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 – 20 are presented for examination.
The drawing is objected to under 37 CFR 1.83(a).
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Ethington; James M (US 20190156298 A1) in the view of Du, Xianping. A data mining methodology for vehicle crashworthiness design. Embry-Riddle Aeronautical University, 2019, further in the view of Fukushige; Takahiro (US 20130346047 A1).
This action is non final rejection.
Priority
Acknowledgment is made for applicant’s claim for the benefit of a prior filed application of PRO 63/149,138 filed on 02/12/2021/
Information Disclosure Statement
The IDS filed on 05/14/2024, 08/22/2024 and 03/24/2025 are reviewed and considered. See attached documents.
Drawings
The drawings are objected to under 37 CFR 1.83(a). The drawings must show every feature of the invention specified in the claims. Therefore, the “fourth set of features of the structure design of the article by increasing a dimensionality of the third set of features” on claim 12, must be shown or the feature(s) canceled from the claim(s). No new matter should be entered.
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Ethington; James M (US 20190156298 A1) in the view of Du, Xianping. A data mining methodology for vehicle crashworthiness design. Embry-Riddle Aeronautical University, 2019, further in the view of Fukushige; Takahiro (US 20130346047 A1).
As of claim 1, Ethington teaches receiving, by one or more processors, a plurality of first sets of design parameters for a structural design of an article; ([0021] The dataset may include numerical values organized as a plurality of repair parameters and a plurality of operational parameters for each of a plurality of systems (e.g., aircraft). The dataset may include a value of every parameter for each system. In some examples, parameters without complete data may be excluded in pre-processing … [0088] Processor unit 404 serves to run instructions that may be loaded into memory 406. Processor unit 404 may comprise a number of processors, a multi-processor core…), As cited above repair parameters and operational parameters are considered as first set of design parameter.
determining, by the one or more processors, a plurality of first sets of features of the structural design of the article based on the first sets of design parameters; ([0028] The selected predictor variables may correspond to mathematical combinations of one or more of the operational data parameters. The predictor variables may be selected by a user and/or may be selected as part of a comparative analysis of operational data parameters, e.g., automatically). In here, predictor variables are interpreted as first sets of features.
determining, by the one or processors, a plurality of second sets of features of the structural design of the article by reducing dimensionality of each of the first sets of features; ([0028], In some examples, predictor variables may be aggregated into derived scalar data features, which may then be used in place of the original predictor variables… [0083] At step 316, the method includes reducing dimensionality of the predictor variables. Dimensionality reduction reduces the number of random variables considered and may improve processing times and training performance in step 318. Any effective methods may be used, including but not limited to feature selection and principal component analysis).
clustering, by the one or more processors, the second set of features into a plurality of clusters; ([0029] Values of the predictor variables (e.g., from the historical database) for each system, in combination with the cluster assignment of each system, may be used to create a labeled dataset. In other words, a labeled dataset may be created that includes predictor variable values as inputs and cluster assignments as outputs).
training, by the one or more processors, a classifier model based on the clusters, ([0029], a labeled dataset may be created that includes predictor variable values as inputs and cluster assignments as outputs. The labeled dataset may be divided into training data and validation data and may be repeatedly divided as needed for training and testing of models. Training data and validation data may be complementary subsets of the labeled dataset).
wherein the classifier model is trained to predict a quality metric of the structural design of the article as a function of an input set of features of, the quality metric corresponding to a subjective evaluation of the structural design; ([0026], At step 16, process 10 includes assigning a repair forecast label to each cluster. Repair forecast labels may correspond to differences in one or more repair data parameters and may include any suitable descriptor(s) configured to convey meaningful information to a forecasting system or user. For example, the labels may include qualitative differences (e.g., more’ or ‘most’), relative quantitative values (e.g., 80th percentile), absolute quantitative values (e.g., greater than 10), and/or the like, or any combination thereof).
receiving, by the one or more processors, a second set of design variables for the structural design of the article; ([0029], The labeled dataset may be divided into training data and validation data, and may be repeatedly divided as needed for training and testing of models. Training data and validation data may be complementary subsets of the labeled dataset… [0068], To avoid bias in the training process, the training dataset and the evaluation dataset may be independent, sharing no input data. The training dataset may also be complementary subsets of the labeled dataset. Generally, the training dataset and the evaluation dataset have no overlap of data and show substantially the same statistical distribution).
