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 .
Remarks
2. Claims 1-20 have been examined. Claims 1-7 and 12-13 have been rejected. Claims 8-11 and 14-17 have been objected to. Claims 18-20 are allowed. This Office action is responsive to the amendment filed on February 4, 2026, which has been entered in the above identified application.
Claim Rejections - 35 USC § 101
3. The correction to claim 18 has been approved, and the rejections to claims 18 and 19 are withdrawn.
Claim Rejections - 35 USC § 103
4. 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.
5. Claims 1, 2, 4, 7 are rejected under 35 U.S.C. 103 as being unpatentable over Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016) in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), and further in view of Raveh et al (U.S. Patent No. 11,640,559).
5-1. Regarding claim 1, Butor teaches a surface mount technology (SMT) manufacturing system for fabricating printed circuit board assemblies (PCBAs), by disclosing inspection strategies during the process of assembly of advanced electronic system comprising printed circuit board assemblies [Page 1, Abstract and 1. Introduction].
Butor teaches said system comprising: a screen printer for depositing solder paste on conductive solder pads on a printed circuit board (PCB), by disclosing a screen printer for the paste deposition [page 2, last paragraph, lines 1-2; figure 2, ‘1. Solder paste deposition-DEK Galaxy’].
Butor teaches a solder paste inspection (SPI) sub-system for inspecting the solder paste deposited on the PCB to identify defects, by disclosing a solder paste inspection [figure 2, ‘2. Solder Paste Inspection SPI’].
Butor teaches a pick-and-place machine for placing circuit components on the solder paste, by disclosing a Pick&Place machine [figure 2, ‘3. Pick&Place I – Siplace SX1’].
Butor teaches a first automated optical inspection (AOI) sub-system for inspecting the PCB after the circuit components are placed on the PCB, by disclosing carrying out a test using AOI after component placement with Pick&Place machines and solder reflow [page 2, last paragraph, lines 5-6; figure 2, ‘6. Automatic Optical Inspextion AOI’].
Butor teaches a reflow soldering oven for bonding component leads both electrically and mechanically to the pads on the PCB, by disclosing a reflow convection oven [figure 2, ‘Reflow furnace – Vitronic Soltec MR933’].
Butor does not expressly teach an analysis engine responsive to process data and variables from each of the screen printer, the SPI sub-system,… and providing feedback signals to each of the screen printer, the SPI sub-system…, said engine providing statistical modeling to provide performance and process feedback control for self-corrections purposes. Grafstrom discloses forming a deposit on a workpiece [Abstract] using a system comprising an apparatus for forming a deposit of viscous medium, and a processing means, which may be implemented as a processing circuitry, configured to generate control instructions for operating the apparatus [page 3, lines 10-13]. The apparatus is a surface mount technology, SMT, tool and the viscous medium a SMT medium used for mounting SMT electrical components on the workpiece [page 10, lines 23-25]. A "workpiece" may be a board (e.g., a printed circuit board (PCB)) [page 7, lines 20-21]. The system comprises an inspection tool, such as an Automated Optical Inspection, AOI, tool [page 3, lines 19-21] and the SMT tool uses external measurement equipment, such as for example, a solder paste inspection, SPI, tool, and/or an internal tool that is integrated in the apparatus, and sensors measuring for example, temperature, pressure or viscosity to generate input for improving or modifying the generation of the control instructions [page 5, lines 3-12]. The system employs a neural network to generate control instructions for operating the apparatus [page 3, lines 13-18]. Examples of neural networks include artificial neural networks, probabilistic graphical models such as Bayesian networks, probabilistic classifiers and/or controllers (e.g. Gaussian mixture models) [page 15, lines 4-14]. Data generated by the SPI tool may be used to train the neural network model to improve the operation of an individual apparatus, to create control instructions for a specific type of viscous medium, to compensate for e.g. ageing effects of the viscous medium, environmental parameters, etc. [page 5, lines 6-10]. Additionally, sensors measuring for example temperature, pressure or viscosity may be employed to generate input for improving or modifying the generation of the control instructions [page 5, lines 10-12]. The environmental parameter and the deposit parameter may be processed through a neural network to generate 130 the control instructions for use when operating the apparatus to form the deposit onto the workpiece [page 25, lines 16-20]. Use of a neural network allows for a more dynamic and efficient optimization of control instructions, since they may be generated for example on a workpiece to workpiece basis, every time a new type or batch of viscous medium is introduced, or whenever the user wishes so [page 3, lines 28-32]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide, for the screen printer and SPI of Butor, an analysis engine for processing data and providing feedback to the screen printer and SPI, as taught by Grafstrom. This would allow for more dynamic and efficient optimization of control instructions.
