DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 1-5, 7-12, 14-19 have been presented for examination based on the amendment filed on 4/27/2026.
Claims 6, 13 and 20 were previously cancelled.
Claims 1, 8 and 15 are currently amended.
Claims 1-5, 7-12, 14-19 are newly rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement.
Claims 1-5, 7-12, 14-19 remain rejected under 35 U.S.C. 101.
Claim(s) 1-5, 7-12, 14-19 are newly rejected under 35 U.S.C. 103 as being unpatentable over US 20200362674 A1 by Alanazi; Amer et al., in view of US 20210133375 A1 by ZAGAYEVSKIY; Yevgeniy et al., further in view of US 20180347326 A1 A1 Shammari; Ahmad T. et al.
This action is made Final.
---- This page is left blank after this line ----
Response to Arguments
(Argument 1) Applicant has argued in Remarks Pg.7:
PNG
media_image1.png
370
656
media_image1.png
Greyscale
(Response 1) The data gathering aspect may be real time, but the computation, even if done in real time is still mathematical concept (reducing the productivity index (an abstract number based on data)). Reducing the productivity index (PI) for each of the laterals by a same percentage (Emphasis is added that PI is an abstract number in a model), whether this is done real time or otherwise is not still mathematical concept. The claims are updated likewise.
Secondly, updating the physics-based model (for updating the PI) and affecting production (e.g. adjusting an inflow control valve setting) cannot be done with the physics-based model according to applicant’s own specification.
PNG
media_image2.png
136
564
media_image2.png
Greyscale
(Argument 2) Applicant has argued in Remarks Pg.8:
PNG
media_image3.png
440
632
media_image3.png
Greyscale
(Response 2) As for the arguments regarding rejection made under 35 USC 103, the newly amended limitation is not taught by the combination of Alanazi in view of Zagayevskiy. Updated rejection is presented below based on updated search.
---- This page is left blank after this line ----
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-5, 7-12, 14-19 are newly rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention. Specifically exemplary claim 1 now recites:
[Claim 1] ….
inputting the production scenarios into a physics-based model of the multilateral wells, wherein the physics-based model is built using one or more well tests where a productivity index is estimated for laterals of the multilateral wells [[by]] through iteratively altering the productivity index in real-time by reducing an intermediate productivity index for each lateral by a same percentage and averaging the reduced intermediate productivity indices until an individual lateral flowrate is matched based on a known reservoir pressure;….
Notice should be made here that while specification and the invention is directed to using the hybrid machine learning based model to predict ICV valve position in real time, the context here is that the physics-based model is performing the calculation is real time. This not only does not have support in the specification (new matter), the specification explicitly recites that the physics-based model cannot be used to perform real time computations (like of PI leading to real time valve control). See specification [0039]:
PNG
media_image2.png
136
564
media_image2.png
Greyscale
Claims 8 and 15 recite similar limitation as claim 1 above and are rejected for new matter for the same reasons. Respective dependent claims are rejected for inheriting the above deficiency.
---- This page is left blank after this line ----
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-5, 7-12, 14-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to mental process without any additional elements that provide a practical application or amount to significantly more than the abstract idea.
Claims 1, 8 and 15:
Step 1: the claims 1, 8 and 15 are drawn to a method, a system and an An apparatus comprising a non-transitory, computer readable, storage medium respectively, falling under one of the four statutory categories of invention.
Step 2A, Prong 1: Taking claim 1 as representative, however analysis is applicable to claim 8 & 15 as well. The claim 1 limitations recite (bolded for abstract idea identification):
Claim 1
Mapping Under Step 2A Prong 1
1. A computer-implemented method, comprising:
obtaining, data associated with well completion, data associated with inflow control valves, and reservoir attributes of multilateral wells;
generating, production scenarios based on the data associated with well completion, the data associated with inflow control valves, and the reservoir attributes of the multilateral wells;
inputting, the production scenarios into a physics-based model of the multilateral wells, wherein the physics- based model is built using one or more well tests, where a productivity index is estimated for laterals of the multilateral wells by iteratively altering the productivity index in real-time by reducing an intermediate productivity index for each lateral by a same percentage and averaging the reduced intermediate productivity indices until an individual lateral flowrate is matched based on a known reservoir pressure;
obtaining, simulation data associated with the multilateral wells output from the physics-based model, the simulation data comprising, for each scenario, per-lateral and total flowrates labeled to corresponding per-lateral inflow control valve settings and reservoir attribute;
predicting multilateral well production parameters using a machine learning model trained using the simulation data associated with the multilateral wells and target parameters associated with the multilateral wells; and
adjusting an inflow control valve setting associated with a multilateral well of the multilateral wells to achieve specific production flow rates from the multilateral well based on the predicted multilateral well production parameters.
