DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
IDS filed 1/13/2025 is being considered by the examiner
Claim Objections
Claims 4 and 17 are objected to because of the following informalities:
Claim 4 recites, "associated with the control line [line 1]". The examiner suggests, "associated with a control line".
Claim 17 recites, "associated with the control line [line 1]". The examiner suggests, "associated with a control line".
Appropriate correction is required.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 2, 4-6, 13-15, 17-21 are rejected under 35 U.S.C. 103 as being unpatentable over Arabshahi et al. [U.S. Pub. 2021/0116896] ("Arabshahi") in view of Freese et al. [U.S. Pub. 2023/0400508] ("Freese").
With regard to claim 1, Arabshahi teaches a method, comprising:
obtaining, by a processing device ("one or more processors to perform operations [par. 0005]"), current sensor data associated with a sensor of a substrate manufacturing system ("monitoring a plurality of sensor outputs while the wafer recipe is executed [par. 0039]" and "feeding data through one or more pattern recognition models [par. 0041]");
determining a ("the outputs of the sensors are forwarded to multiple models simultaneously for processing … each model may be trained to recognize a particular pattern of sensor outputs, or 'fingerprint,' that indicates a fault is likely being generated in the semiconductor wafer [par. 0042]");
responsive to determining that the ("each of the plurality of models may be trained to identify when conditions associated with the semiconductor processing system indicate that a fault is likely to be formed [par. 0042]" and "as soon as conditions inside the processing chamber evolve into a fault condition, the abnormal output 708 may change to indicate the presence of a fault [par. 0058];" it is implied that the decision is based on a threshold criterion, e.g., similar to how thresholds are used in [fig. 3] to determine a fault, the models that provide outputs indicating abnormal or normal conditions must be based on a threshold criterion), performing at least one of generating an alert or performing a corrective action ("the fault output may also include writing a fault indication in a process log, sending a record of the sensor measurements to a data store for analysis and/remodel training, stopping the process being executed by the wafer recipe, sounding an alarm, and/or any other method of alerting users and/or other systems of the possible fault condition [par. 0044]").
Although Arabshahi teaches using various models and techniques to analyze the current sensor data, including determining changing trajectory of sensor values [par. 0053], Arabashi does not explicitly teach determining a slope value.
In an analogous art (substrate processing), Freese teaches determining a slope value ("iteratively determining the slope of the signal SIG is going to cross the alarm limit Sa [par. 0197]" and "estimating when the alarm limit will be met by the signal SIG based on the slope of the signal SIG [par. 0204]").
It would have been obvious to one having ordinary skill in the art at the time of filing the invention to have included Freese's teachings of determining slope, with Arabshahi's teachings of models to detect patterns, for the benefit of quickly identifying when sensor values are projected to exceed a threshold.
With regard to claim 2, the combination above teaches the method of claim 1. Arabshahi in the combination further teaches wherein a slope of the fault detection limit comprises a first value for a first duration of a production run and a second value for a second duration of the production run (see [fig. 3] where thresholds 302d and 304d have different flops during different durations of the production run).
Arabshahi further teaches, "the wafer recipe may include setpoint values and associated times that control how environmental conditions change throughout the processing step. Wafer recipes may include a time series of setpoint temperatures, a time series of setpoint pressures, a time series of gas flow rates into the processing chamber, a time series of gas flow ratios, and/or setpoints for any other environmental condition that may be controlled in the processing chamber [par. 0038]."
It would have been obvious to one having ordinary skill in the art at the time of filing the invention to have included the concept of thresholds that vary during the production run as taught by Arabshahi, into the models as taught by Arabshahi, for the benefit of determining when a fault has occurred depending on values expected during certain time periods.
With regard to claim 4, the combination above teaches the method of claim 1. Freese in the combination further teaches determining that a y-intercept value associated with the control line satisfies a further threshold criterion associated with the fault detection limit ("iteratively determining the slope of the signal SIG is going to cross the alarm limit Sa [par. 0197]" and "estimating when the alarm limit will be met by the signal SIG based on the slope of the signal SIG [par. 0204]"); and
responsive to determining that the y-intercept value satisfied a further threshold criterion associated with the fault detection limit, performing at least one of generating the alert or performing the corrective action (" An alarm may be generated when the signal SIG crosses the alarm limit Sa [par. 0172]").
With regard to claim 5, the combination above teaches the method of claim 1. Freese in the combination further teaches wherein the slope value is updated in response to receiving additional sensor data ("iteratively determining the slope of the signal SIG is going to cross the alarm limit Sa [par. 0197]").