Butor-Grafstrom do not expressly teach the analysis engine responsive to process data and variables from… the pick-and-place machine,… and provide feedback signals to… the pick-and-place machine…, said engine providing statistical modeling to provide performance predictability and process feedback control for self-correction purposes. Cvijetinovic discloses a nozzle performance analytics system [column 8, line 55 to column 9, line 8] that collects data from pick and place equipment at one or more industrial facilities, generates real-time and time-series performance vector data for respective pick and place machines, and analyzes the machine-specific performance vectors to predict future operational trends or future times of failure for pick and place nozzles or heads [column 9, lines 9-27; column 14, lies 63-67; column 15, lines 54-64]. Data collected includes real-time and historical statistical and operational data [column 9, line 66 to column 10, line 16]. As analytics system tracks cumulative behavior and performance of each nozzle over time, a vector analysis component can generate performance model data for each nozzle that represents expected behavior of the model over time, and can be generated based on monitored historical performance of each nozzle [column 18, line 50 to column 19, line 2]. The analytics system can be configured to deliver control outputs directed to selected control devices in response to detecting or predicting a nozzle failure or performance degradation, such as altering operation of a selected pick and place machine to either mitigate the identified performance issue or to slow the performance degradation until the defective nozzle can be repaired or replaced [column 10, lines 38-45]. This would decrease the likelihood of a failed or improper pick. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the pick and place machine of Butor-Grafstrom, an analysis engine to process information from the pick and place machine and provide feedback based on statistical modeling for taking corrective action, as taught by Cvijetinovic. This would decrease the likelihood of a failed or improper pick.
Butor-Grafstrom-Cvijetinovic do not expressly teach the analysis engine responsive to process data and variables from… the first AOI sub-system and the reflow soldering oven and providing feedback signals to… the first AOI sub-system and the reflow soldering oven, said engine providing statistical modeling to provide performance predictability and process feedback control for self-correction purposes. Raveh discloses use of AI-based software configured to monitor a production process or product of manufacturing machine [column 3, lines 48-54]. Self-motivated learning is performed to iteratively improve an accuracy of a classification model [column 4, lines 29-31]. The accuracy of the classification model may be improved by selecting an algorithm, such as which neural network to use [column 5, lines 1-10]. This includes use of a Deep Neural Network (DNN) [column 14, lines 51-57]. Capabilities of the disclosed solution may be embedded into the internal algorithmic process of manufacturing a product and Quality Assurance thereof [column 5, lines 45-55]. The classification model may be utilized for a supervised image classification task, such as for AOI classification, In-Process-Quality-Control (IPCQ) tasks, QI task or the like [column 7, lines 20-25]. The AI-based software may be configured to determine, such as based on visual input, if a machine functions properly by utilizing a classification model to classify products image, classify items within the images, identify defects within items or products, or the like [column 3, lines 48-59]. AI-based software may be used for Automated Optical Inspection (AOI), such as Printed Circuit Board (PCB) manufacture [column 3, lines 60-63]. Use of the artificial intelligence system having automatic improvement of artificial intelligence classification models on the processes involved in PCB manufacture would improve the overall quality of the fabrication process by providing more accurate identification of defects and more optimized processes, such as soldering of components. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the AOI and reflow oven of Butor-Grafstrom-Cvijetinovic, an artificial intelligence system to process information from the AOI and reflow oven and provide feedback based on statistical modeling for taking corrective action for both devices, since Raveh discloses that such a system would improve the overall quality of the fabrication process.
5-2. Regarding Claim 2, Butor-Grafstrom-Cvijetinovic-Raveh teach all the limitations of claim 1, wherein the analysis engine employs a self-learning Markov decision process (MDP) model that manages sequential decision process outcomes in which states and transitions are quantified in calculated rewards during the transition between two Markov states and provides multi-agent reinforcement learning, by disclosing that the neural network may use a “Deep reinforcement learning” algorithm [Grafstrom, page 28, lines 3-11; Raveh, column 12, lines 45-64, column 14, lines 51-57], which fundamentally relies on a Markov Decision Process.
5-3. Regarding claim 4, Butor-Grafstrom-Cvijetinovic-Raveh teach all the limitations of claim 1, further comprising an auto-insertion machine for inserting additional components on the PCB that are not able to be placed by the pick-and-place machine, by disclosing a second Pick&Place machine [figure 2, ‘4 Pick&Place II – Siplace SX2’].