See Step 2A Prong 2 and 2B.
Abstract Idea/Mental Step: The generation of production scenario appears to judgement/ opinion based on observation1 (viz the metadata). See MPEP 2106.04(a)(2)(III).
Abstract Idea/Mathematical concept/Mental Step: The inputting the production scenario into physics based model is a mental step which simply implies running the model (forming opinion) with data (observation of meta data and physics based model). See MPEP 2106.04(a)(3). The physics-based model it self if mathematical concept requiring executing of equation related physics2 of the well/reservoir. See MPEP 2106.04(a)(I)(B) & (C).
The inputting of data can also be considered under Step 2A Prong 2.
The estimation of productivity index is considered abstract idea (mental step based on possible mathematical concept) to form an opinion (adhoc matching decision based on known reservoir pressure) based on observation (estimated productivity index). Altering the productivity index (an abstract number) in the model by same percentage in real time is still performing mathematical calculation/concept related to abstract concept of physics-based modeling. The altering does not alter any actual adjusting in real time based on productivity index.
This may also be treated as an idea of solution under Step 2A Prong 2.
See Step 2A Prong 2 and 2B.
Abstract Idea/Mathematical concept: predicting parameters using machine learning model at high level of generality with no details of what is predicted, how it is predicted and how the model is trained for the specific application (multilateral well production parameters) is abstract idea at best. See MPEP 2106.04(a)(2)(I)(C). Also see Example 47 Claim 2 as generic training and outputting the parameter is patent ineligible. The training process3 involves evaluating data to determine best hyperparameters (weights). The plain meaning is that random search algorithms, which compute neural network hyperparameters are a series of mathematical calculations.
See Step 2A Prong 2 and 2B.
Under its broadest reasonable interpretation, these covers a mental process including an observation, evaluation, judgment or opinion that could be performed in the human mind or with the aid of pencil and paper. That is, nothing in the claim element precludes the step from practically being performed in the mind or with the aid of pencil and paper but for the recitation of generic computer components (Processor). Also the mathematical concepts disclosed may also be performed in the mind or with the aid of pencil and paper.
Claims 8 & 15 recites similar bolded limitations as above and those are rejected likewise.
Step 2A, Prong 2: In accordance with this step, the judicial exception is not integrated into a practical application.
In claim 1, all the steps recite additional element as at least one processor. The processor(s) is(are) recited at a high level of generality, i.e., as a generic computer performing generic computer functions. See MPEP 2106.05(f).
Claim 1
Mapping Under Step 2A Prong 2
1. A computer-implemented method, comprising:
obtaining, data associated with well completion, data associated with inflow control valves, and reservoir attributes of multilateral wells;
generating, production scenarios based on the data associated with well completion, the data associated with inflow control valves, and the reservoir attributes of the multilateral wells;
inputting, the production scenarios into a physics-based model of the multilateral wells, wherein the physics- based model is built using one or more well tests,
where a productivity index is estimated for laterals of the multilateral wells by iteratively altering the productivity index in real-time by reducing an intermediate productivity index for each lateral by a same percentage and averaging the reduced intermediate productivity indices until an individual lateral flowrate is matched based on a known reservoir pressure;
obtaining, simulation data associated with the multilateral wells output from the physics-based model, the simulation data comprising, for each scenario, per-lateral and total flowrates labeled to corresponding per-lateral inflow control valve settings and reservoir attribute;
predicting multilateral well production parameters using a machine learning model trained using the simulation data associated with the multilateral wells and target parameters associated with the multilateral wells; and
adjusting an inflow control valve setting associated with a multilateral well of the multilateral wells to achieve specific production flow rates from the multilateral well based on the predicted multilateral well production parameters.
