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 .
2. The amendment filed on 03/26/2026 has been received and fully considered.
3. Claims 1-20 are presented for examination.
Response to Arguments
4. Applicant's arguments filed on 03/26/2026 have been fully considered but they are not persuasive. Regarding applicant’s assertions that: “This subject matter does not involve a mental process capable of being performed in the human mind. The claimed method requires generation of predictive models, conditioning one model on output of another, and generating a production profile that is utilized for extraction of the target well. Such operations cannot practically be performed mentally and instead involve structured processing of large-scale well production datasets and dynamic operational parameters. Accordingly, claim 1 does not recite any of the judicial exception groupings under Prong One of Step 2A. The Applicant respectfully submits that claim 1 satisfies Step 2A, Prong One” and that “Claim 1 recites, inter alia, a two-model architecture including a static model and a decline model, generation of the decline model using dynamic well data and output of the static model, and generation of a predicted well production profile which is utilized throughout an extraction process of the target well.
These elements do not merely use a computer as a tool to carry out calculations. Rather, the claim defines a structured modeling architecture that improves techniques for predicting time- evolving resource production behavior of oil wells and applies those predictions to determine operational development parameters.”, the Examiner respectfully disagrees and asserts that the claims are, as a whole, are clearly directed to an abstract idea “a combination of mental process and mathematical concept” and do not recite anything that goes beyond the judicial exception. The Examiner respectfully notes that the claims do not in any way provide any improvement to a technological field, as asserted by the Applicant nor integrate the recited abstract to practical application. In fact, there absolutely no way to improve the functionality of the general processor by the step set forth in the claims. While the claims are amended to recite a generating step of which a certain profile could be utilized to an extraction and thus could further amount to post-solution activities, it is noted by the examiner that to transform an abstract idea, law of nature or natural phenomenon into "a patent-eligible application", the claim must recite more than simply the judicial exception "while adding the words 'apply it.'" Mayo, 132 S. Ct. at 1294, 101 USPQ2d at 1965. As such, the claims are clearly abstract, as constructed. As per applicant’s assertions that: “Independent claim 1 recites, inter alia, "generating the predicted well production profile for the target well using the first final resource production rate and the second final resource production rate, wherein the predicted well production profile is utilized to control a production process for extraction of the target well." The Applicant respectfully submits that the asserted combination of Benhallam and Chung fails to teach at least these features.”, the Examiner respectfully disagrees and asserts that Benhallam does provide for the generated well production profile (see para [0085], The forecasting step (904) estimates the production potential of each target zone in each new drill location. It should be noted that the rate at which these production potentials are generates are different at each location. The predicted attribute is user-selected and can be either initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells. [0121] The forecasting module 112 of computer system 101 then forecasts, for each new drill location, a potential production rate for one or more target zones (1840), and estimates the probability of exceeding or falling short of the forecasted potential production rate using neighborhood characterization data 142 indicating an amount that wells in neighboring zones are producing (1850). The neighborhood production data 142 thus provides an indication of how much material other zones in a given area are producing. This data may be used to estimate whether a new drill location will likely fall short of or exceed a forecast for that location.), wherein the predicted well production profile is utilized to control a production process for extraction of the target well (see para [0114], Once a recompletion opportunity has been identified by the well target identifier 124, the computer system 101 sends a signal 128 to initiate production at that well (1770). Further commands may also be sent controlling how the oil is produced. Indeed, in at least some embodiments, controls from the computer system directly control equipment and/or processes at the hydrocarbon extraction site 130.); para [0015] of Chung further provides evaluating “well performance”. For example, well performance metrics may include initial rate (or decline in rate) of hydrocarbon production, final rate (or decline in rate) of hydrocarbon production, lifetime total hydrocarbon production, the net present value of the lifetime hydrocarbon production, or return on investment (ROI) from hydrocarbon production.. Therefore, the combination of the cited references clearly renders obvious the limitations, contrary to applicant’s assertions.
Claim Rejections - 35 USC § 101
5. 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.