Ethington does not explicitly teach training, by the one or more processors, a predictive model based on the first sets of design parameters and the second sets of features, wherein the predictive model is trained to predict a third set of features of the structural design of the article as a function of an input set of design parameters for the structural design of the article; determining, by the one or more processors, the third set of features of the structural design of the article based on the second set of design variables and the predictive model; and determining, by the one or processors, the quality metric of the structural design of the article based on the third set of features and the classifier model.
While Du teaches training, by the one or more processors, a predictive model based on the first sets of design parameters and the second sets of features, wherein the predictive model is trained to predict a third set of features of the structural design of the article as a function of an input set of design parameters for the structural design of the article;(section 2.3.2 “Surrogate based optimization”, The surrogate model is a mathematical model that learns the relationship between the design variables and the response from a dataset and then predicts the responses of new designs…Page 145, “introduction of four regression machine learning algorithms” In this appendix, four MLAs, i.e. the GPR, SVM, RFR, and ANN, are briefly introduced in the context of engineering design. The design variables are presented as [Symbol font/0xCE]Rn x (real value space with n dimensions), and the responses are noted as [Symbol font/0xCE]Rm y . The surrogate modeling aims to establish a prediction function between design variables and responses…section 7.3.3 “uncertain dataset mining” The frequencies for SEA values of the 1,000 new designs created with GPR are plotted in Figure 7.5. The labeling criteria are defined as good (“g”): SEA > 1,200 J/kg; intermediate (“m”): 800 J/kg < SEA ≤ 1,200 J/kg and poor (“p”): SEA ≤ 800 J/kg), as it is cited above different predictive model are trained and Specific energy absorption (SEA) is interpreted as third set of features.
determining, by the one or processors, the quality metric of the structural design of the article based on the third set of features and the classifier model (section 4.4.1, “data labeling”, After the simulation dataset was created, based on the requirements specified in the safety regulations or the designer’s preference, the predicted SEA of each DOE was classified and labeled as “good/g,” “intermediate/m” or “poor/p.” In this study, the vehicle performance was labeled as “g” if SEA ≥ 2,000 J/kg, as “m” when 1,500 J/kg ≤ SEA < 2,000 J/kg, and as “p” for SEA < 1,500 J/kg).
Ethington and Du are considered to be analogous to the claimed invention since it teaches machine learning based repair forecasting, clustering and designing of S shaped beam for a vehicle. Therefore, it would be obvious for a person of ordinary skill in the art before the effective filing date to integrate Du’d teaching of training a predictive model and predict design parameter such as Specific energy absorption and based in this it determines the quality metric of the design into Ethington model to use inputs of the first and second set of features with a classifier.
The motivation would have been improve processing times and training performance by performing dimensionality reduction reduces the number of random variables considered (Ethington, [0083]) and it develops a new data-driven, knowledge-based methodology to design a complex mechanical system, such as a crashworthy vehicle, effectively and efficiently (Du section 9.1, conclusion).
The modified model of Ethington and Du does not explicitly teach determining, by the one or more processors, the third set of features of the structural design of the article based on the second set of design variables and the predictive model.
While Fukushige teaches determining, by the one or more processors, the third set of features of the structural design of the article based on the second set of design variables and the predictive model; ([0100], The test data includes feature values, aerodynamic performance values, and categories of each test design…[0173], The performance estimating unit 44 reads test data from the learning data storage unit 12 and calculates the Cd values of the vehicle shape of the test data using the approximate model created by the approximate model creating unit 43 based on the feature values of the test data. The performance estimating unit 44 reads the Cd values of the vehicle shape of the test data acquired through the CFD from the learning data storage unit 12), as it is cited above the test data is considered as a second set of design variable and CD value is interpreted as third set of features.
Fukushige is considered as analogous to the claimed invention since it teaches performance prediction based on feature values. Therefore, it would be obvious for a person of ordinary skill in the art before the effective filing date to integrate Fukushige’s teaching of using test data into the approximate model and predict a third set of features like CD value into the modified model to train and test the predicted model.
The motivation would have been to improve the prediction accuracy by using test and train data and by creating an approximate model and use the approximate model when predicting the aerodynamic performance (Fukushige, [0243]).