As per MPEP 2106.05(g) the obtaining various data as claimed is considered as data gathering/extra solution activity.
As per MPEP 2106.05(f) use of generic computer components where the processor invokes computers or other machinery merely as a tool to perform an existing process. The limitation related to processor is removed from claim 1, however is implied in the claims 8 and 15 is mapped as generic component.
As per MPEP 2106.05(g) the inputting various data as claimed is considered as data gathering/extra solution activity. Use of processor is considered as above under MPEP 2106.05(f). The limitation related to processor is removed from claim 1, however is implied in the claims 8 and 15 is mapped as generic component.
Under MPEP 2106.05(f)(1) the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished.
Here the solution (matching based on the known reservoir pressure) does not provide steps taken iteratively to alter the productivity index (for each lateral) to cause the matching to happen, thereby failing to show details how the result is accomplished. Further productivity index (PI) is an output of the model. It is unclear how the output now is reduced to create matching, or in another words how & what is done to PI to reduce the PI. Alternately, PIcomingled = Average of (PI laterals), to match the PI comingled to actual PI, what is done with individual PI is not claimed and how does it affect actual values is not claimed. Also unclaimed is the relationship between the lateral PI and control valves.
As per MPEP 2106.05(g) the obtaining various data as claimed is considered as data gathering/extra solution activity. Use of processor is considered as above under MPEP 2106.05(f). Under MPEP 2106.05(f)(1) the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished. , where there is specific solution (what are the “multilateral well production parameters”?) is disclosed and no mechanism other than simply stating that is obtained using trained machine learning model. The input parameters (“simulation data associated with the multilateral wells and target parameters associated with the multilateral wells”) does not show the mechanism that would be required to train the model nor what and how the output of multilateral well production parameters are generated. Providing the details of the simulation data (as to what it is) is not sufficient show the mechanism to provide details of the physics-based model.
Further under MPEP 2106.05(h) use of machine learning model generically to output production parameters with no details how the neural network is trained and how the production parameters are derived from the trained neural network lead to simply the use of neural network in the field of predict multilateral well production parameters
Under MPEP 2106.05(g) predicting is considered as an output of the neural network and broad recitation of Production parameters as output is considered as post solution/extra-solution activity.
Under MPEP 2106.05(g)/(h), this is post solution activity and field of use.
Examiner Note: There is no nexus claimed/established between the “multilateral well production parameters” and the “inflow control valve (ICV) settings”, therefore this remains field of use as indicated above. This may further be idea of solution because it is not clear what the multilateral well production parameters are and how the ICV setting are computed to adjust the ICV. This would be beneficial to overcome the rejection under this statute.
In particular, the claim(s) recites the additional elements of a processor for the system claim, at a high-level of generality (i.e. a generic processor performing generic functions of computing and executing information such that it amounts to no more than mere instructions to apply the exception using a generic computer component). Accordingly, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.05(f).
Additionally, Claim 8 recites additional elements of “at least one processor, and at least one non-transitory storage media storing instructions that, when executed by the at least one processor”, which are recited at a high level of generality, i.e., as a generic computer performing generic computer functions. As per MPEP 2106.05(f) they do not integrate the judicial exception into practical application.
Additionally, Claim 15 recites additional elements (underlined) of “apparatus comprising a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations”, which are recited at a high level of generality, i.e., as a generic computer performing generic computer functions. As per MPEP 2106.05(f) they do not integrate the judicial exception into practical application.
Step 2B: As discussed above the claim fails to integrate of the abstract idea into a practical application and it for same reasons, the additional element of using a computer/processor to perform the claimed steps amounts to no more than mere instructions to apply the exception using a generic computer/processing component, and do not contribute significantly more to the judicial exception, as mapped above. The claims 1, 8 & 15 are therefore considered to be patent ineligible.
