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 .
Status of Claims
Applicant's submission filed on 8/22/25 has been entered. Claims 1-20 are cancelled. Claims 21-43 are presented for examination.
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.
Note: The following analysis is based on the Revised Guidance titled “2019 Revised Patent Subject Matter Eligibility Guidance (Vol. 84, No. 4).
STEP 1
Are the claims directed to a process, machine, manufacture or composition of matter?
Claims 21-43 are all directed to a statutory category (e.g., a process, machine, manufacture, or composition of matter). The answer is YES.
STEP 2A. Prong 1
The claims disclose the abstract idea of the process of determining a schedule for acquiring materials to be used by a manufacturing or processing plant to produce one or more products.
.
Exemplary claim 21 recites the following abstract concepts that are found to include “abstract idea”:
“obtaining information defining multiple orders for one or more items over time;
determining, based on the obtained information, whether to retrain a first machine learning algorithm;
responsive to determining not to retrain the first machine learning algorithm, determining a classification threshold and a prediction mask, wherein the determined classification threshold and prediction mask are parameters of the first machine learning algorithm
identify one or more of the orders that are likely to change based on classification of the orders;
determine one or more lengths of time that the one or more orders are likely to change based on regression of the orders; and
generating a schedule identifying materials to be used to manufacture or produce the one or more items based on the one or more lengths of time that the one or more orders are likely to change.”
controlling one or more manufacturing or production operations that are performed according to the schedule;”
The remaining limitations are no more than computer elements (i.e., a processing device, processors) to be used as a tool to perform this abstract idea.
The recited limitations cover a process that, under its broadest reasonable interpretation, covers subject matter viewed as a certain method of organizing human activity with the additional recitation of generic computer components. For example, but for the “causes one or more processors” language, “obtain, identify, determine, generate” etc.. in the context of this claim encompasses the user manually obtaining the information, identifying orders that are likely to change and the length of time they are likely to change based on a pattern (classification and regression), generating a schedule to acquire materials.
The practice of obtaining, identifying, determining, and generating data, based classification is a commercial or legal interaction long prevalent in our system of commerce. The claims recite the idea of performing various conceptual steps generically resulting in the generation of a plan/schedule. As determined earlier, none of these steps recites specific technological implementation details, but instead get to this result by receiving, selecting and determining data. Thus, the claims are directed to a certain method of organizing human activity
STEP 2A, Prong 2
Are there additional elements or a combination of elements in the claim that apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that it is more than a drafting effort designed to monopolize the exception?
The claim recites one additional element: that a machine learning algorithm is used to perform the identifying, determining steps, and the first machine learning algorithm is trained using multiple types of features.
The machine learning algorithm in the steps is recited at a high level of generality, i.e., as a generic processor performing a generic computer function of processing data (identify, by a machine learning algorithm, orders). This generic processor limitation is no more than mere instructions to apply the exception using a generic computer component.
Accordingly, this additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
The claim is directed to an abstract idea.
STEP 2B
The next issue is whether the claims provide an inventive concept because the additional elements recited in the claims provide significantly more than the recited judicial exception. Taking the claim elements separately, the function performed by the computer system at each step of the process is purely conventional. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of “using a machine learning algorithm to identify one or more of the orders that are likely to change” and “causing the first machine learning algorithm to determine, … one or more lengths of time that the one or more orders are likely to change based on regression of the orders” amount to no more than mere instructions to apply the exception using a generic computer component. Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The claim is not patent eligible.
Considered as an ordered combination, the computer components of Applicants' claims add nothing that is not already present when the steps are considered separately. The claimed invention does not focus on an improvement in computers as tools, but rather certain independently abstract ideas that use computers as tools. {Elec. Power, 830 F.3d at 1354). (Step 2B: NO).
There is no indication that the processor or machine learning algorithm is anything other than a generic, off-the-shelf computer component, and the Symantec, TLI, and OIP Techs. Court decisions cited in MPEP 2106.05(d)(II) indicate that mere collection or receipt of data over a network is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is here).
Independent claims 28 and 35 recite similar limitations as claim 21 and are therefore rejected under the same rationale.