Regarding the limitation said analysis engine being responsive to data and variables from the auto-insertion machine and providing feedback signals to the auto-insertion machine for self-correction purposes, as disclosed above with respect to claim 1, Cvijetinovic discloses using machine learning to process information from a pick and place machine to predict future times of failure [Cvijetinovic, column 9, lines 9-27; column 14, lies 63-67; column 15, lines 54-64] and take corrective action [Cvijetinovic, column 10, lines 38-45] to decrease the likelihood of a failed or improper pick. This could also be used on the second pick and place machine of [Cvijetinovic, figure 2] to obtain such advantages. Also, as discussed above, Raveh discloses use of AI-based software configured to monitor a production process or product of manufacturing machine [Raveh, column 3, lines 48-54]. Self-motivated learning is performed to iteratively improve an accuracy of a classification model [Raveh, column 4, lines 29-31] and such AI-based software may be used for In-Process-Quality-Control (IPCQ) tasks, QI task or the like [Raveh, column 7, lines 20-25]. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the second pick and place machine, an artificial intelligence system to process information from the second pick and place machine, and provide feedback for taking corrective action, since Cvijetinovic and Raveh disclose that such a system would improve the overall quality of the fabrication process.
5-4. Regarding claim 7, Butor-Grafstrom-Cvijetinovic-Raveh teach all the limitations of claim 1, further comprising an in-circuit testing machine for performing electrical testing on the PCB, by disclosing functional circuit test [figure 2, ’10. Functional Inspection FCT’].
As for the limitation said analysis engine being responsive to data and variables from the in-circuit testing machine and providing feedback signals to the in-circuit testing machine for self-correction purposes, as discussed above, Raveh discloses use of AI-based software configured to monitor a production process or product of manufacturing machine [Raveh, column 3, lines 48-54]. Self-motivated learning is performed to iteratively improve an accuracy of a classification model [Raveh, column 4, lines 29-31] and such AI-based software may be used for In-Process-Quality-Control (IPCQ) tasks, QI task or the like [Raveh, column 7, lines 20-25]. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the functional circuit test machine, an artificial intelligence system to process information from the functional circuit test machine, and provide feedback for taking corrective action, since Raveh discloses that such a system would improve the overall quality of the fabrication process.
6. Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016) in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), in view of Raveh et al (U.S. Patent No. 11,640,559), and further in view of Samsudin et al (Pub. No. US 2020/0367367).
6-1. Regarding Claim 3, Butor-Grafstrom-Cvijetinovic-Raveh teach all the limitations of claim 1. While Butor-Grafstrom-Cvijetinovic-Raveh were able to teach the first AOI sub-system in claim 1 as the AOI that performs inspection after the Pick&Place machines and solder reflow [Butor, figure 2, ‘6. Automatic Optical Inspextion AOI’] because such an AOI of Butor is performed after the circuit components are placed on the PCB, such an interpretation would not hold given the limitation of claim 3, which recites the system according to claim 1 further comprising a second AOI sub-system for inspecting the PCB after the PCB has been to the reflow soldering oven, said analysis engine being responsive to data and variables from the second AOI sub-system and providing feedback signals to the second AOI sub-system for self-correction purposes. Based on this limitation, the AOI shown in [Butor, figure 2, ‘6. Automatic Optical Inspextion AOI’] is now being interpreted as the second AOI sub-system. Therefore, with respect to claim 3, Butor-Grafstrom-Cvijetinovic-Raveh do not expressly teach in addition to the second AOI sub-system, a first automated optical inspection (AOI) sub-system for inspecting the PCB after the circuit components are placed on the PCB. Samsudin discloses that after picking and placing surface mount technology (SMT) components on a PCB, an automated optical inspection (AOI) may be used for inspecting the SMT components to check for catastrophic failure (e.g. a missing component), quality defects and other parameters or characteristics [paragraph 24, lines 12-16; paragraph 30, lines 10-16]. Samsudin also discloses that implementing the reflowing may include inspecting of the stacked assembly after reflow using an AOI [paragraph 48, lines 6-9]. This would help prevent defects in the fabrication process. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide, in addition to an AOI after reflow, an AOI after the circuit components are placed on the PCB, as taught by Samsudin. This would help prevent defects in the fabrication process.