Claims 2, 9 & 16 recite “wherein multilateral well production parameters comprise inflow control valve settings corresponding to laterals of the respective multilateral well” further defining the what data is gathered. This type of limitation merely confines the use of the abstract idea to a particular technological environment (type of data gathered using extrasolution activity) and thus fails to add an inventive concept to the claims. MPEP 2106.05(g) & (h).
Claims 3, 10 & 17 recite “ dividing the simulation data into training and validation datasets; and training the machine learning model using a randomized search algorithm applied to the simulation data associated with the multilateral wells output from the physics-based model” , which are considered as abstract idea based on mental step (dividing limitation) and performing mathematical calculations (training step). The claims do not disclose any additional limitations that integrate the judicial exception into practical element or add significantly more to the judicial exception.
Claims 4, 11 & 18 recite “ comprising validating the trained machine learning model by evaluating R-squared values and mean absolute error (MAE) values.” , which are considered as abstract idea based on performing mathematical calculations (validating by performing mathematical calculation to determine R-squared and MAE steps). The claims do not disclose any additional limitations that integrate the judicial exception into practical element or add significantly more to the judicial exception.
Claims 5, 12 & 19 recite “ wherein the target parameters are defined by a predetermined well type” , which are considered as abstract idea based on mental step of picking the target parameters based on well type (See specification [0028] ). The claims do not disclose any additional limitations that integrate the judicial exception into practical element or add significantly more to the judicial exception.
Claims 6, 13 & 20 (Cancelled)
Claims 7 & 14 recite “ wherein the production scenarios are generated as permutations of the data associated with well completion, the data associated with inflow control valves, and the reservoir attributes of the multilateral wells..” , which are considered as abstract idea based on mental step which is based observation of datum as claimed (See specification [0022]-[0026] ). The claims do not disclose any additional limitations that integrate the judicial exception into practical element or add significantly more to the judicial exception.
----- This page is left blank after this line -----
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 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-5, 7-12, 14-19 are rejected under 35 U.S.C. 103 as being unpatentable over US 20200362674 A1 by Alanazi; Amer et al., in view of US 20210133375 A1 by ZAGAYEVSKIY; Yevgeniy et al., further in view of US 20180347326 A1 A1 Shammari; Ahmad T. et al.
Regarding Claims 1, 8 and 15 (Updated 7/1/2026)
Alanazi teaches (Claim 1) A computer-implemented (Alanazi: Fig.31, Fig.10 & [0094]) method, comprising/ (Claim 8) A system, comprising: at least one processor, and at least one non-transitory storage media storing instructions that, when executed by the at least one processor (Alanazi: Fig.31) , cause the at least one processor to/ (Claim 15) An apparatus comprising a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations (Alanazi: Fig.31, Fig.10 & [0094]) comprising:
obtaining data associated with well completion (Alanazi: smart well completions (SWCs) data gathering for modeling Fig.4 element 402 [0041]-[0042] data collection, [0061]-[0062] [0048]-[0049], [0094] with Fig.10) , data associated with inflow control valves (Alanazi: as inflow control devices (ICVs) data gathered for model initialization Fig.4 [0041]-[0042] data collection ) , and reservoir attributes of multilateral wells (Alanazi: such as reservoir fluid properties (PVT), permeability (K) Fig.10 with [0094]) ;
generating production scenarios based on the data associated with well completion, the data associated with inflow control valves, and the reservoir attributes of the multilateral wells (Alanazi: [0042] "... The acquired data was used to calibrate the model. After calibration, an optimization algorithm (in this case, a genetic algorithm (GA)) was used to generate different production scenarios and optimize the performance of each lateral...."; Fig.4B & [0067]"... At 452, comingled production scenarios are generated using an optimization algorithm, for example, the genetic algorithm optimization procedure 300....");