5.1 Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 2A- Prong One
The claim(s) recite(s) a method for predictive decline modeling for a target well of a subterranean reservoir, comprising: The step of: “generating a static model based on an input data set including historical production data corresponding to one or more wells”; “generating a decline model wherein the decline model is generated using the historical production data, dynamic well data, and at least one output of the static model”; under the broadest reasonable interpretation fall under a mental process; likewise, the steps of: “generating a predicted well production profile for a target well by: “determining, using the static model and one or more well features of the target well, a predicted initial resource production rate for the target well”; “determining, using the decline model and the predicted initial resource production rate, a first final resource production rate for the target well at a first time interval”; and “determining, using the decline model and the first final resource production rate at the first time interval, a second final resource production rate at a second time interval subsequent to the first time interval”; generating the predicted well production profile for the target well using the first final resource production rate and the second final resource production rate”, under the broadest reasonable interpretation fall under a mental process or otherwise a mathematical concept. Therefore, the claims are directed to an abstract idea, by use of generic computer components and thus are clearly directed to an abstract idea, as constructed.
Step 2A Prong Two
This judicial exception is not integrated into a practical application because the additional limitation such as: “one or more tangible non-transitory media”, “computer-executable instructions”, “computer system”, “a predictive decline modeling system”, “at least one processor”, either alone or in combination, all serve to gather and process data and do not add anything more significantly to the judicial exception, but are mere instructions to apply the exception using a generic computer component that are well known, routine, and conventional activities (see specification at para [0059-0063], and fig.1, 6, 9) which can be of any type, including general-purpose computer (para [0062], The computer system 900 may be a conventional computer, a distributed computer, or any other type of computer, such as one or more external computers made available via a cloud computing architecture) previously known in the industries. Merely adding a programmable computer to perform generic computer functions does not automatically overcome an eligibility rejection. Alice, 573 U.S. at 223-24. Furthermore, the use of a general-purpose computer to apply an otherwise ineligible algorithm does not qualify as a particular machine. See Ultramerciallnc. v. Hulu, LLC, 772F.3d 709, 716-17 (Fed. Cir. 20l4); In re TLI Commc 'ns LLC v. AV Automotive, LLC, 823 F.3d 607, 613 (Fed. Cir. 2016) (mere recitation of concrete or tangible components is not an inventive concept); Eon Corp. IP Holdings LLC v. AT&T Mobility LLC, 785; and the additional limitation of: “wherein the predicted well production profile is utilized to control a production process for extraction of the target well” could amount to post-solution activities that are well-known, routine and conventional and are not sufficient to amount to significantly more than the judicial exception (See further MPEP 2106.05(d)(i-iv)-f); thus are not patent eligible under 35 USC 101.
Step 2B
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, as previously discussed above with reference to the integration of abstract idea into a practical application, the additional elements of: “one or more tangible non-transitory media”, “computer-executable instructions”, “computer system”, “a predictive decline modeling system”, either alone or in combination, all serve to gather and process data and do not add anything more significantly to the judicial exception, but are mere instructions to apply the exception using a generic computer component that are well known, routine, and conventional activities (see specification at para [0059-0063], and fig.1, 6, 9) which can be of any type, including general-purpose computer (para [0062], The computer system 900 may be a conventional computer, a distributed computer, or any other type of computer, such as one or more external computers made available via a cloud computing architecture) previously known in the industries. Merely adding a programmable computer to perform generic computer functions does not automatically overcome an eligibility rejection. Alice, 573 U.S. at 223-24. Furthermore, the use of a general-purpose computer to apply an otherwise ineligible algorithm does not qualify as a particular machine. See Ultramerciallnc. v. Hulu, LLC, 772F.3d 709, 716-17 (Fed. Cir. 20l4); In re TLI Commc 'ns LLC v. AV Automotive, LLC, 823 F.3d 607, 613 (Fed. Cir. 2016) (mere recitation of concrete or tangible components is not an inventive concept); Eon Corp. IP Holdings LLC v. AT&T Mobility LLC, 785; and the additional limitation of: “wherein the predicted well production profile is utilized to control a production process for extraction of the target well” could amount to post-solution activities that are well-known, routine and conventional activities and are not sufficient to amount to significantly more than the judicial exception (See further MPEP 2106.05(d)(i-iv)-f); thus are not patent eligible under 35 USC 101.Therefore, using computer components amount to no more than mere instructions to perform the abstract, and thus are not sufficient to amount to significantly more than the recited abstract, as constructed.