As for said analysis engine being responsive to data and variables from the first and second AOI sub-systems, and providing feedback signals to the first and second AOI sub-systems for self-correction purposes, as discussed above, Raveh discloses use of AI-based software configured to monitor a production process or product of manufacturing machine [column 3, lines 48-54]. Self-motivated learning is performed to iteratively improve an accuracy of a classification model [column 4, lines 29-31]. Capabilities of the disclosed solution may be embedded into the internal algorithmic process of manufacturing a product and Quality Assurance thereof [column 5, lines 45-55]. The classification model may be utilized for a supervised image classification task, such as for AOI classification, In-Process-Quality-Control (IPCQ) tasks, QI task or the like [column 7, lines 20-25]. The AI-based software may be configured to determine, such as based on visual input, if a machine functions properly by utilizing a classification model to classify products image, classify items within the images, identify defects within items or products, or the like [column 3, lines 48-59]. AI-based software may be used for Automated Optical Inspection (AOI), such as Printed Circuit Board (PCB) manufacture [column 3, lines 60-63]. Use of the artificial intelligence system having automatic improvement of artificial intelligence classification models on the processes involved in PCB manufacture would improve the overall quality of the fabrication process by providing more accurate identification of defects. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the first and second AOI, an artificial intelligence system to process information from the first and second AOI and provide feedback for taking corrective action for both devices, since Raveh discloses that such a system would improve the overall quality of the fabrication process.
7. Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016) in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), in view of Raveh et al (U.S. Patent No. 11,640,559), and further in view of Zheng et al (U.S. Patent No. 10,939,600).
7-1. Regarding claim 5, Butor-Grafstrom-Cvijetinovic-Raveh teach all the limitations of claim 1. Butor-Grafstrom-Cvijetinovic-Raveh do not expressly teach the claim further comprising a wave solder machine for bulk soldering the PCB. Zheng is directed to fabricating printed circuit board assemblies (PCBs) using wave soldering by simultaneously connecting multiple electronic components and connectors on a circuit board [column 1, lines 5-8; column 2, lines 23-37]. This would expand the types of circuit board configuration that can be fabricated by allowing though-hole components to be soldered. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide a wave solder machine, as taught by Zheng. This would expand the types of circuit board configuration that can be fabricated.
As for said analysis engine being responsive to data and variables from the wave soldering machine, and providing feedback signals to the wave soldering machine for self-correction purposes, as discussed above, Raveh discloses use of AI-based software configured to monitor a production process or product of manufacturing machine [Raveh, column 3, lines 48-54]. Self-motivated learning is performed to iteratively improve an accuracy of a classification model [Raveh, column 4, lines 29-31] and such AI-based software may be used for In-Process-Quality-Control (IPCQ) tasks, QI task or the like [Raveh, column 7, lines 20-25]. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the wave soldering machine, an artificial intelligence system to process information from the wave soldering machine, and provide feedback for taking corrective action, since Raveh discloses that such a system would improve the overall quality of the fabrication process.
8. Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016) in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), in view of Raveh et al (U.S. Patent No. 11,640,559), and further in view of Adler et al (Pub. No. US 2021/0012499).
8-1. Regarding Claim 6, Butor-Grafstrom-Cvijetinovic-Raveh teach all the limitations of claim 1, further comprising an in-line X-ray inspection machine for performing an X-ray inspection process on the PCB, by disclosing X-ray Inspection [Butor, figure 2, 9. X-Ray Inspection XCT’].
Butor-Grafstrom-Cvijetinovic-Raveh do not expressly teach said analysis engine being responsive to data and variables from the in-line X-ray inspection machine and providing feedback signals to the in-line X-ray inspection machine for self-correction purposes. Adler discloses inspection of objects using X-rays [paragraph 2] where inline measurements may be performed on one or more systems or tools, such as an assembling tool [paragraph 30, lines 1-5]. During an assembling process performed by an assembly tool, the X-ray system may inspect the assembled components or devices during one or more steps in the assembling process [paragraph 30, lines 20-23]. Reference images for components may be captured before the components are connected to the printed circuit boards [paragraph 43, lines 6-8]. One or more artificial intelligence modules and/or machine learning models may be used to identify defects [paragraphs 29, 44]. The artificial intelligence module and the machine learning algorithm may work in real time [paragraph 46]. This would provide higher quality products by allowing for the inspection of hidden defects. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide to the system of Butor-Grafstrom-Cvijetinovic-Raveh, x-ray inspection using artificial intelligence analysis to identify defects, as taught by Adler. This would provide higher quality products.