inputting the production scenarios into a physics-based model of the multilateral wells (Alanazi: [0067]"... At 454, laterals are set up and tested under a comingled production scenario....") , wherein the physics- based model is built using one or more well tests (Alanazi: [0046] "... [0046] The model can either calculate the PI based on one of many conventional techniques using reservoir description parameters, or the [physics based] model can correlate the value based on well test results...."; the model is physics based model because at least it uses equations to compute well productivity index (PI) as in [0047];[0062] showing various physics based models; [0097]-[0098]) where a productivity index is estimated for laterals (Alanazi: Fig.10 shows the actual laterals (not the model as alleged) modeling aspect for which is shown in Fig.3 and 4A-4B; Specifically Figs.3 shows "[0054]... an example of a genetic algorithm optimization procedure 300 for determining optimum valve settings as given by the well model...", 4A shows "[0061]... FIG. 4A, the process 400 includes a model initialization and equipment assessment phase 402 (including steps 408-416) and a single lateral calibration phase 404 (including steps 418-428). A comingled production calibration process phase 406 follows the single lateral calibration phase 404. ..."4B shows the shows the comingled aspect modeling) of the multilateral wells (Alanazi : Fig.10 actual multilateral well; Fig.4A-4B modeling for each lateral individually (Fig.4A section 404) and comingled (Fig.4B section 406)) through iteratively altering the productivity index [presumably inferred as for each lateral, from the arguments pg. 9 arguing against comingled model calibration] until an individual lateral flowrate is matched (Alanazi : Fig.4A element 422, 424, 426 iteratively assesses and alters PI/pressure for each of the lateral (annotated by “Single Lateral Calibration 404”) where the calibration is done against field test PDHMs (pressure) and model output, where if the match is not found both the productivity index (PI) and modeled reservoir pressure (in Fig.4A step 426) are reassessed iteratively to be within a threshold (in step 424); This process is individually repeated for each of the laterals (in steps 428 & 420 onwards loop); [0062] also states this regarding matching the flow "... Modeling SWC will mimic downhole conditions and help to predict flow contributions from each lateral by estimating downhole parameters, such as PI, FBHP, oil rate, and water cut. The model can be constructed using surface network modeling software, which uses nodal-network analysis of a steady-state system. Surface network modeling software can include, for example, General Allocation Program (GAP) from Petroleum Experts (Petex), and Pipeline Simulation (PIPESIM) from Schlumberger (SLB)....") based on a known reservoir pressure (Alanazi: [0062] "... [0062] The process 400 can include a workflow beginning at the field, continuing through the model optimization procedure, and ending by applying a model solution to the field. The SWC model starts from permanent downhole measurement (PDHMs) point as a top node and down to lateral total depth. ..."[0068] "... Approaches using the process 400 can benefit from well-maintained surface and subsurface equipment, such as PDHMs and multi-phase flow meters (MPFMs), which provide accurate measurements of well rates and downhole pressure [known reservoir pressure]...."; here the mapping is made show the field test PDHMs (Fig.4A, step 422) include the known reservoir pressure);
obtaining simulation data associated with the multilateral wells output from the physics-based model (Alanazi: [0064]), the simulation data comprising, for each scenario, per-lateral and total flowrates labeled (Alanazi : [0064] Fig.4A model construction 416 using input from step 414 to include per lateral PIs and SWC (SWC has inflow control data information ([0093] "... A typical SWC includes at least one feed-through isolation packer 1002, ICVs 1004, and a permanent downhole monitoring system (PDHMS) 1006....") ) to corresponding per-lateral inflow control valve settings and reservoir attribute (Alanazi : Fig.4A elements 414, 416 & [0093 as above; [0064] "... SWC data is reviewed and the laterals' PIs are estimated. At 416, a preliminary network model is constructed, for example, using well static data and reservoir and fluid properties. The preliminary network model, for example, can be part of an approach which starts first by building a preliminary model (not yet calibrated) using the well's previous production and pressure data. Then, the model is calibrated. After calibration, the model is capable of providing recommended downhole choke valve settings guided by user objectives....");