5.2 Dependent claims 2-12, 14-18, and 20 merely include limitations pertaining to further mathematical computations (claims 2, 14), “wherein generating the predicted well production profile includes recursive calculations generating resource production rates for a series of time intervals” (mathematical concept). (claims 3, 14); “wherein the static model is generated with supervised machine learning using the historical production data as feature inputs and a target variable being an initial resource production rate having a 30 days-averaged Initial Production (IP30) value” (mental process); (claim 4); “wherein the historical production data represents one or more of a geological feature, well completion parameters, reservoir properties, production data, injection data, and fluid data” (WURC data gathering and processing); (claims 5, 15); “wherein the decline model is generated with a neural network using the historical production data and the dynamic well data as feature inputs and a target variable being resource production rate at time (t)” (mental process or otherwise a mathematical concept); (claim 6) “wherein the neural network includes two to seven dense layers and between 100 and 600 neurons per layer” (mental process); (claim 7); “wherein the dynamic well data includes one or more of a resource production rate for a previous time interval, a bottom hole pressure at the target well, a shut-in bottom hole pressure at the target well, and an average draw down pressure at the target well” (WURC data gathering and processing); (claims 8, 17) “identifying a subset of data from the historical production data associated with a shut-in period of days; and removing the subset of data associated with the shut-in period of days from the historical production data) (mental process); (claim 9) “wherein the decline model has an elapsed days feature variable that is reset by an occurrence of an acid job or a recompletion at the target well” (mental process); (claims 10, 18) “wherein the input data set is generated from an initial data set filtered based on a well age or a type of well” (mental process); (claim 11) “wherein the first time interval or the second time interval is based on a user input indicating a desired length of time, or a comparison of different lengths of time affecting an absolute percentage error of the predicted well production profile” (mental process); “developing the target well based on the predicted well production profile” (WURC post-solution activities), all of which further amount to further mental process and/or mathematical concept similar to that already recited by the independent claims and already addressed above and thus are further not patent eligible under 35 USC 101.
Claim Rejections - 35 USC § 103
6. 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.
6.0 Claim(s) 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Benhallam et al. (USPG_PUB No. 2019/0325331), in view of Chung et al. (USPG_PUB No. 2021/0224669).
6.1 In considering claims 1, 13, and 19, Benhallam et al. teaches a method for predictive decline modeling for a target well of a subterranean reservoir, the method comprising:
generating a static model based on an input data set including historical production data corresponding to one or more wells (see fig. 1,9,17 para [0057], (historical completion data 141) corresponding to one or more wells (hydrocarbon wells 131) used as input to generate the static model. 'The platform implements inter-disciplinary field data including well surveys, well logs, well formation tops, completion and perforation data, production data, fluid contacts data, static geologic models, simulation models, and other types of data; [0096] “the used of machine learning techniques like supervised neural networks to obtain the static model“ Step 1205 forecasts production for each of the identified and validated targets using a range of different techniques. These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks, analytical models specifically designed for horizontal wells, and simulation models. [0098] Artificial Neural Networks (ANN's) are used to develop a model [decline model] that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. [0110], 'The geological map generator 108 may generate a geological map 110 based on this petrophysical log data 135. The data accessing module 107 may also access historical completion data 141 for the hydrocarbon extraction site 130 to identify uncontacted net pay intervals (1720)); generating a decline model wherein the decline model is generated using the historical production data, dynamic well data, and at least one output of the static model (Fig. 1, 9 & 17; (historical completion data 141) as well as dynamic well data (hydrocarbon wells 131, para [0057], In generating the disclose production decline model, The platform implements inter-disciplinary field data including well surveys, well logs, well formation tops, completion and perforation data, production data, fluid contacts data, static geologic models, simulation models, and other types of data'; accounting for the dynamic data used to generate a production decline model, [0071], In addition, the process of identifying these uncontacted pay intervals can be augmented with dynamic data sources; [0072], 'Another way to integrate dynamic data is through the use of simulation models. This is possible as long as the simulation model was fed as an input to the platform'; [0085], 'The forecasting step (904) estimates the production potential of each target zone in each new drill location. The predicted attribute is user-selected and can be either initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells'; [0096] “the used of machine learning techniques like supervised neural networks to obtain the static model“ Step 1205 forecasts production for each of the identified and validated targets using a range of different techniques. These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks, analytical models specifically designed for horizontal wells, and simulation models. [0098] Artificial Neural Networks (ANN's) are used to develop a model [decline model] that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. [0110], 'The geological map generator 108 may generate a geological map 110 based on this petrophysical log data 135. The data accessing module 107 may also access historical completion data 141 for the hydrocarbon extraction site 130 to identify uncontacted net pay intervals (1720). These uncontacted net pay intervals represent material 132 remaining in hydrocarbon wells 131 on the hydrocarbon site (1720)); and generating a predicted well production profile for a target well (see fig. 9, 17; para [0085], 'The forecasting step (904) estimates the production potential of each target zone in each new drill location. The predicted attribute is user-selected and can be either initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells) by: determining, using the static model and one or more well features of the target well, a predicted initial resource production rate for the target well (see fig. 9, 17; para [0085], 'The forecasting step (904) estimates the production potential of each target zone in each new drill location. The predicted attribute is user-selected and can be either initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells. The predicted attribute is user-selected and can be either initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells' [0096] “the used of machine learning techniques like supervised neural networks to obtain the static model“ Step 1205 forecasts production for each of the identified and validated targets using a range of different techniques. These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks, analytical models specifically designed for horizontal wells, and simulation models. [0098] Artificial Neural Networks (ANN's) are used to develop a model [decline model] that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals. [0112], 'Method 1700 next includes forecasting, using statistical neighborhood methods and one or more machine learning algorithms, projected production results for one or more recompletion opportunities at the hydrocarbon site, the projected production results including initial production estimates and/or ultimate recovery estimates (1740). The forecasting module 112 of computer system 101 may use statistical neighborhood methods and/or machine learning algorithms 113 to forecast projected production results 114. These projected results may indicate how much remaining material 132 is likely to be produced during an Initial term (i.e. initial production estimate 115) and over the life of the well (i.e. ultimate recovery estimate 116)); determining, using the decline model and the predicted initial resource production rate, a first final resource production rate for the target well at a first time interval (Fig. 9 & 17; [0067], For brownfields with complex stratigraphy, a machine-assisted stratigraphic correlation tool may be implemented that correlates user-provided formation tops across the field. This tool uses time series similarity assessment algorithms to find the optimal time alignment between two time series'; [0085], 'The forecasting step (904) estimates the production potential of each target zone in each new drill location. The predicted attribute is user-selected and can be either initial production {IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells'; The training set used to train the model include past completion intervals, and the training attributes are derived or calculated specifically over these intervals. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals; [0098] Artificial Neural Networks (ANN's) are used to develop a model [decline model] that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. The training set used to train the model include past completion intervals, and the training attributes are derived or calculated specifically over these intervals. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals. [0112], using statistical neighborhood methods and one or more machine learning algorithms, projected production results for one or more recompletion opportunities at the hydrocarbon site, the projected production results including initial production estimates and/or ultimate recovery estimates (1740). The forecasting module 112 of computer system 101 may use machine learning algorithms 113 to forecast projected production results 114. These projected results may indicate how much remaining material 132 is likely to be produced during an initial term (i.e. initial production estimate 115) and over the life of the well (i.e. ultimate recovery estimate 116).'; [0129], the forecasting module 112 may be used to forecast an initial production rate 115 for the selected horizontal well placement candidate .placed in the identified location on the hydrocarbon extraction site using analytical, simulation, or machine learning models (1980).'; an initial [first] production rate can be determined [first time interval], as well as a ultimate recovery estimate over the life of the well [second time interval]); and determining, using the decline model and the first final resource production rate at the first time interval, a second final resource production rate at a second time interval subsequent to the first time interval (Fig. 9 & 17; [0067], a machine-assisted stratigraphic correlation tool may be implemented that correlates user-provided formation tops across the field. This tool uses time series similarity assessment algorithms to find the optimal time alignment between two time series; [0085], 'The forecasting step (904) estimates the production potential of each target zone in each new drill location which indicated multiple time interval. Note that initial production (IP) is selected as the target attribute to predict and an EUR is estimated using decline parameters of active neighborhood wells'; [0097], forecast a production attribute (IP and/or EUR) for any type of identified opportunity. The predicted attribute is pre-defined and can be initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, the EUR is estimated using decline parameters of active neighborhood wells [0096]-[0098] the used of machine learning techniques like supervised neural networks to obtain the static model which in turned can be fed to the artificial neural network (ANN) to derive the decline model, The training set used to train the model include past completion intervals, and the training attributes are derived or calculated specifically over these intervals. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals. [0112], 'Method 1700 next includes forecasting, using statistical neighborhood methods and one or more machine learning algorithms, projected production results for one or more recompletion opportunities at the hydrocarbon site, the projected production results including initial production estimates and/or ultimate recovery estimates (1740). The forecasting module 112 of computer system 101 may use statistical neighborhood methods and/or machine learning algorithms 113 to forecast projected production results 114. These projected results may indicate how much remaining material 132 is likely to be produced during an initial term (i.e. initial production estimate 115) and over the life of the well (i.e. ultimate recovery estimate 116).'; [0128], [0129], 'Once the well candidates 121 have been identified, the forecasting module 112 may be used to forecast an initial production rate 115 and an initial production rate can be determined [first time interval]. as well as an ultimate recovery estimate [second production rate] over the life of the well [second time interval]); generating the predicted well production profile for the target well using the first final resource production rate and the second final resource production rate (see para [0085], The forecasting step (904) estimates the production potential of each target zone in each new drill location. The predicted attribute is user-selected and can be either initial production (IP) or estimated ultimate recovery (EUR). Note that if IP is selected as the target attribute to predict, an EUR is estimated using decline parameters of active neighborhood wells. [0121] The forecasting module 112 of computer system 101 then forecasts, for each new drill location, a potential production rate for one or more target zones (1840), and estimates the probability of exceeding or falling short of the forecasted potential production rate using neighborhood characterization data 142 indicating an amount that wells in neighboring zones are producing (1850). The neighborhood production data 142 thus provides an indication of how much material other zones in a given area are producing. This data may be used to estimate whether a new drill location will likely fall short of or exceed a forecast for that location.), wherein the predicted well production profile is utilized to control a production process for extraction of the target well (see para [0114], Once a recompletion opportunity has been identified by the well target identifier 124, the computer system 101 sends a signal 128 to initiate production at that well (1770). Further commands may also be sent controlling how the oil is produced. Indeed, in at least some embodiments, controls from the computer system directly control equipment and/or processes at the hydrocarbon extraction site 130.). While Benhallam et al. does not specifically state that the term of a decline model, he provides for using decline parameters as part of its model generation and the use Artificial Neural Networks (ANN's) to develop a model that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes, also at para [0057], The platform implements inter-disciplinary field data including well surveys, well logs, well formation tops, completion and perforation data, production data, fluid contacts data, static geologic models, simulation models, and other types of data; and thus would have been obvious to a person of skilled in the art.
Nevertheless, Chung et al. teaches a method of which a decline model is generated applying a production prediction algorithm to one or more parameters for a hydrocarbon well to generate a production decline curve for the hydrocarbon well (see para [0036] The production prediction engine 106 is configured to generate a production decline curve (e.g., including predicted production data) in response to the cleaned historical data received from the data preparation engine 104. In one example, the production prediction engine 106 generates the production decline curve by fitting an Arps equation (e.g., an Arps hyperbolic to exponential decline curve) to historical production data. The result of applying the Arps equation is the predicted production decline curve, which forecasts oil and/or gas production volumes going forward from a particular start date, based on historical data from prior to the start date. See further [0040-0041]).
Benhallam et al. and Chung et al. are analogous art because they are from the same field of endeavor and that the model analyzes by Chung et al. is similar to that of Benhallam et al. Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.2 Regarding claims 2 and 14, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein generating the predicted well production profile includes recursive calculations generating resource production rates for a series of time intervals (see Benhallam et al. para [0067]. This tool uses time series similarity assessment algorithms to find the optimal time alignment between two time series'; [0095], This tracking is performed using an algorithm that recursively tracks pay cells in multiple user-specified directions. [0121], forecasting module 112 of computer system 101 then forecasts, for each new drill location, a potential production rate for one or more target zones (1840), and estimates the probability of exceeding or falling short of the forecasted potential production rate using neighborhood characterization data 142 indicating an amount that wells in neighboring zones are producing (1850). See Chung et al. 0032] The data preparation engine 104 is also configured to resolve discrepancies regarding the initial production date for a given well (e.g., mismatching other publicly available data). In some examples, the data preparation engine 104 is configured to calculate ancillary values related to the well parameters 102, such as average daily production rates per month (e.g., monthly volume divided by days on production) for oil, gas, and/or water for a given well; or a moving average (e.g., a 3-month moving average) of water/gas and/or water/oil ratios for a given well. [0075]). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.3 As per claims 3 and 15, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the static model is generated with supervised machine learning using the historical production data as feature inputs (see para [0067], a machine-assisted stratigraphic correlation tool may be implemented that correlates user-provided formation tops across the field. This tool uses time series similarity assessment algorithms to find the optimal time alignment between two time series'; [0076), 'The production forecasting phase of the workflow forecasts production for missed net pay opportunities. [0096), 'Step 1205 forecasts production for each of the identified and validated targets using a range of different techniques. These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks, analytical models specifically designed for horizontal wells, and simulation models'; These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks, analytical models specifically designed for horizontal wells, and