The dependent claims when analyzed as a whole are held to be patent ineligible under 35 U.S.C. 101 because the additional recited limitations fail to establish that the claims are not directed to an abstract idea. The claims provide minimal technical structure or components for further consideration either individually or as ordered combinations with the independent claims. As such, additional recited limitations in the dependent claims only refine the identified abstract idea further. Further refinement of an abstract idea does not convert an abstract idea into something concrete.
Accordingly, a conclusion that the collecting step is well-understood, routine, conventional activity is supported under Berkheimer Option 2.
See MPEP 2106.05(d)(II) The courts have recognized the following computer functions as well-understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity.
i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); TLI Communications LLC v. AV Auto. LLC, 823 F.3d 607, 610, 118 USPQ2d 1744, 1745 (Fed. Cir. 2016) (using a telephone for image transmission); OIP Techs., Inc., v. Amazon.com, Inc., 788 F.3d 1359, 1363, 115 USPQ2d 1090, 1093 (Fed. Cir. 2015) (sending messages over a network); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350,1355,112 USPQ2d 1093,1096 (Fed. Cir. 2014) (computer receives and sends information over a network); but see DDR Holdings, LLC v. Hoteis.com, L.P., 773 F.3d 1245, 1258, 113 USPQ2d 1097, 1106 (Fed. Cir. 2014) ("Unlike the claims in Ultramercial, the claims at issue here specify how interactions with the Internet are manipulated to yield a desired result-a result that overrides the routine and conventional sequence of events ordinarily triggered by the click of a hyperlink." (emphasis added));
iv. Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306,1334,115 USPQ2d 1681,1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363,115 USPQ2d at 1092-93.
The claims are ineligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 21-23, 25-30, 32-37, 39, 40 rejected under 35 U.S.C. 103 as being unpatentable over Gopinath et al. (20150379536A1), in view of Godfrey et al. (US 20200279192 A1), in further view of Ohlsson et al. (US 20200143313 A1).
Re-claim 21, Gopinath et al. teach a method comprising:
-obtaining information defining multiple orders for one or more items over time;
(see e.g. paragraphs [0057] A data preprocessor 186 and a predictor 182 can be configured to receive point-of-sale data,
[0061] point-of-sale data and/or inventory data might be received and stored in a coarse time-wise data format (e.g., levels by week, or levels by month) and can be converted into a highly granular time-wise data formats (e.g., levels by day or by hour). The received data can be stored persistently, and thus a history can be assembled.
[0063] The point-of-sale data refers to an item 117 or items sold on a given date or dates and/or sold within specified date ranges (e.g., daily sales 164, weekly sales 176, monthly sales 170, etc.).
[0047] As shown, a single instance 115 is used for combining data from various points of sale (e.g., store_A), and to provide an ordering forecast to a particular supplier or manufacturer (e.g., MFG_P, MSG_Q, MFG_R, etc.) This architecture supports processing orders from any combinations of orders from stores and orders from other points in the distribution network, if such orders are available.
[…] - identify one or more of the orders that are likely to change based on classification of the orders;
(see e.g. paragraphs [0057] In some cases, the inventory can be predicted to be too low (e.g., relative to a period of high demand from the retail outlets), and a predictor 182 might increment an existing order, or might issue a supplemental order to the manufacturer of the items that are drawn down during periods of high demand from the retail outlets. In some cases, and as shown in FIG. 1B1, the predictor 182 includes a time- wise and quantity-wise model for ordering, and an additional module processes the time-wise and quantity- wise model to generate one or more orders that comport with the order formatting and submission characteristics as may be specified by the various manufacturers. [0041] Similarly, the time frames referring to the need for orders to be fulfilled might vary substantially, depending on the nature of the agent or entity. Strictly as one example, a point-of-sale agent or entity might raise orders based on a forecast through an upcoming month. Or, a distribution center agent or entity might raise orders based on a forecast through an upcoming quarter, or even longer period of time. As shown, the granularity of quantities and time frames increases through the progression from supplier to points of sale.)
