and/or any other relevant information. [0103] The computing system can identify a modification risk of the package, and further predict downstream effects, such as the delay risk and confidence associated with the identified modification of the transit configuration [0105] Non-limiting examples of corrective modifications include identifying an alternate origin for the load based on the first carrier, identifying a different carrier for the load, identifying a different service level for the load,) The broadest reasonable interpretation of “accountability” is information about the accountable parties, in view of specification [0012], “accountabilities (supplier, carrier, business, other).” Therefore, information about the carrier, the shipper, and the source/location falls under accountability information. The BRI of “rule” is any information regarding previous business rules that define the process, such as “load restrictions/constraints,” “shipment priority,” and “industry patterns or histories” as taught above in Rajagopal. Action information refers to information about the corrective actions the effects of those actions, which is also taught by Rajagopal in at least [0103] and [0105].
-receives, based at least on the passed information, predictions about degradations of performance associated with the first logistics process from the model, (Rajagopal [0044] estimated shipment performance, [0049] provide deeper insights and analytics with reporting of on-time performance, cycle and transit times, detention and demurrage, among others; etc. [0072] In some aspects, the predictive tracking management system may identify On-time, In Full (OTIF) metrics, fulfillment lead times, etc. trending in relation to defined targets. OTIF metrics are a performance indicator for measuring how many orders were delivered on time and in full. In some aspects, when a metric is lagging, the predictive tracking management system may be configured to link causes of delay and address a root cause of the delay. [0093] The load prediction engine 716 is trained to use identify factors that may cause delay to the load and output various information, such as a delay probability and a confidence of the delay probability to a recommendation engine 718.) Identifying performance metrics and measuring when a metric is lagging, resulting in the prediction of the root cause satisfies the limitation.
-receives suggested applicable actions from the model to resolve the predicted degradations, (Rajagopal [0026] The predictive tracking management system can include a machine learning (ML) model or other artificial intelligence (AI) network that is capable of predicting the route a load (e.g., a package, parcel, or freight) is going to travel from pickup to delivery and/or throughout a trip from pickup to delivery and issues that may adversely affect the transit of the load. [0027] Based on the predicted route, the ML model can predict an estimate that the load will be delayed and an estimated delivery date and time. The ML model can also provide a confidence in the estimation and may provide information to the customer that identifies risks. In some aspects, the ML model can also provide suggestions for alternate carriers based on delivery requirements, alternate delivery destinations, alternate shipping originations, and so forth. The ML model can also monitor the load based on the route after consignment to the carrier, and may provide options to remediate a potentially delayed load. [0093] The recommendation identifies potential resolutions that can be addressed through the carrier, such as escalating a service level or attempting to request the carrier to reroute the load. [0104] At block 810, the computing system may, in response to determining that the predicted delivery date is after the requested delivery date, identify at least one corrective modification to the transit configuration for the load. )
-feeds the predictions and suggested actions to an enterprise resource planning (ERP) system (Rajagopal [0048] In some examples, the supply chain management system 100 can provide integration (e.g., via automated processing, via application-programming interfaces, etc.) between various systems and enable data sharing between the supply chain management system 100 and ... enterprise resource planning (ERP) platform, etc. [0068] At block 415, after the consignment of the load to the carrier, the predictive tracking management system may identify risks and exceptions to shipments based on a combination of predictive analytics and identification of issues within the carrier ... The predictive tracking management system may be configured to provide notifications to a client system when an unacceptable risk for the load, which is consigned to the carrier, is identified Non-limiting examples of a client system include a ... a manufacturing execution system (MES), a TMS, a warehouse management system (WMS), an ERP system, a bespoke system, and so forth. [0071] At block 425, the predictive tracking management system may be configured to orchestrate resolution options to resolve any issues. In some cases, the predictive tracking management system can be configured to orchestrate an external system, enterprise system, or third party companies or system, such as a WMS or MES to identify potential resolutions.) Rajagopal’s sending notifications to an ERP system about a predicted unacceptable risk is an example of feeding predictions to an ERP system. Orchestrating resolution options on an enterprise system is an example of feeding suggested actions to the ERP system.
-for analysis based at least on at least the predictions and the suggested actions, and (Rajagopal [0072] In some aspects, when a metric is lagging, the predictive tracking management system may be configured to link causes of delay and address a root cause of the delay. For example, shipments on Thursday may have a higher chance of being delayed, which causes goods or materials to be unavailable. In some cases, the predictive tracking management system uses an ML model to identify the root causes based on a variety of unstructured data. In some aspects, the various recommendations may be tracked through a workflow engine for supervising various predictive tracking management system tasks. [0077] The load prediction engine 522 may be configured to use the classified data, rules, and routes from the carrier engine 514, and carrier predictions from the carrier prediction engine 516 to predict various aspects of the shipment. In one illustrative example, the load prediction engine 522 may identify a likelihood of delay associated with that route based on the various events and information collected by the predictive tracking management system 500...) Using Rajagopal’s “rules” data from [0077], is an example of analyzing business rules based on carrier predictions, and suggested actions(recommendations/corrective actions). For example “shipping on a Thursday” is an example of a business rule, with predictions and suggestions to modify such rules found in at least [0072-0074] in the form of recommendations.