simulation models), and a target variable being an initial resource production rate having a 30 days-averaged Initial Production (IP30) value (see Chung et al. para [0031] For example, the data preparation engine 104 is configured to resolve monthly reported negative days on production, reported days on production greater than the number of days in the month, and/or missing or otherwise unavailable days on production. For example, negative days on production or greater than 31 days on production for a given month is classified as erroneous, and it may be assumed that such months include approximately 30.4375 days (e.g., an average number of days per month). The data preparation engine 104 is also configured to classify certain data as erroneous, for example production data that indicates a negative volume of oil, gas, and/or water for a given month. Such data is considered erroneous (e.g., not possible) and it may be assumed that such negative production volumes are 0 for the given month.). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.4 With regards to claim 4, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the historical production data represents one or more of a geological feature, well completion parameters, reservoir properties, production data, injection data, and fluid data (see Chung et al. para The well parameters 102 may include well location (e.g., wellhead latitude and longitude, average horizontal section depth), borehole parameters (e.g., geometry), and production parameters (e.g., historical production data) for one or more hydrocarbon wells. In some examples, the well parameters 102 include data sufficient to generate a predicted decline curve for a given well. In particular, the well parameters 102 may also include historical monthly production volume(s) for oil, gas, and/or water; days on production for the well; true vertical depth (TVD); and lateral length. [0024] Examples of this disclosure include predicting or forecasting future hydrocarbon production for a given well or group of wells based on historical production data for the well or group of wells. Other examples of this disclosure include predicting or forecasting future hydrocarbon production additionally or alternately based on well spacing (or the distance between a well being analyzed and other wells already analyzed), completion data for the well being analyzed (or other wells in the vicinity of the well being analyzed), and/or other parameters or data for the well being analyzed. See further Benhallam et al. para [0005] Embodiments described herein are directed to identifying and implementing hydrocarbon production opportunities including recompletion opportunities, new vertical drill target opportunities, and horizontal or deviated well target opportunities. In one embodiment, a computer system accesses petrophysical log data obtained at a hydrocarbon extraction site to generate a geological map of the site. The computer system also accesses historical completion data for the hydrocarbon extraction site to identify uncontacted net pay intervals that represent material remaining in hydrocarbon wells on the hydrocarbon site.). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.5 As per claims 5 and 16, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the decline model is generated with a neural network using the historical production data and the dynamic well data as feature inputs and a target variable being resource production rate at time (t) (see Chung et al. [0036] The production prediction engine 106 is configured to generate a production decline curve (e.g., including predicted production data) in response to the cleaned historical data received from the data preparation engine 104. In one example, the production prediction engine 106 generates the production decline curve by fitting an Arps equation (e.g., an Arps hyperbolic to exponential decline curve) to historical production data. The result of applying the Arps equation is the predicted production decline curve, which forecasts oil and/or gas production volumes going forward from a particular start date, based on historical data from prior to the start date. [0037] For example, qi represents an initial production rate, Di represents an initial decline rate, b represents a forecast model curvature (e.g., a b-factor), and Df represents a final decline rate. The production decline curve for production as a function of time (e.g., q(t)) generated by the production prediction engine 106. [0098] Artificial Neural Networks (ANN's) are used to develop a model that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. The training set used to train the model include past completion intervals, and the training attributes are derived or calculated specifically over these intervals. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals.). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.6 Regarding claim 6, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the neural network includes two to seven dense layers and between 100 and 600 neurons per layer (see Benhallam et al. para [0096], forecasts production for each of the identified and validated targets using a range of different techniques. These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks which consists of a multi-layer perception (MLP), and thus would have been obvious to a person of skilled in the art. See further Chung et al. For example, k-means clustering is used to cluster wells according to well factors (e.g., qi, Di, b, and Df boundary conditions explained above) and a deep neural network algorithm is applied to predict the production decline curve for the given well based on the evenness of those well factors.[0059] The above description of the prediction system 100 of FIG. 1 generally refers to determinations related to a single hydrocarbon well, or at least a single hydrocarbon well at a time. However, in other examples of this description, the prediction system 100 is also configured to analyze a group or package of wells. For example, functionality of the prediction system 100 is applied to determine prediction accuracy metrics for each well in a group of wells (e.g., the wells in a basin). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [005