predicting multilateral well production parameters using a machine learning model (Alanazi: [0068]-[0069]) trained (Alanazi : training as calibrating the model [0051] "... Optimization algorithms (for example, genetic algorithms) can be implemented as stochastic methods to assure repeatability. The algorithms can be run multiple times, with the number of iterations used being sufficient for the size of the problem. For example, the number of iterations used can be based on rules of thumb correlating the number of iterations with a number of variables (for example, ICVs). The optimized solution can reflect the current state of the calibrated model and can be implemented immediately to begin realizing production gains. After a production interval has occurred in which well/reservoir conditions are expected (or likely) to change, the calibration and optimization of the model can be repeated to reflect these changes. In some implementations, automated processes can be used to apply the recommendations to the field, for example, as part of a closed-loop automated optimization....") using the simulation data associated with the multilateral wells and target parameters associated with the multilateral wells (Alanazi: Fig.4C [0070]-[0087] use of various machine learning and genetic algorithmic techniques; e.g. [0078] "... For the well control optimization problem, for example, x.sub.opt contains the ICV setting(s) of all valves in the smart wells. Objective functions in smart multilateral well completions can include, for example, cumulative oil production, cumulative oil flow rate, and economic implications, which can consider a net present value (NPV) of a project (for example, the multilateral well)...."); See Fig.5-7, 10 and 12 showing how the genetic algorithm creates simulation data which is used in training/calibrating and predicting ICV values); and
adjusting an inflow control valve setting associated with a multilateral well of the multilateral wells to achieve specific production flow rates from the multilateral well based on the predicted multilateral well production parameters (Alanazi: Fig.4 B element 462 and 464; [0061]-[0077] flow detailing how the optimizing changes to downhole inflow control valve (ICV) settings for surface ICVs and subsurface ICVs in one or more laterals of the multilateral well is made based on the predictions) .
Alanazi does not explicitly disclose the trained in the context to train the neural network and in real-time by reducing an intermediate productivity index for each lateral by a same percentage and averaging the reduced intermediate productivity indices(Emphasis on bolded).
Zagayevskiy explicitly disclose predicting multilateral well production parameters using a machine learning model trained using the simulation data associated with the multilateral wells and target parameters associated with the multilateral wells (Zagayevskiy: Fig.2A elelements [0014] "... a reduced parameter size are generated based on statistical and machine learning (deep neural network) approximation of a full physics flow simulator, i.e. the processing of a full physics flow algorithmic model...."); [0025] "... The plurality of reservoir data models 72 is previously generated data models generated by the full-physics flow simulator 64. The machine learning and training and validation module 74 can iteratively generate reduced parameter space algorithmic models from the computationally complex algorithmic model using at least one training dataset and at least one validation dataset, i.e. reservoir models, and at least one selected from a group of: input variables 56, output variables 58, updated reservoir model 76, history matching input variables 78, the optimized reservoir model 84, and optimization input variables 86....";).
Zagayevskiy does not explicitly teach in real-time by reducing an intermediate productivity index for each lateral by a same percentage and averaging the reduced intermediate productivity indices(Emphasis on bolded).
Shammari teaches in real-time by reducing an intermediate productivity index for each lateral by a same percentage and averaging the reduced intermediate productivity indices (Shammari: [0024]
PNG
media_image4.png
140
560
media_image4.png
Greyscale
Also see [0035] [0038])
It would have been obvious to one (e.g. a designer) of ordinary skill in the art before the effective filing date of the claimed invention to apply the teachings of Zagayevskiy to Alanazi reduce the parameter space to allow reservoir management teams to quickly make informed and effective decisions, there is a need for systems that can generate less computationally complex workflows yet still maintain an acceptable degree of accuracy by using machine learning in optimizing the reservoir management (Zagayevskiy: Abstract & [0002]; Alanazi: Abstract). The motivation to combine would have been that Zagayevskiy & Alanazi are analogous art to the instant claims in the field of reservoir management/completion to optimize the choke valve (ICV) for such management using machine learning (Zagayevskiy: [0014][0025] – showing machine learning using physics based model, [0032][0028] & Alanazi: Abstract [0070]-[0087] showing machine learning, [0041]-[0042]).