- generating a schedule to acquire materials to be used to manufacture or produce the one or more items based on the one or more lengths of time that the one or more orders are likely to change.
(see e.g. paragraph [0058] The manufacturers in turn use such data in their own procurement systems so as to procure enough raw materials and/or partially-finished goods be able to manufacture enough quantities, and in time to satisfy the flow of orders.
[0061] Further, a calendar is superimposed over a forecast period, and a predicted days-of-supply and/or store lead-time can be calculated and used by an order management and/or by a manufacturing procurement system.)
--controlling one or more manufacturing or production operations that are performed according to the schedule;
(see e.g. paragraph [0058] The manufacturers in turn use such data in their own procurement systems so as to procure enough raw materials and/or partially-finished goods be able to manufacture enough quantities, and in time to satisfy the flow of orders.
[0061] Further, a calendar is superimposed over a forecast period, and a predicted days-of-supply and/or store lead-time can be calculated and used by an order management and/or by a manufacturing procurement system.)
Although Gopinath et al. teach identifying one or more of the orders that are likely to change based on classification of the orders;
Gopinath et al. do not explicitly teach using machine learning algorithm.
However, Godfrey et al. teach using machine learning algorithm. -determining, based on the obtained information, whether to retrain a first machine learning algorithm;
(see e.g. [0016] A given machine learning model may be utilized to provide a prediction with respect to some set of input data. In an example, for a given machine learning model to provide an accurate prediction, data that the machine learning model has utilized to make the prediction should have a similar distribution as the training data on which was used to train the model. In practice, however, data distributions can change over a period of time. Thus, deploying a model, in practice, typically is not a one-time occurrence and can involve retraining the model with new data and then deploying the retrained model. Consequently, it may be advantageous to determine whether incoming data has a distribution that significantly deviates from the distribution of the training data in order to determine whether retraining the model would be beneficial.
--responsive to determining not to retrain the first machine learning algorithm, determining a classification threshold and a prediction mask, wherein the determined classification threshold and prediction mask are parameters of the first machine learning algorithm;
(see e.g. [0046] As further shown in the statement 410, a threshold value (e.g., “X”) can be selected, which is included in a relationship where if a score is greater than or equal to the value of the threshold value, then the machine learning model (e.g., the upstream model) will predict that the score belongs in a class corresponding to fraudulent activity (e.g., “FRAUD”). In other words, the threshold value can correspond to a boundary value of an output variable (e.g., the score) that is utilized to assign the output variable in one class or another class (e.g., fraud or non-fraud).
[0028] As illustrated, the electronic device 110 includes constraint data 212 corresponding to a data specification for providing semantics (e.g., providing meaning to output values that may be interpreted by a downstream model) to respective values from a distribution of values. In an example, such semantics are defined based on a set of constraints that, for a binary classification problem, indicate respective probabilities for an output of the model to be in one class or another class. In particular, an example set of constraints can include respective indicators of confidence to corresponding score values for assigning a classification to a particular score value. In the context of machine learning, such semantics enable a machine learning model to interpret and associate meaning to the values which can facilitate a more accurate analysis of the data in order to provide, in an example, a prediction or classification. An example of constraint data is discussed in more detail in FIG. 3 below. As further shown, the server 120 includes constraint data 212 for storing information corresponding to one or more sets of constraints.
[0057] FIG. 7 illustrates a flow diagram of an example process 700 for determining a classification based on a set of constraints and a score received from a client model in accordance with one or more implementations.
[0031] As discussed further herein, such a score value may be utilized in conjunction with a threshold value and/or a set of constraints to determine a particular classification in a binary classification model. Moreover, the destination device model 260 can utilize server-side signals 270 (or utilize a rule-based mechanism) in conjunction with the assessment received from the source device model 220 in order to make a decision or initiate an action to be performed by the server 120.
[0058] The server 120 receives an assessment value from a source electronic device (e.g., the electronic device 110 or the electronic device 115) where the assessment value is provided from an output of a first machine learning model deployed on the source electronic device (710). The server 120 determines, using a second machine learning model deployed on the server 120, a classification of the assessment value based on a set of constraints, where the set of constraints is utilized by the source electronic device and the destination electronic device to at least define a probability that the assessment value corresponds to a particular classification (712). Further, the server 120 performs an action based at least in part on the classification of the assessment value (714).