6.7 As per claim 7, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the dynamic well data includes one or more of a resource production rate for a previous time interval, a bottom hole pressure at the target well, a shut-in bottom hole pressure at the target well, and an average draw down pressure at the target well (see Benhallam et al. “historical production data”, Chung et al. para [0032], Other ancillary values for a well, such as surface location, bottomhole location, and lateral length may be obtained by accessing or “scraping” directional surveys, plat maps, or other types. In some examples, the data preparation engine 104 is configured to resolve double entries in the well parameters 102 (e.g., by removing an extraneous entry for a given date in the historical data for a given well), the rules for removal of which may vary by state. The data preparation engine 104 is configured to apply shut-in correction in some examples. [0040] In one non-limiting example, it is assumed that historical well parameters 102 are available for a well from two years prior to a present date (e.g., approximately two years of historical data in one example), and the start date for the prediction by the production prediction engine 106 is the present date. The production prediction engine 106 selects values of qi, Di, b, and Df that generate a decline curve for the previous two years, in which an error between the generated decline curve for the previous two years is reduced or minimized when compared to the historical data for the previous two years. Subsequently, the production prediction engine 106 generates a predicted production decline curve using the selected values of qi, Di, b, and Df beginning from the present date (e.g., the start date) and continuing until a specified end date, or beyond.). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.8 Regarding claims 8 and 17, the combined teachings of Benhallam et al. and Chung et al. teaches the step of identifying a subset of data from the historical production data associated with a shut-in period of days (see Chung et al. para [0031] For example, the data preparation engine 104 is configured to resolve monthly reported negative days on production, reported days on production greater than the number of days in the month, and/or missing or otherwise unavailable days on production. For example, negative days on production or greater than 31 days on production for a given month is classified as erroneous, and it may be assumed that such months include approximately 30.4375 days (e.g., an average number of days per month). [0032]. In some examples, the data preparation engine 104 is configured to resolve double entries in the well parameters 102 (e.g., by removing an extraneous entry for a given date in the historical data for a given well), the rules for removal of which may vary by state. The data preparation engine 104 is also configured to apply shut-in correction in some examples.); and removing the subset of data associated with the shut-in period of days from the historical production data (see Chung et al. para [0032], In some examples, the data preparation engine 104 is configured to resolve double entries in the well parameters 102 (e.g., by removing an extraneous entry for a given date in the historical data for a given well), the rules for removal of which may vary by state. The data preparation engine 104 is also configured to apply shut-in correction in some examples further Benhallam et al. data filtering [0038], Still further, the computer system includes a well target identifier configured to search the 3D map to identify optimal target locations for horizontal or deviated wells according to the RPOS 3D map and well placement constraints including azimuth, target length, deviation range or others as defined by a user. Also part of the computer system are an optimization module configured to perform an interference analysis designed to filter through and select an optimal set of non-interfering well candidates, a forecasting module configured to forecast an initial production rate for the selected horizontal well placement candidate placed in the identified location on the hydrocarbon extraction site using analytical, simulation, or machine learning models, and a production initiator configured to initiate hydrocarbon production at the selected horizontal well placement candidate in the identified location.). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.9 As per claim 9, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the decline model has an elapsed days feature variable that is reset by an occurrence of an acid job or a recompletion at the target well (see Chung et al. para [0018] In this description, “well performance” may refer to a metric to evaluate a well. For example, well performance metrics may include initial rate (or decline in rate) of hydrocarbon production, final rate (or decline in rate) of hydrocarbon production, lifetime total hydrocarbon production, the net present value of the lifetime hydrocarbon production, or return on investment (ROI) from hydrocarbon production. [0039] In these examples, the production prediction engine 106 is configured to numerically solve for values of the above parameters (e.g., qi, Di, b, and Df) to reduce or minimize the error between the historical data provided by the data preparation engine 104 and a historical portion of the decline curve. For example, the production prediction engine 106 selects values of qi, Di, b, and Df that generate a decline curve for a time period that corresponds with a time period for which cleaned historical data is available (e.g., received from the data preparation engine 104), in which an error between the generated decline curve for that time period is reduced or minimized when compared to the historical data for that time period. The selected values of qi, Di, b, and Df may then be used to generate the predicted production data, or decline curve, that begins at a given start date and continues until a given end date, or beyond. Benhallam et al. para [0034] Continuing this embodiment, the computer system forecasts, using statistical characterization methods and machine learning algorithms, projected production results for recompletion opportunities at the hydrocarbon site. The projected production results include initial production estimates and/or ultimate recovery estimates. The computer system also filters a list of recompletion opportunities for selection according to geological feasibility, mechanical feasibility and/or engineering feasibility, and initiates hydrocarbon production at the recompletion opportunity.) . Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.10 With regards to claims 10 and 18, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the input data set is generated from an initial data set filtered based on a well age or a type of well (see Benhallam et al. [0006], he computer system also filters a list of recompletion opportunities for selection according to geological feasibility, mechanical feasibility and/or engineering feasibility, and, at least in some embodiments, initiates hydrocarbon production at the recompletion opportunity. Chung et al. para [0039] In these examples, the production prediction engine 106 is configured to numerically solve for values of the above parameters (e.g., qi, Di, b, and Df) to reduce or minimize the error between the historical data provided by the data preparation engine 104 and a historical portion of the decline curve. For example, the production prediction engine 106 selects values of qi, Di, b, and Df that generate a decline curve for a time period that corresponds with a time period for which cleaned/filtered historical data is available (e.g., received from the data preparation engine 104), in which an error between the generated decline curve for that time period is reduced or minimized when compared to the historical data for that time period.). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.11 As per claim 11, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the first time interval or the second time interval is based on a user input indicating a desired length of time, or a comparison of different lengths of time affecting an absolute percentage error of the predicted well production profile (see Chung et al. abstract, The method also includes comparing the predicted production data for a time period between the start time and the end time to historical production data for the hydrocarbon well from the time period [0039] In these examples, the production prediction engine 106 is configured to numerically solve for values of the above parameters (e.g., qi, Di, b, and Df) to reduce or minimize the error between the historical data provided by the data preparation engine 104 and a historical portion of the decline curve. For example, the production prediction engine 106 selects values of qi, Di, b, and Df that generate a decline curve for a time period that corresponds with a time period for which cleaned historical data is available (e.g., received from the data preparation engine 104), in which an error between the generated decline curve for that time period is reduced or minimized when compared to the historical data for that time period. [0080] The method 700 continues in block 704 with comparing the predicted production data for a time period between the start time and the end time to historical production data for the hydrocarbon well from the time period. The method 700 also continues in block 706 with determining a prediction accuracy for the hydrocarbon well based on the comparison. As explained above, in an example, the backtesting engine 108 is configured to compare the predicted production decline curve with the corresponding (e.g., occurring at the same time) historical production data from the same time period. A prediction accuracy is generated or determined based on the comparison of the predicted production decline curve with the corresponding historical production data from the same time period). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.12 Regarding claim 12, the combined teachings of Benhallam et al. and Chung et al.
teaches the step of developing the target well based on the predicted well production profile (see Benhallam et al. para [0098] Artificial Neural Networks (ANN's) are used to develop a model that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. The training set used to train the model include past completion intervals, and the training attributes are derived or calculated specifically over these intervals. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
6.13 As per claim 20, the combined teachings of Benhallam et al. and Chung et al. teaches that wherein the static model is generated based on an input data set including historical production data corresponding to one or more wells and the decline model is generated based on the historical production data and dynamic well data (see Benhallam et al. para [0096] “the used of machine learning techniques like supervised neural networks to obtain the static model“ Step 1205 forecasts production for each of the identified and validated targets using a range of different techniques. These include statistical methods that leverage spatial and temporal neighborhood fluid data, machine learning techniques like supervised neural networks, analytical models specifically designed for horizontal wells, and simulation models. [0098] Artificial Neural Networks (ANN's) are used to develop a model [decline model] that relates the production performance to a series of data attributes including petrophysical interpretations, geological properties, and various engineering attributes. Once the model is trained, the testing dataset is collected by deriving the testing attributes specifically over the target uncontacted pay intervals. Further see Chung et al. para [0036] The production prediction engine 106 is configured to generate a production decline curve (e.g., including predicted production data) in response to the cleaned historical data received from the data preparation engine 104. In one example, the production prediction engine 106 generates the production decline curve by fitting an Arps equation (e.g., an Arps hyperbolic to exponential decline curve) to historical production data. The result of applying the Arps equation is the predicted production decline curve, which forecasts oil and/or gas production volumes going forward from a particular start date, based on historical data from prior to the start date. See further [0040-0041]). Therefore, it would have been obvious to a person of skilled in the art at the time of filing of the applicant’s invention to combine the method of Chung et al. with that of Benhallam et al. because Chung et al. teaches suing the prediction modification engine 112 may leverage machine learning techniques to improve the accuracy of the decline curves generated by the production prediction engine 106 (see para [0056]).
Conclusion
7. Claims 1-20 are rejected and 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.
8. Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANDRE PIERRE-LOUIS whose telephone number is (571)272-8636. The examiner can normally be reached M-F 9:00 AM-5:00 PM.
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, EMERSON C PUENTE can be reached at 571-272-3652. 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.
/ANDRE PIERRE LOUIS/Primary Patent Examiner, Art Unit 2187 June 14, 2026
Prosecution Insights