Claim 14, is similar to claim 1, but it recites additional element which Ethington teaches, one or more processors; ( [0080, Processor unit 404 may comprise a number of processors, a multi-processor core, and/or a particular type of processor or processor ) and one or more non-transitory computer-readable media including one or more sequences of instructions which, when executed by the one or more processors ([0004], In some embodiments, a data processing system may include instructions stored in a memory and executed by a processor…[0100], computer-readable storage media 424 is a non-transitory, physical or tangible storage device used to store program code 418 rather than a medium that propagates or transmits program code 418).
Claim 20, is also similar to the claim 1, but it recites additional element which Ethington teaches A non-transitory computer-readable medium including one or more sequences of instructions which, when executed by one or more processors (([0004], In some embodiments, a data processing system may include instructions stored in a memory and executed by a processor…[0100], computer-readable storage media 424 is a non-transitory, physical or tangible storage device used to store program code 418 rather than a medium that propagates or transmits program code 418).
Therefore claims 14 and 20 are rejected under the same rational as claim 1.
As of claim 2, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1 and Ethington also teaches , wherein the article a structural component of a vehicle ([0178], the illustrative embodiments and examples described herein are particularly useful for vehicle fleet repair forecasting and repair part demand forecasting).
As of claim 3, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 2, and Du also teaches wherein at least one of the first sets of design parameters comprises an indication of a thickness of a portion of the structural component (Section 4.3.1, “component parameterization and geometry modeling”, To conduct a detailed geometry design on the S-beam, its profile must be parameterized. The fully parameterized S-beam model is shown in Figure 4.2. The total length of the beam is 1,000 mm. The geometric parameters, including upper segment length (UL), upper radius (UR), slope segment angle (AN), bottom segment radius (DR), height (H), tube cross-section length (CL) and thickness (T), are design variables).
As of claim 4, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 2, and Du also teaches wherein at least one of the first sets of design parameters comprises an indication of a location of a portion of the structural component (Section 4.3.1, “component parameterization and geometry modeling”, The fully parameterized S-beam model is shown in Figure 4.2… tube cross-section length (CL) and thickness (T), are design variables. Points r and f are 27 two center points at the ends. Points a-d are the transition points between different segments of this beam The following geometric constraints were used: H – (UR + DR) × (1 - cos(AN)) > 0 in the vertical direction, which makes the length of segment bc greater than zero. 1,000 – (UL + H/tan(AN) + (UR + DR) × tan(AN/2)) > 0 in the horizontal direction, which guarantees the length of segment ra is greater than zero. The ranges of the design variables are as follows: 100 mm ≤ UL ≤ 400 mm; 150 mm ≤ UR ≤ 600 mm; 20° ≤ UL ≤ 40°; 150 mm ≤ DR ≤ 600 mm; 150 mm ≤ H ≤ 300 mm; 50 mm ≤ CL ≤ 70 mm; and 1.5 mm ≤ T ≤ 3.0 mm).
As of claim 5, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 2, and Du also teaches wherein at least one of the first sets of features comprises an indication of one or more deformations of the structural component in response to an applied force (Section 3.3.1 “step1 design at the system level”, The geometric design of each component is changed, in principle, to modify its average stiffness (avgstiff) under the corresponding crash condition. This parameter can be regarded as a feature of each component, since it is closely related to energy absorption, acceleration and intrusion).
Claim 15 is in the same scope as claim 5, so claim 15 is rejected under the same ration as claim 5.
As of claim 6, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 2, and Du also teaches wherein at least one of the second sets of features comprises an indication of one or more deformations of the structural component in response to an applied force (Section 6.3 “structural analysis Cases for evaluation of MLAs”, In the TbPT model, the circular cross-section areas are taken as design variables with the range from 0.6 to 225.8 cm2, and the other dimensions are listed in Figure 6.2(a). Two loads with the same magnitude (444.8 kN) are applied on joints T and S, and the vertical displacement of joint S is taken as the response of this system). As cited above the design variable of cross section area is used and a force is applied so, change in cross section is interpreted as an indication of the deformation due to applied force.
Claim 16 is in the same scope as claim 6, so claim 16 is rejected under the same ration as claim 6.
As of claim 7, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1, and Ethington teaches wherein the dimensionality of each of the first sets of features is reduced by at least one of principal component analysis or an auto-encoder ([0083] At step 316, the method includes reducing dimensionality of the predictor variables. Dimensionality reduction reduces the number of random variables considered, and may improve processing times and training performance in step 318. Any effective methods may be used, including but not limited to feature selection and principal component analysis).