--wherein the second machine learning algorithm is trained using multiple types of features.
(see e.g. [0024] As discussed further below, a machine learning model may be trained by the server 120, and then deployed to a client such as the electronic device 110. Further, the electronic device 110 may provide one or more machine learning frameworks for developing applications using such machine learning models. In an example, such machine learning frameworks can provide various machine learning algorithms and models for different problem domains in machine learning.
[0026] The machine learning model deployed on the server 120 can then perform one or more machine learning algorithms using the input provided from the electronic device 110.).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Gopinath et al., and include the steps cited above, as taught by Godfrey et al., because training a model involves extensive computing resources, and retraining can adversely impact the interoperability between the respective machine learning models by impacting the accuracy of the machine learning models until retraining and redeployment of the retrained models are completed.(see e.g. [0017], [0018]).
Gopinath et al., in view of Godfrey et al., do not explicitly teach using machine learning algorithm.
However, Ohlsson et al. teach --using a second machine learning algorithm to identify one or more of the orders that are likely to change based on classification of the orders;
(see e.g. [0046] In order to stay competitive in a dynamic market, businesses often need to constantly reconfigure supply chains and to manage the various types of uncertainties that are continually being introduced (e.g., both from demand-side as well as from supply-side).
[0016]. The system can process the inventory dataset using a trained machine learning model to generate a prediction of the plurality inventory variables that are characterized by having one or more future uncertainty levels. The system can provide the processed inventory dataset to an optimization algorithm. The optimization algorithm can be used to predict a target inventory level for optimizing an inventory holding cost.
[0012] In some embodiments, the trained algorithm comprises a machine learning algorithm. In some embodiments, the machine learning algorithm is selected from the group consisting of a support vector machine (SVM), a naïve Bayes classification, a linear regression, a quantile regression, a logistic regression, a random forest, and a neural network.
[0061] The method for inventory management and optimization may further use the statistical distribution to generate the prediction of the stochastic variables having future uncertainty. The prediction may comprise a distribution of the stochastic variables having future uncertainty. For example, the distribution of the stochastic variables having future uncertainty may include a distribution over a future duration of time. The future duration of time may correspond to a planning horizon for the material being optimized. The planning horizon may vary depending on the customer, supplier, and the material.
[0054] For example, variables having future uncertainty (e.g., stochasticity) may include inventory level, supply factors, supplier orders, demand factors, demand forecast, material consumption, transit time, lead time, material requirements planning (MRP), inventory holding cost, and shipping cost.
Note: “The variables having future uncertainty including a distribution over a future duration of time/. a planning horizon” encompasses “lengths of time that the one or more orders are likely to change”.
Although Gopinath et al. teach - generating a schedule to acquire materials to be used to manufacture or produce the one or more items […]
(see e.g. paragraph [0058] The manufacturers in turn use such data in their own procurement systems so as to procure enough raw materials and/or partially-finished goods be able to manufacture enough quantities, and in time to satisfy the flow of orders.
[0061] Further, a calendar is superimposed over a forecast period, and a predicted days-of-supply and/or store lead-time can be calculated and used by an order management and/or by a manufacturing procurement system.)
Ohlsson et al. explicitly teach --generating a schedule to acquire materials to be used to manufacture or produce the one or more items based on the one or more lengths of time that the one or more orders are likely to change."
(see e.g. [0075] In some embodiments, methods and systems for inventory management and optimization may perform optimization of a demand forecast. In the case of manufacturers, the demand forecast (or planned consumption) for each item over a planning horizon can be computed using the forecasted demand for the finished products, the product configurations, and a bill of materials (BOM). BOM is a dynamic (time-varying) hierarchical graph that provides a list of all the items and intermediate assemblies required to manufacture a finished product, along with their quantities.
[0086] For example, the probabilistic graphical model can be used to compute optimal scheduled arrivals at any given time t into the future for such a supply chain network, or an optimal safety stock and safety time for each facility in the network.