It would have been obvious to one (e.g. a designer) of ordinary skill in the art before the effective filing date of the claimed invention to apply the teachings of Shammari to Zagayevskiy & Alanazi to compute productivity index and "... based on the final productivity index for each lateral, wellhead pressures (WHP), inline control valves (ICVs), or both may be adjusted to achieve a desired productivity from the multilateral completion (block 110). For example, an engineer at a well site may adjust the wellhead pressure at a wellhead of the multilateral well and may adjust the setting (e.g., between closed and 100% open) of one or more inline control valves to achieve a desired production rate (that is, a certain volume of produced fluid per time). ..."( Shammari: [0025]). Further motivation to combine would have been that Shammari, Zagayevskiy & Alanazi are analogous art to the instant claims in the field of reservoir management/completion to optimize the choke valve (ICV) for such management using machine learning (Shammari: [0025], Zagayevskiy: [0014][0025] – showing machine learning using physics based model, [0032][0028] & Alanazi: Abstract [0070]-[0087] showing machine learning, [0041]-[0042]).
Regarding Claims 2, 9 and 16
Alanazi teaches wherein multilateral well production parameters comprise inflow control valve settings corresponding to laterals of the respective multilateral wells (Alanazi: [0068]-[0077]; See Fig.4B element 462-464 and Fig.10 showing inlet control valves (ICVs) for laterals 1012 of the respective multilateral wells being controlled) .
Regarding Claims 3, 10 and 17
Zagayevskiy teaches comprising: dividing the simulation data into training and validation datasets (Zagayevskiy: [0025] ) ; and training the machine learning model using a randomized search algorithm applied to the simulation data associated with the multilateral wells output from the physics-based model (Zagayevskiy: [0032] a plurality of trained algorithmic models from the physics based model as disclosed in [0025]) .
Regarding Claims 4, 11 and 18
Zagayevskiy teaches comprising validating the trained machine learning model by evaluating R-squared values and mean absolute error (MAE) values (Zagayevskiy: [0035] as mean square error MSE) .
Regarding Claims 5, 12 and 19
Alanazi teaches wherein the target parameters are defined by a predetermined well type (Alanazi: [0049] showing gas, oil and water zones for multilateral wells; [0106]-[0108] showing restricting the well-type production for gas and oil) .
Regarding Claims 6, 13 and 20 (Cancelled)
Regarding Claim 7
Alanazi teaches wherein the production scenarios are generated as permutations of the data associated with well completion, the data associated with inflow control valves, and the reservoir attributes of the multilateral wells (Alanazi: [0042][0067][0069], [0079]-[0087] use of genetic algorithm to generate scenarios population; [0094] based on collected data associated with well completion, ICV specification as data associated with inflow control valve; PVT & Permeability as reservoir attributes) .
---- This page is left blank after this line ----
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.
---- This page is left blank after this line ----
Communication
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AKASH SAXENA whose telephone number is (571)272-8351. The examiner can normally be reached Mon-Fri, 7AM-3:30PM.
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, RYAN PITARO can be reached on (571) 272-4071. 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.
AKASH SAXENA
Primary Examiner
Art Unit 2188
/AKASH SAXENA/Primary Examiner, Art Unit 2188 Thursday, July 2, 2026
1 See Specification [0022] "...For example, metadata includes a well's completion details, ICV details, reservoir attributes and the like.... In embodiments, this metadata dictates the type of production scenarios that can be generated." [0023]-[0026] offer no explanation of any specific methodology that would render generation of production scenario to not be mental step based on observation of metadata.
2 See Specification [0015] "...Generally, physics-based models are defined by one or more governing physics model equations to incorporate process variations using independent process parameters...."
3 See Specification [0032] "... Generally, initial machine learning model hyperparameters are set by a data scientist ahead of training and control implementation aspects of the model. The hyperparameters are distinct from the model parameters (e.g., input and output parameters), which are learned during training. In examples, hyperparameter tuning enables determining the combination of hyperparameter values for a machine learning model that performs the best (as measured on a validation dataset) for a problem. In embodiments, the randomized search algorithm fine tunes one or more hyperparameters, such as neural network size, number of nodes in each layer, learning rate, activation function, solver, and max number of iterations...."
Prosecution Insights