Claim 17 is in the same scope as claim 7, so claim 17 is rejected under the same ration as claim 7.
As of claim 8, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1, and Ethington also teaches wherein the predictive model is trained using at least one of at least square regression analysis, Gaussian process regression analysis, or a neural network ([0030] One or more supervised learning algorithms may be used to train a model on the training data. These algorithms may include K Nearest Neighbor, Support Vector Machine, Naïve Bayes, Neural Networks, and/or comparable algorithms).
Claim 18 is in the same scope as claim 8, so claim 18 is rejected under the same ration as claim 8.
As of claim 9, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1, and Ethington also teaches where in the classifier is trained Using at least one of a logistic regression analysis, a random forest analysis, or a neural network ([0064] A library of algorithms appropriate to classification problems may be included in the training module. For example, appropriate algorithms may include artificial neural networks, learned decision trees, and/or support vector machines. In some examples, the library may also include ensemble aggregation methods).
Claim 19 is in the same scope as claim 9, so claim 19 is rejected under the same ration as claim 9.
As of claim 10, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1, and Ethington also teaches wherein the classifier model is trained based on input by a user ([0064] A library of algorithms appropriate to classification problems may be included in the training module. For example, appropriate algorithms may include artificial neural networks, learned decision trees, and/or support vector machines. In some examples, the library may also include ensemble aggregation methods. The algorithms of the library may also be further tailored by choosing specific parameters such as learning rate, number of trees, or kernel type. Training module 116 may test one algorithm with multiple parameters, may select parameters or algorithms best suited to the labeled dataset, may receive selections from the user, or may use any appropriate method to select a set of algorithms with which to train and evaluate models).
As of claim 11, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1, and Du also teaches wherein the input comprises, for at least some of the clusters, an indication of a respective quality metric assigned to that cluster by a user (Section 5.3.4 “Data mining on the design dataset”, Before the data could be mined, the design alternatives were labeled in terms of simulated SEA and CFE performances. The labeling criteria can be set in an arbitrary way based on the costumer’s requirements or the designer’s experience. In the present study, the following criteria were used: g: SEA > 1,800 J/kg and CFE > 0.1 m: 1,400 < SEA ≤ 1,800 J/kg and 0.08 < CFE ≤ 0.1 p: SEA ≤ 1,400 J/kg and CFE ≤ 0.08).
As of claim 12, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 1, and Du also teaches determining, by the one or more processors, a fourth set of features of the structural design of the article by increasing a dimensionality of the third set of features (Section 5.1 “introduction”
PCA is a mathematical procedure to reduce high dimensional data by expressing them with a set of linearly uncorrelated orthogonal basis vectors, or principal components (PCs). The original dataset can be recovered using a decompression procedure by multiplying the orthogonal basis with the principal component scores (PCSs) …By changing the value of the PCSs, the geometry of the part can be modified by the decompression process of PCA).
As of claim 13, the modified model of Ethington-Du- Fukushige teaches all the limitations of claim 12, Du also teaches wherein the dimensionality of the third set of features is increased by at least one of principal component analysis or an auto-encoder (Section 5.2.1, “Node based shape design by principal component analysis(PCA)”, the PCSs can be used as the design variables in the shape design to govern the locations of control points implicitly. Through the inverse process of PCA with calculated and selected k PCSs and PCs, the updated geometry model can be reconstructed).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
NAKHJAVANI; Omid (US 20180341850 A1, Date Published, 2018-11-29), is similar to the claimed invention since it teaches a method for predicting preliminary design load cases using an artificial neural network, the method including obtaining, with a processor, historical data corresponding to an existing structural component.
HODEL BENJAMIN (AU 2015318047 A1, Date Published 2017-04-13) is similar to the claimed invention since it teaches a method for developing machine operation classifiers for a machine. The method includes receiving training data, determining one or more training features based on the training data values and determining one or more training labels (36) associated with one or more training features.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABRHAM A. TAMIRU whose telephone number is (571)272-6987. The examiner can normally be reached Monday - Friday 8:00am - 5:00pm.
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 at 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.
/ABRHAM ALEHEGN TAMIRU/Examiner, Art Unit 2188
/RYAN F PITARO/Supervisory Patent Examiner, Art Unit 2188