[0064] Manufacturers typically use rule-based systems to determine the timing and size of orders that need to be placed with suppliers. The static rules may not account for any variability which may yield sub-optimal inventory levels. Using artificial intelligence based (e.g., machine learning) approaches, inventory management and optimization can be formulated as a constrained optimization problem, where a goal is to minimize a cost function (e.g., an inventory holding cost) under one or several constraints, such as a confidence level of availability of stock.
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Gopinath et al., in view of Godfrey et al., and include the steps cited above, as taught by Ohlsson et al., because by applying machine learning to accurately manage and predict inventory variables with future uncertainty (see e.g. [0048]).
Re-claim 22, Gopinath et al. teach the method of Claim 21, wherein the multiple types of features comprise (i) one or more order-specific features of the orders, (ii) one or more time-varying aggregations of the orders, and (iii) one or more static features each associated with multiple ones of the orders.
(see e.g. paragraph [0057] In some cases, the inventory can be predicted to be too low (e.g., relative to a period of high demand from the retail outlets), and a predictor 182 might increment an existing order, or might issue a supplemental order to the manufacturer of the items that are drawn down during periods of high demand from the retail outlets. In some cases, and as shown in FIG. 1B1, the predictor 182 includes a time-wise and quantity-wise model for ordering, and an additional module processes the time-wise and quantity-wise model to generate one or more orders that comport with the order formatting and submission characteristics as may be specified by the various manufacturers.)
Re-claim 23, Gopinath et al. teach the method of Claim 22, wherein: the one or more static features comprise indications of how similar different items are based on prior behaviors of one or more entities; and the prior behaviors are represented as embeddings, each embedding generated using words that represent items and sentences that represent sequences of items requested by the one or more entities.
(see e.g. paragraphs [0072] A datastore comprising a history of orders 210 and/or a history of ordering patterns 211 is accessed. An operation (see operation 205) performs various statistical analysis on the history of ordering patterns 211.
[0081] FIG. 4A shows time periods T1, T2, and T3 over a continuous time period depicted by the abscissa (x-axis). The origin (y-axis) depicts volume or quantities of an item. Superimposed onto the graph of case study 4A00 is an indication of a distributor's initial replenishment quantity (see point 406). The chart depicts depletion of the distributor's initial replenishment quantity over time (see line segment 408). As shown, the depletion of the quantity over time is inverse to the ongoing demand line (see line segment 410). At some point in time, the distributor's initial replenishment amount becomes depleted (e.g., see the boundary between time period T1 and time period T2).
[0073] The statistical analysis (e.g., correlation analysis) performed in operation 205 facilitates identification of ordering patterns, and such ordering patterns might be of a nature that they re-occur. For example, a point-of-sale outlet might sell a large number of strawberry pop-tarts during the hurricane season. And, since the hurricane season re-occurs annually, the annually-recurring pattern of ordering of strawberry pop-tarts emerges from analysis of the ordering history.
[0074] In an exemplary situation, the identified historical ordering patterns (such as “last year's hurricane season”) are applied to future time windows such as “the upcoming hurricane season” (see operation 208). During this or subsequent steps, a data structure can be prepared to organize items and corresponding quantities into a forecast that observes inventory model parameters 213. Inventory model parameters include delivery intervals, maximum volume to be delivered at any interval, service levels, etc.)
Re-claim 25, Gopinath et al. anticipates the method of Claim 21, wherein generating the schedule comprises: decrementing an order estimation based on the orders and the one or more lengths of time that the one or more orders are likely to change; and allocating the decremented order estimation over a specified horizon.
(see e.g. paragraphs [0087] The selection of a safety stock strategy, and/or modifications to safety stock parameters (e.g., to increase levels 434, or to decrease levels 436), can be performed iteratively. A replenishment plan can be generated using the selected safety stock strategy, and selected safety stock parameters.
[0041] the time frames referring to the need for orders to be fulfilled might vary substantially, depending on the nature of the agent or entity.)