-implements a change based on an instruction received from the ERP system based at least on the analysis. (Rajagopal [0080] the resolution engine 528 can be integrated into the client system 540 {[0073] the client system 540 can be various systems such as an ERP} and can be configured to identify alternative resolutions, such as diverting resources from one location to the intended recipient of the delayed load. In some cases, the resolution engine 528 may be able to resolve changes to the transit configuration by changing service level, intercepting the load with an agent, changing delivery location, and so forth. [0107] At block 814, the computing system may identify at least one corrective modification to the transit configuration for the load. For example, the carrier may allow modification of the load while in transit, and the computing system may identify corrective actions (e.g., escalate the service level) to minimize the delay probability. [0071] orchestrate resolution options to resolve any issues. In some cases, the predictive tracking management system can be configured to orchestrate an external system, enterprise system, or third party companies or system, such as a WMS or MES to identify potential resolutions. In one aspect, the predictive tracking management system may be configured to execute a workflow engine to manage execution of various recommendation systems ...) Resolving changes by intercepting the load, changing delivery location, or coordinating the corrective action with the carrier is an example of implementing a change in business rules based on corrective modifications. Since the resolution engine is implemented into the client system(which we know from [0073] includes ERP systems), then we know that the instructions come from the ERP systems in Rajagopal.
However, Rajagopal does not explicitly teach:
- that the feeding of the predictions and suggested actions to an enterprise resource planning (ERP) system for the analysis based at least on the predictions and the suggested actions is specifically “for the analysis of business rules based at least on the predictions and suggested actions” (Though Rajagopal does use business data, and rules as part of the analysis (see [0077]), the analysis isn’t of the business rules in particular.)
-implements a change in business rules.
Alternatively, Najmi discloses a self-learning supply chain management system that enables root cause analysis of supply chain execution failures and provides tools to proactively resolve supply chain disruptions. Najmi teaches: - the analysis of business rules based at least on the predictions and suggested actions(Najmi [0039] Business rules configuration manager 234 provides for business configuration analysis 304 by providing a user interface to compute, monitor, and change any one of business rules 348, model attributes 350, and/or optimization settings 352. [0087] In some embodiments, self-learning system 110 uses pattern detection, machine learning and statistical process control techniques to monitor and detect when members of a given segment fail to conform to expected and predicted behavior patterns. In some embodiments, self-learning system 110 detects deviations from anticipated behavior. [0056] The learning cycle process 502 compiles information from self-learning supply chain system 110 to generate contingency plans for supply chain disruptions which were not known to occur by storing, for example, the types of levers used to overcome the supply chain disruption and the effects the levers had in remedying the disruption. These solutions to supply chain disruptions comprise risk management, adaptability, agility, and continuous alignment of business objectives and ongoing execution.)
-implements a change in business rules based at least on the analysis.(Najmi [0056] Self-learning supply chain system 110 uses rules to solve, monitor, and analyze performance across PDCA cycles. In some embodiments, this validates and refines planning assumptions on an ongoing basis. Learning cycle process 502 comprises updates, refinements, and reconfiguration of the assumptions, business rules, and planning models. [0057] Self-learning system 110 reconfigures rules and parameters 222 across a supply chain management software suite using, for example, industry and business model-specific templates and wizards...Self-learning system 110 automates decision support workflows and prioritizes and resolves supply chain disruptions or one or more supply chain entities 120. [0059] Self-learning system 110 continuously refines many factors to drive superior performance on an ongoing basis, to enable self-learning. Factors include, for example, assumptions, models, business rules, diagnosis paths, levers, resolution paths, and performance scorecards.) As seen in Najmi, the business rules themselves are analyzed in order to be updated, refined and reconfigured to optimize performance.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the present disclosure to further modify Rajagopal by adding the analysis of business rules and implementing a change in business rules based on the analysis as seen in Najmi. By simply substituting Rajagopal’s step of “feeds the predictions and suggested actions to an enterprise resource planning (ERP) system for analysis(of root causes)” to instead be an analysis of business rules instead as taught by Najmi, one would arrive at the predictable outcome of “feed(ing) the predictions and suggested actions to an enterprise resource planning (ERP) system for analysis of business rules based at least on at least the predictions and the suggested actions, and implements a change in business rules based on an instruction received from the ERP system based at least on the analysis.” This outcome is predictable because Rajagopal already analyses causes of delay, therefore, adding an analysis of the business rules can simply be performed by inserting the business rule data into the machine learning model. One would have been motivated to perform this combination by the benefit of increasing performance and accurately anticipating degradations. (Najmi [0056] Learning cycle process 502 replaces unknown unknowns with known unknowns, which may thereby incorporate contingencies into a supply chain plan. An unknown unknown is a supply chain disruption that a planner is not aware might occur. A known unknown, by contrast, are those supply chain disruptions which are known to occur, even if the timing or extent of the disruption is unknown ahead of time... These solutions to supply chain disruptions comprise risk management, adaptability, agility, and continuous alignment of business objectives and ongoing execution.)