9. Claims 12-13 are rejected under 35 U.S.C. 103 as being unpatentable over Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016), in view of Samsudin et al (Pub. No. US 2020/0367367), in view of Zheng et al (U.S. Patent No. 10,939,600), in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), in view of Raveh et al (U.S. Patent No. 11,640,559), and further in view of Adler et al (Pub. No. US 2021/0012499).
9-1. Regarding claim 12, Butor teaches the claim of a surface mount technology (SMT) manufacturing system for fabricating printed circuit board assemblies (PCBAs), by disclosing inspection strategies during the process of assembly of advanced electronic system comprising printed circuit board assemblies [Page 1, Abstract and 1. Introduction].
Butor teaches said system comprising: a screen printer for depositing solder paste on conductive solder pads on a printed circuit board (PCB), by disclosing a screen printer for the paste deposition [page 2, last paragraph, lines 1-2; figure 2, ‘1. Solder paste deposition-DEK Galaxy’].
Butor teaches a solder paste inspection (SPI) sub-system for inspecting the solder paste deposited on the PCB to identify defects, by disclosing a solder paste inspection [figure 2, ‘2. Solder Paste Inspection SPI’].
Butor teaches a pick-and-place machine for placing circuit components on the solder paste, by disclosing a Pick&Place machine [figure 2, ‘3. Pick&Place I – Siplace SX1’].
Butor teaches… a reflow soldering oven for bonding component leads both electrically and mechanically to the pads on the PCB, by disclosing a reflow convection oven [figure 2, ‘Reflow furnace – Vitronic Soltec MR933’].
Butor teaches a second AOI sub-system for inspecting the PCB after the PCB has been to the reflow soldering oven, by disclosing carrying out a test using AOI after component placement with Pick&Place machines and solder reflow [page 2, last paragraph, lines 5-6; figure 2, ‘6. Automatic Optical Inspextion AOI’].
Butor teaches an auto-insertion machine for inserting additional components on the PCB that are not able to be placed by the pick-and-place machine, by disclosing a second Pick&Place machine [figure 2, ‘4 Pick&Place II – Siplace SX2’].
Butor teaches… an in-line X-ray inspection machine for performing an X-ray inspection process on the PCB, by disclosing X-ray Inspection [Butor, figure 2, 9. X-Ray Inspection XCT’].
Butor teaches an in-circuit testing machine for performing electrical testing on the PCB, by disclosing functional circuit test [figure 2, ’10. Functional Inspection FCT’].
Butor does not expressly teach a first automated optical inspection (AOI) sub-system for inspecting the PCB after the circuit components are placed on the PCB. Samsudin discloses that after picking and placing surface mount technology (SMT) components on a PCB, an automated optical inspection (AOI) may be used for inspecting the SMT components to check for catastrophic failure (e.g. a missing component), quality defects and other parameters or characteristics [paragraph 24, lines 12-16; paragraph 30, lines 10-16]. Samsudin also discloses that implementing the reflowing may include inspecting of the stacked assembly after reflow using an AOI [paragraph 48, lines 6-9]. This would help prevent defects in the fabrication process. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide, in addition to an AOI after reflow, an AOI after the circuit components are placed on the PCB, as taught by Samsudin. This would help prevent defects in the fabrication process.
Butor-Samsudin do not expressly teach a wave solder machine for bulk soldering the PCB. Zheng is directed to fabricating printed circuit board assemblies (PCBs) using wave soldering by simultaneously connecting multiple electronic components and connectors on a circuit board [column 1, lines 5-8; column 2, lines 23-37]. This would expand the types of circuit board configuration that can be fabricated by allowing though-hole components to be soldered. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide a wave solder machine, as taught by Zheng. This would expand the types of circuit board configuration that can be fabricated.