[0051] Ordering safety stock can be effective in some situations, however in some situations the stock is expensive and/or perishable, and in many situations the ordering of safety stock tends to induce unwanted effects such as undesirable financial effects (e.g., resulting from carrying charges) and/or unwanted operational effects (e.g., excess stock on hand, spoiled or expired goods, etc.). Techniques that combine wholesale ordering techniques (e.g., ordering based on distribution center forecasts and/or inventory levels) with retail-level ordering techniques (e.g., ordering based on point-of-sale data) can be employed to improve accuracy and avoid a range of unwanted effects.
[0056] As earlier indicated, the practice of over-ordering (e.g., by wholesale channels and/or distribution channels) to amass inventory, “just in case” still sometimes fails to mitigate the potential for lost sales during periods of higher demand Further as earlier indicated, amassing the excess stock may be expensive or may be impractical.
*******The Examiner notes that Gopinath et al. teach increasing or decreasing levels of safety stocks and caution against over-ordering. Although Gopinath et al. teach that [0057] In some cases, the inventory can be predicted to be too low (e.g., relative to a period of high demand from the retail outlets), and a predictor 182 might increment an existing order, or might issue a supplemental order to the manufacturer of the items that are drawn down during periods of high demand from the retail outlets.”, it is considered an obvious variation of Gopinath et al. to decrementing an order estimation. No unpredictable results are foreseen since there are a finite number of identified predictable solutions with a reasonable expectation of success to choose from: 1) increment the order estimation, 2) decrement the order estimation or 3) apply no change to the order estimation.
Re-claim 26, Gopinath et al. teach the method of Claim 25, wherein allocating the decremented order estimation over the specified horizon comprises allocating the decremented order estimation to times within the specified horizon that are not associated with the orders.
(see e.g. paragraph [0057 n some cases, the inventory can be predicted to be too low (e.g., relative to a period of high demand from the retail outlets), and a predictor 182 might increment an existing order, or might issue a supplemental order to the manufacturer of the items that are drawn down during periods of high demand from the retail outlets. In some cases, and as shown in FIG. 1B1, the predictor 182 includes a time-wise and quantity-wise model for ordering, and an additional module processes the time-wise and quantity-wise model to generate one or more orders that comport with the order formatting and submission characteristics as may be specified by the various manufacturers. – [0061] Further, a calendar is superimposed over a forecast period, and a predicted days-of-supply and/or store lead-time can be calculated and used by an order management and/or by a manufacturing procurement system.)
Re-claim 27, Gopinath et al. teach the method of Claim 21, further comprising: generating multiple schedules; and using one or more metrics to compare the multiple schedules.
(see e.g. paragraph [0069] The aforementioned forecast can be used to generate a replenishment plan (e.g., a time-sequenced safety stock plan) such that an order for a particular item 117 can be broken down into quantities and corresponding time frames so as to, for example, schedule deliveries in accordance with delivery intervals (see operation 212). In some cases, the time frames include calculation of lead times (e.g., lead time from placement of the order to the availability at the final point-of-sale location). When a replenishment plan is ready, it can be communicated to suppliers and/or manufacturers (see operation 214).
[0090] As shown, the stock level calculator accesses inventory model parameters 113 (e.g., based on a selected inventory model and/or selected replenishment strategy), then combines the point-of-sale consumption data with the distribution-level order data to generate a time-phased replenishment plan for the identified items (e.g., when a particular quantity of a particular item or set of items should be delivered to a particular location).
Re-claim 41, Gopinath et al. teach the method of Claim 21, further comprising: placing one or more second orders for the materials to be used.
(see e.g. [0040] - As shown, an agent for materials in storage and or an agent handling materials that are in-transit may place orders (e.g., orders from transit 112) to the supplier to cover quantities damaged in storage or transit. Another case covers the situation where an agent or sub-manufacturer dealing with partially complete materials may place orders to a supplier (e.g., to cover defective materials).
The Examiner notes for partially complete materials at the manufacturer, an additional order is placed.
Claims 28 and 35 recite similar limitations as claim 21 and are therefore rejected under the same arts and rationale.