Regarding Claim 18:
The combination of Rajagopal and Najmi teaches the system of claim 16,
However, Rajagopal fails to teach:
-wherein prior to passing the information to the model the system trains the model to recognize anomalies in the information.
Alternatively, Najmi teaches:
-wherein prior to passing the information to the model the system trains the model to recognize anomalies in the information.(Najmi [0061] Self-learning system 110 comprises a plurality of closed loop performance monitoring systems 602, 604, 606, and 608. These systems strategically mine relevant data generated during the normal course of business of one or more supply chain entities 120 and use this data to provide early detection of deviations from a supply chain plan, validate assumptions, expand and assess options, and improve performance. [0079] In some embodiments, self-learning system 110 enables early detection of sources, or suspected sources, of risk and capitalizes on opportunities to start proactively detecting the sources to maximize available reaction time... In a similar manner, a structured analysis method of self-learning system 110 captures likely failure patterns and accumulates and refines root cause analysis maps based on historical data. [0030] In this document, the terms “disruptions,” “problems,” “perturbations,” “changes,” or “events” may refer to any positive or negative deviation, condition, pattern, or occurrence within the supply chain plan or during execution of the plan that can motivate action by a supply chain planner.) Refining a self-learning system based on historic data is an example of training a model. A deviation from a supply chain plan is an example of detecting anomalies, since deviation is used interchangeably with terms like “disruptions,” “problems,” “perturbations,” “changes,” or “events.”
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the present disclosure to further modify Rajagopal by adding the training of a model to determine anomalies in data as taught by Najmi. One of ordinary skill in the art would have been motivated by the benefit of the self-learning system of Najmi improving in accuracy over time. (Najmi [0046])
Regarding Claim 19:
The combination of Rajagopal and Najmi teach the system of claim 16,
Furthermore, Rajagopal teaches:
- wherein the system promotes analysis of logistics processes and execution of related business rules in an autonomous manner that does not require changes to existing business processes.(Rajagopal [0048] In some examples, the supply chain management system 100 can provide integration (e.g., via automated processing, via application-programming interfaces, etc.) between various systems and enable data sharing between the supply chain management system 100 and other shipper and/or supply chain systems such as, for example... enterprise resource planning (ERP) platform, etc. [0068] In some aspects, the predictive tracking management system may be integrated into the client system to monitor information and receive information from the client system to engage carrier shipping services, freight shipping services, and so forth...[0071] At block 425, the predictive tracking management system may be configured to orchestrate resolution options to resolve any issues. In some cases, the predictive tracking management system can be configured to orchestrate an external system, enterprise system, or third party companies or system... In one aspect, the predictive tracking management system may be configured to execute a workflow engine to manage execution of various recommendation systems associated with different parties. [0073] In some aspects, the predictive tracking management system 500 is configured to integrate with a client system 540 and uses various third-party services 550, information services 560, and public APIs 570 to predict load delivery issues and exceptions.) The analysis of logistics process and execution of related business rules in Rajagopal does not require changes to existing business processes, because it is merely integrated(via automated processing) and enable data sharing to monitor the information. As seen in Rajagopal [0071] and [0073], the predictions can be performed on third party systems, meaning that the predictions can be performed with existing historic data(without changing existing business processes).
Regarding Claim 20:
Rajagopal teaches the system of claim 16,
-wherein the system provides output from natural language processing (NLP) systems about the first logistics process to the model to supplement the metric, accountability, rule, and action information.(Rajagopal [0076] For example, the event identification engine 518 may be configured to extract structured data from third-party services 550. Non-limiting examples of third-party services include ... generational chat services (e.g., ChatGPT), ML-based classification services, and so forth... In some cases, the event identification engine 518 can also connect to various information services, such as a news service or a weather service, to monitor for alerts and other pertinent information that may affect shipment. The event identification engine 518 can also receive information from the public API. For example, the public API can be a weather service or an API associated with the carrier shipping service and receive structured data. In some cases, the public API can be a propriety service using a large language model (LLM) such as a generative pre-trained transformer (GPT) that is trained to summarize, extract, or process information.) Use of large language models such as ChatGPT to summarize, extract, or process information (to supplement the event identification engine 518 configured to identify events pertinent to predictive tracking management system), is an example of providing output from NLP systems, about the first logistics process to the model. Since the Predictive tracking management system includes the model that processes the metric, accountability, rule, and action information (see Rajagopal [0026]), the limitation has been satisfied.
13. Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Rajagopal (US 20240362576 A1), in view of Adeel Najmi (US 20160217406 A1) hereinafter Najmi, further in view of Dominik Held (US 20120096044 A1) hereinafter Held.
Rajagopal teaches the system of claim 16,
-Furthermore, Rajagopal teaches
-machine learning model (Rajagopal [0076] For example, the event identification engine 518 may be configured to extract structured data from third-party services 550. Non-limiting examples of third-party services include commercial weather services, commercial traffic prediction services, generational chat services (e.g., ChatGPT), ML-based classification services, and so forth.)
However, Rajagopal fails to teach:
- wherein the system further predicts table headers of data sources comprising at least one of ERP system, master data, and external data via an ML model trained on a standard or user-defined dictionary of table fields and using data source meta data comprising at least one of database name, table name and field name as features to predict a standard table field corresponding to a data source schema field, such mapping promoting preparation of a local dictionary mapping of each data source element to a standard table field name, the present approach further using data source field names and data relevance to assist in suggestive process resulting in identification of corresponding table field.
Held discloses systems for data allocation identification, including determining a database allocation and identifying a domain directly associated with the entity. This is analogous to functions in recited in claim 17 of the present claims, since they both relate to the structuring of unstructured data using dictionaries. Held teaches:
- wherein the system further predicts table headers of data sources (Held [0003] The method may further comprise determining at least one header table field directly associated with the entity by identifying a directly associated type definition linked to the directly associated domain, wherein the directly associated type definition refers to the header table field, wherein each header table field is in a corresponding header table. [0054] If 100 different items are order using a single sales order, the database may include one sales order in the header table, and 100 records in at least one detail table. The record in the header table and all the records in the line item table may include the same linking field (i.e. the records include a linking field with the same document number).) Determining a header table field is an example of predicting table headers. The data sources are mapped to the databases and records about an entity as taught in Held.
-comprising at least one of ERP system, master data, and external data (Held (ERP System): [0058] These tasks can be carried out using the technical definition of the entity in a database of an enterprise resource planning system. In particular, these tasks can be performed by determining the database allocation of the entity. Master Data: [0061] In the present example, a domain 101 is directly associated with the entity. The domain 101 may be understood as an object, a technical object (i.e. an object with technical attributes) or a unique repository object (i.e. the domain is used in a repository and each domain is unique). The direct association between the domain 101 and the entity may be implemented by creating a link between the domain 101 and the entity in a data dictionary, e.g. the SAP dictionary. Also, the domain 101 may be associated with a system, e.g. a database system. In some cases, the domain 101 is linked to one or more customizing tables and the entity may define an entity category. The customizing tables linked to the domain 101 may include all the members of the entity category that are available on the system, as well as identifiers of the members and links to other information. External data:[0105] The method described above can be applied on databases storing any kind of data, such as business data or technical data.) Held’s domain is a master database because it stores the link between the entity in the data dictionary, and includes all the members, as well as identifiers of the members and links to other information.
-via a model trained on a standard dictionary of table fields (Held [0060] FIG. 1 depicts a data model of relationships in a database and a partial database allocation of an entity. [0061] The direct association between the domain 101 and the entity may be implemented by creating a link between the domain 101 and the entity in a data dictionary, e.g. the SAP dictionary. [0066] FIG. 2 depicts specific relationships in a database and a partial database allocation. In other words, FIG. 2 depicts a specific implementation of the data model shown in FIG. 1. [0074] The data dictionary may be implemented as table or database metadata, or as a distinct data structure. The data dictionary information may be understood to create a direct association between the EKORG type definition 203, the ADDI_EKORG type definition 205, the HEKORG type definition 207, and the EKORG domain 201. In other words, the links in the data dictionary directly associate the type definitions with the domain.) The model of relationships in a database is an example of a model “trained on a standard dictionary” because it stores the relationships and mappings of data to their standardized forms.
-and using data source meta data comprising at least one of database name, table name and field name as features to predict a standard table field corresponding to a data source schema field, (Held data source metadata: [0095] Domains and type definitions may be stored as metadata for each individual table. Database name: [0071] In another example, a VKORG domain (i.e. the domain with the name VKORG) is linked to a sales organization entity. The VKORG domain uses a customizing table with the name TVKO for storing mem