Butor-Samsudin-Zhang do not expressly teach an analysis engine responsive to process data and variables from each of the screen printer, the SPI sub-system,… and providing feedback signals to each of the screen printer, the SPI sub-system,… said engine providing statistical modeling to provide performance predictability and process feedback control for self-correction purposes. Grafstrom discloses forming a deposit on a workpiece [Abstract] using a system comprising an apparatus for forming a deposit of viscous medium, and a processing means, which may be implemented as a processing circuitry, configured to generate control instructions for operating the apparatus [page 3, lines 10-13]. The apparatus is a surface mount technology, SMT, tool and the viscous medium a SMT medium used for mounting SMT electrical components on the workpiece [page 10, lines 23-25]. A "workpiece" may be a board (e.g., a printed circuit board (PCB)) [page 7, lines 20-21]. The system comprises an inspection tool, such as an Automated Optical Inspection, AOI, tool [page 3, lines 19-21] and the SMT tool uses external measurement equipment, such as for example, a solder paste inspection, SPI, tool, and/or an internal tool that is integrated in the apparatus, and sensors measuring for example, temperature, pressure or viscosity to generate input for improving or modifying the generation of the control instructions [page 5, lines 3-12]. The system employs a neural network to generate control instructions for operating the apparatus [page 3, lines 13-18]. Examples of neural networks include artificial neural networks, probabilistic graphical models such as Bayesian networks, probabilistic classifiers and/or controllers (e.g. Gaussian mixture models) [page 15, lines 4-14]. Data generated by the SPI tool may be used to train the neural network model to improve the operation of an individual apparatus, to create control instructions for a specific type of viscous medium, to compensate for e.g. ageing effects of the viscous medium, environmental parameters, etc. [page 5, lines 6-10]. Additionally, sensors measuring for example temperature, pressure or viscosity may be employed to generate input for improving or modifying the generation of the control instructions [page 5, lines 10-12]. The environmental parameter and the deposit parameter may be processed through a neural network to generate 130 the control instructions for use when operating the apparatus to form the deposit onto the workpiece [page 25, lines 16-20]. Use of a neural network allows for a more dynamic and efficient optimization of control instructions, since they may be generated for example on a workpiece to workpiece basis, every time a new type or batch of viscous medium is introduced, or whenever the user wishes so [page 3, lines 28-32]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide, for the screen printer and SPI of Butor-Samsudin-Zheng, an analysis engine for processing data and providing feedback to the screen printer and SPI, as taught by Grafstrom. This would allow for more dynamic and efficient optimization of control instructions.
Butor-Samsudin-Zheng-Grafstrom do not expressly teach the analysis engine responsive to process data and variables from… the pick-and-place machine,… and providing feedback signals to… the pick-and-place machine…, said engine providing statistical modeling to provide performance predictability and process feedback control for self-correction purposes. Cvijetinovic discloses a nozzle performance analytics system [column 8, line 55 to column 9, line 8] that collects data from pick and place equipment at one or more industrial facilities, generates real-time and time-series performance vector data for respective pick and place machines, and analyzes the machine-specific performance vectors to predict future operational trends or future times of failure for pick and place nozzles or heads [column 9, lines 9-27; column 14, lies 63-67; column 15, lines 54-64]. Data collected includes real-time and historical statistical and operational data [column 9, line 66 to column 10, line 16]. As analytics system tracks cumulative behavior and performance of each nozzle over time, a vector analysis component can generate performance model data for each nozzle that represents expected behavior of the model over time, and can be generated based on monitored historical performance of each nozzle [column 18, line 50 to column 19, line 2]. The analytics system can be configured to deliver control outputs directed to selected control devices in response to detecting or predicting a nozzle failure or performance degradation, such as altering operation of a selected pick and place machine to either mitigate the identified performance issue or to slow the performance degradation until the defective nozzle can be repaired or replaced [column 10, lines 38-45]. This would decrease the likelihood of a failed or improper pick. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the pick and place machine of Butor-Samsudin-Zheng-Grafstrom, an analysis engine to process information from the pick and place machine and provide feedback based on statistical modeling for taking corrective action, as taught by Cvijetinovic. This would decrease the likelihood of a failed or improper pick.