Claims 29 and 36 recite similar limitations as claim 22 and are therefore rejected under the same arts and rationale.
Claims 30 and 37 recite similar limitations as claim 23 and are therefore rejected under the same arts and rationale.
Claims 32 and 39 recite similar limitations as claim 25 and are therefore rejected under the same arts and rationale.
Claims 33 and 40 recite similar limitations as claim 26 and are therefore rejected under the same arts and rationale.
Claim 34 recites similar limitations as claim 27 and are therefore rejected under the same arts and rationale.
Claims 42 and 43 recite similar limitations as claim 41 and are therefore rejected under the same arts and rationale.
Claims 24, 31, 38 rejected under 35 U.S.C. 103 as being unpatentable over Gopinath et al. (20150379536A1), in view of Godfrey et al. (US 20200279192 A1), in view of Ohlsson et al. (US 20200143313 A1), in further view of Renz et al. (US2003/0093307 A1).
Re-claim 24, Gopinath et al., in view of Godfrey et al., in view of Ohlsson et al., do not teach the limitation as claimed.
However, Renz et al. teach a method of Claim 21, wherein the regression of the orders is based on a defined asymmetrical loss function. (see e.g. paragraphs [0012], [0083]).
[0067] The agent estimates a set of functions f.sub.t+.quadrature.t(I.sub.- t.sup.r), .quadrature.t.epsilon.{1, . . . , T}, each of which minimizes an appropriately selected distance function d between an approximation and the actual order: d(f.sub.t+1(I.sub.t.sup.r)-o.sub.t+1). The agent model assumes that the planned/forecasted orders already in the system can be updated (quantity changed, orders cancelled, etc.) when new information becomes available and that there is a causal relationship between the orders in the near-term future and current point-of-sales data.
[0083]--a classification and regression decision tree algorithm. This algorithm check is particularly useful in order to account for certain variability, such as, for example, seasonal or daily variations in consumer demand.)
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Gopinath et al., in view of Godfrey et al., in view of Ohlsson et al., and include the steps cited above, as taught by Renz et al., in order to account for variables in inventory demand and orders (see e.g. paragraphs [0080], [0083])
Claims 31 and 38 recite similar limitations as claim 24 and are therefore rejected under the same arts and rationale.
Response to Arguments
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. --Deshpande et al. (US 10832194 B2)
Applicant’s arguments with respect to claims 21-40 have been considered but are not persuasive.
Applicant’s argument:
Applicant respectfully submits that causing a machine learning algorithm to specifically use a "classification threshold and prediction mask" to determine "one or more lengths of time that the one or more orders are likely to change based on regression of the orders" is a process that could not be considered as generic computer technology performing a generic computer function, as the machine learning algorithm uses specific determined parameters (e.g., "classification threshold and prediction mask") to generate outputs, which is not a generic computer function.
Therefore, for at least the foregoing reasons, Applicant respectfully submits that claims 21-43 are allowable under 35 U.S.C. § 101.
Examiner’s response:
The Examiner acknowledges Applicant’s arguments, but disagrees. Using any type of parameters as input to an algorithm to generate outputs is not a technological improvement, and is a generic function. Generally, Machine Learning models are trained to perform specific tasks such as making predictions. The addition of specific parameters as input to a model does not alter the abstract nature of the claims and does not add an inventive component that renders the claims patentable.
Applicant’s argument:
Applicant respectfully submits that the cited references fail to disclose at least "causing the first machine learning algorithm to determine ... one or more lengths of time that the one or more orders are likely to change based on regression of the orders; [and] generating a schedule to acquire materials to be used to manufacture or produce the one or more items based on the one or more lengths of time that the one or more orders are likely to change."
Examiner’s response:
Gopinath et al., in view of Ohlsson et al. teach the limitations as described above.
Conclusion
THIS ACTION IS MADE FINAL. 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 LUNA CHAMPAGNE whose telephone number is (571)272-7177. The examiner can normally be reached M-F 8:00-5:00.
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, Florian Zeender can be reached on 571 272-6790. 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.
/LUNA CHAMPAGNE/Primary Examiner, Art Unit 3627 November 5, 2025