Butor-Samsudin-Zheng-Grafstrom-Cvijetinovic do not expressly teach the analysis engine responsive to process data and variables from… the first AOI sub-system, the reflow soldering oven, the second AOI sub-system, the auto-insertion machine, the wave solder machine,… and the in-circuit testing machine and providing feedback signals to.. the first AOI sub-system, the reflow soldering oven, the second AOI sub-system, the auto-insertion machine, the wave solder machine,… and the in-circuit testing machine, said engine providing statistical modeling to provide performance predictability and process feedback control for self-correction purposes. Raveh discloses use of AI-based software configured to monitor a production process or product of manufacturing machine [column 3, lines 48-54]. Self-motivated learning is performed to iteratively improve an accuracy of a classification model [column 4, lines 29-31]. The accuracy of the classification model may be improved by selecting an algorithm, such as which neural network to use [column 5, lines 1-10]. This includes use of a Deep Neural Network (DNN) [column 14, lines 51-57]. Capabilities of the disclosed solution may be embedded into the internal algorithmic process of manufacturing a product and Quality Assurance thereof [column 5, lines 45-55]. The classification model may be utilized for a supervised image classification task, such as for AOI classification, In-Process-Quality-Control (IPCQ) tasks, QI task or the like [column 7, lines 20-25]. The AI-based software may be configured to determine, such as based on visual input, if a machine functions properly by utilizing a classification model to classify products image, classify items within the images, identify defects within items or products, or the like [column 3, lines 48-59]. AI-based software may be used for Automated Optical Inspection (AOI), such as Printed Circuit Board (PCB) manufacture [column 3, lines 60-63]. Use of the artificial intelligence system having automatic improvement of artificial intelligence classification models on the processes involved in PCB manufacture would improve the overall quality of the fabrication process by providing more accurate identification of defects and more optimized processes. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use, for the first AOI, the reflow oven, the second AOI, the second pick and place machine, the wave solder machine, and the functional circuit test machine of Butor-Samsudin-Zheng-Grafstrom-Cvijetinovic, an artificial intelligence system to process information from the first AOI, the reflow oven, the second AOI, the second pick and place machine, the wave solder machine, and the functional circuit test machine, and provide feedback based on statistical modeling for taking corrective action for each of the devices, since Raveh discloses that such a system would improve the overall quality of the fabrication process.
Butor-Samsudin-Zheng-Grafstrom-Cvijetinovic-Raveh do not expressly teach the artificial intelligence (Al) / machine learning (ML) analysis engine responsive to process data and variables from… the in-line X-ray inspection machine… and providing feedback signals to… the in-line X-ray inspection machine…, said engine providing statistical modeling to provide performance predictability and process feedback control for self-correction purposes. Adler discloses inspection of objects using X-rays [paragraph 2] where inline measurements may be performed on one or more systems or tools, such as an assembling tool [paragraph 30, lines 1-5]. During an assembling process performed by an assembly tool, the X-ray system may inspect the assembled components or devices during one or more steps in the assembling process [paragraph 30, lines 20-23]. Reference images for components may be captured before the components are connected to the printed circuit boards [paragraph 43, lines 6-8]. One or more artificial intelligence modules and/or machine learning models may be used to identify defects based on a generated statistical process control chart [paragraphs 29, 44; paragraph 46, lines 1-7]. The artificial intelligence module and the machine learning algorithm may work in real time [paragraph 46, lines 7-11]. This would provide higher quality products by allowing for the inspection of hidden defects. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to provide to the system of Butor-Samsudin-Zheng-Grafstrom-Cvijetinovic-Raveh, x-ray inspection using artificial intelligence analysis to identify defects, as taught by Adler. This would provide higher quality products.
9-2. Regarding Claim 13, Butor-Samsudin-Zheng-Grafstrom-Cvijetinovic-Raveh-Adler teach all the limitations of claim 12, wherein the analysis engine employs a self-learning Markov decision process (MDP) model that manages sequential decision process outcomes in which states and transitions are quantified in calculated rewards during the transition between two Markov states and provides multi-agent reinforcement learning, by disclosing that the neural network may use a “Deep reinforcement learning” algorithm [Grafstrom, page 28, lines 3-11; Raveh, column 12, lines 45-64, column 14, lines 51-57], which fundamentally relies on a Markov Decision Process.
Allowable Subject Matter
10. Claims 8-11 and 14-17 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
11. Claims 18-29 are allowed.
Response to Arguments
12. The Examiner acknowledges the Applicant’s amendments to claims 12, 18, and 20. Examiner notes that the last limitation of claim 1 has been amended even though the status of the claim is listed as “Original.”
Regarding independent claim 1, Applicant alleges that Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016) in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), and further in view of Raveh et al (U.S. Patent No. 11,640,559) do not expressly teach a single analysis engine responsive to process data and variables from a screen printer, an SPI sub-system, a pick-and-place machine, an AOI sub-system and a reflow soldering oven and providing feedback signals to each of the devices because (1) the Gastyrom control unit 32 only controls one device in an SMT manufacturing system and (2) Cvijetnovic and Raveh do not disclose a single analysis engine that receives process data and variables and provides feedback signals to all the claimed devices.
Contrary to Applicant’s arguments, the combination of Butor, in view of Grafstrom, in view of Cvijetnovic, and further in view of Raveh teach the claim limitation. Grafstrom, Cvijetnovic, and Raveh each teach an analysis engine that uses a statistical modeling to provide performance predictability and process feedback control for self-correction purposes to improve an aspect of fabricating a printed circuit board assembly [Gafstrom, Abstract; page 7, line 20; page 3, lines 13-18; page 5, lines 6-10; page 15, lines 4-14] [Cvijetnovic, column 8, line 55 to column 9, line 8; column 9, line 66 to column 10, line 16; column 15, lines 54-64; column 18, line 50 to column 19, line 2] [Raveh, column 3, lines 48-54, 60-63; column 4, lines 29-31; column 5, lines 1-10, 45-55; column 14, lines 51-57]. The analysis engine of Grafstrom processes data and variables from a screen printer and solder paste inspection sub-system [Grafstrom, page 5, lines 3-12, page 25, lines 16-20]. The analysis engine of Cvijetnovic processes data and variables from a pick-and-place machine [Cvijetinovic, column 9, lines 9-27; column 9, line 66 to column 10, line 1; column 14, lines 63-67]. The analysis engine of Raveh processes data and variables from different stages in the manufacture of printed circuit boards, such as in AOI sub-system [Raveh, column 5, line 67 to column 6, line 8], and which would include the reflow furnace stage of Butor [figure 2, ‘Reflow furnace – Vitronic Soltec MR933’]. It would have been obvious to one of ordinary skill in the art to use the teachings of Grafstrom, Cvijetnovic, and Raveh to provide a single analysis engine that would process data and variables from each of the devices of Butor. This would not be outside the capability of an ordinary skilled person in the art especially since the claim does not require input or output of data from one of the devices to another of the devices. That is, the analysis of data and variables for each device may be carried out independent from another device to optimize each individual stage of the manufacturing process. To overcome such an obviousness rejection, Examiner suggests amending the claim to include additional details regarding the processing of data and variables of the devices by the analysis engine such that there exists some interdependence between results of the analysis.
Regarding independent claim 12, Applicant alleges that Butor et al (“Diagnosis of electronic systems in SMT technological line,” 2016), in view of Samsudin et al (Pub. No. US 2020/0367367), in view of Zheng et al (U.S. Patent No. 10,939,600), in view of Grafstrom (WO 2020/207893), in view of Cvijetinovic et al (U.S. Patent No. 10,624,251), in view of Raveh et al (U.S. Patent No. 11,640,559), and further in view of Adler et al (Pub. No. US 2021/0012499) do not teach a single analysis engine responsive to process data and variables from a screen printer, an SPI sub-system, a pick-and-place machine, two AOI sub-systems, a reflow soldering oven, an auto-insertion machine, a wave solder machine, an in-line X-ray inspection machine and an in-circuit testing machine, and providing feedback signals to each of the devices based on the same arguments from claim 1, and additional because Samsudin, Zheng, and Adler do not teach a single analysis engine that receives process data and variables and provides feedback signals to all the claimed devices.
Examiner notes that Samsudin was used in the rejection to teach a first automated optical inspection (AOI) sub-system for inspecting the PCB after the circuit components are placed on the PCB and Zheng was used in the rejection to teach a wave solder machine for bulk soldering the PCB. Contrary to Applicant’s arguments, as discussed above, Grafstrom, Cvijetinovic, and Raveh each teach an analysis engine that uses statistical modeling to provide performance predictability and process feedback control for self-correction purposes to improve an aspect of fabricating a printed circuit board assembly, with each reference disclosing processing data and variables from one or more devices that are part of the manufacturing process. Adler also teach an analysis engine that uses statistical modeling to provide performance predictability and process feedback control for self-correction purposes to improve an aspect of fabricating a printed circuit board assembly [Adler, paragraphs 25, 29, 44, 46]. The analysis engine of Adler processes data and variables from an in-line X-ray inspection machine [Adler, paragraphs 2, 30]. It would have been obvious to one of ordinary skill in the art to use the teachings of Grafstrom, Cvijetnovic, Raveh, and Adler to provide a single analysis engine that would process data and variables from each of the devices of Butor. This would not be outside the capability of an ordinary skilled person in the art especially since the claim does not require input or output of data from one of the devices to another of the devices. That is, the analysis of data and variables for each device may be carried out independent from another device to optimize each individual stage of the manufacturing process. To overcome such an obviousness rejection, Examiner suggests amending the claim to include additional details regarding the processing of data and variables of the devices by the analysis engine such that there exists some interdependence between results of the analysis.
Conclusion
13. 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.
14. Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALVIN H TAN whose telephone number is (571)272-8595. The examiner can normally be reached M-F 10AM-6PM.
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, Scott Baderman can be reached at 571-272-3644. 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.
/ALVIN H TAN/Primary Examiner, Art Unit 2118
Prosecution Insights