With regard to claim 6, the combination above teaches the method of claim 1. Freese in the combination teaches the method further comprising:
obtaining, a plurality of datasets each comprising sensor output data from a respective sensor of a plurality of sensors each associated with a corresponding process chamber of a plurality of process chambers ("the HI module 230 also performs aggregation, which may be local and/or semi-local based, station based, device based, module based, processing module based, and/or tool based. The aggregation may be for a group of similar and/or different sensors, related sensors and/or unrelated sensors, etc. The HI module 230 selects lowest correlation and/or aggregation values, as further described below. The HI module 230 monitors distributions, means, standard deviations, and shifts in parameters and HI values. The HI module 230 correlates aggregation values for: the same components, devices, modules, sub-systems, processing stations; and values for different components, devices, modules, sub-systems, processing stations. In some embodiments, the HI module 230 evaluates and correlates parameters and aggregation values to provide health index scoring, which may include comparing aggregation values and selecting a lowest aggregation value [par. 0086]" and "to and from chambers of substrate processing stations [par. 0067]");
combining the plurality of datasets into an aggregate dataset ("the HI module 230 also performs aggregation, which may be local and/or semi-local based, station based, device based, module based, processing module based, and/or tool based. The aggregation may be for a group of similar and/or different sensors, related sensors and/or unrelated sensors, etc. The HI module 230 selects lowest correlation and/or aggregation values, as further described below. The HI module 230 monitors distributions, means, standard deviations, and shifts in parameters and HI values [par. 0067]");
generating a distribution of the aggregate dataset ("the HI module 230 may generate distributions of the sensor data [par. 0103]"); and
identifying the fault detection limit based on a deviation value generated from the distribution ("an HI value may be generated based on the percentage of the parameter distribution within the HI boundaries ... The corresponding HI value decreases as the standard deviation increases [par. 0108]").
Freese further teaches, "Due to the numerous sensors, complexity of the tool, and the interrelationships between features of the tool, it can be difficult to identify, locate and determine what is causing an alarm condition [par. 0068]" and
Arabshahi in the combination further teaches, "For example, a temperature sensor and a pressure sensor may both be within a normal operating envelope, but the combination of conditions between the temperature and/or pressure may cause a fault. Monitoring individual sensors for these conditions may necessarily miss these types of fault conditions. In contrast, a model-based process holistically captures and analyzes all of the sensor outputs as inputs to the model analysis. Thus small variations and/or process-related correlations of outputs between different sensors may be recognized by a model where they would be missed by existing methods [par. 0065]".
It would have been obvious to one having ordinary skill in the art at the time of filing the invention to have included Freese's teachings of aggregating sensor data to determine fault limits, with the teachings of Arabshahi, for the benefit determining when a fault occurs based on multiple sensors.
With regard to claim 13, the combination above teaches the method of claim 1. Arabshahi in the combination teaches the method further comprising:
responsive to determining that a duration of a production run satisfies a further threshold criterion ("determining whether a fault is present in the semiconductor wafer [par. 0046]" and "This fault may be detected at any subsequent point in the wafer manufacturing process [par. 0047]"), updating a value associated with the fault detection limit ("After determining that these conditions caused a fault in the semiconductor wafer, the same conditions can be used to begin training a new model to recognize when such conditions exist in the future [par. 0047]").
With regard to claim 14, the combination above teaches claim 1. Claim 14 recites limitations having the same scope as those pertaining to claim 1; therefore, claim 14 is rejected along the same grounds as claim 1.
Claim 14 differs from claim 1 where claim 14 recites the additional elements (which Arabshahi in the combination teaches) of a memory device and a processing device coupled to the memory device to perform the operations ("a non-transitory machine-readable medium may include instructions that, when executed by one or more processors, cause the one or more processors to perform operations [par. 0004]").
With regard to claims 15 and 17-19, the combination above teaches claims 2 and 4-6. Claims 15 and 17-19 recite limitations having the same scope as those pertaining to claims 2 and 4-6, respectively; therefore, claims 15 and 17-19 are rejected along the same grounds as claims 2 and 4-6.
With regard to claim 20, the combination above teaches claim 1. Claim 20 recites limitations having the same scope as those pertaining to claim 1; therefore, claim 20 is rejected along the same grounds as claim 1.
Claim 20 differs from claim 1 where claim 20 recites the additional elements (which Arabshahi in the combination teaches) of a non-transitory computer-readable storage medium comprising instructions that, when executed by a processing device operatively coupled to a memory, performs operations ("a non-transitory machine-readable medium may include instructions that, when executed by one or more processors, cause the one or more processors to perform operations [par. 0004]").
With regard to claim 21, the combination above teaches claim 2. Claim 21 recites limitations having the same scope as those pertaining to claim 2; therefore, claim 21 is rejected along the same grounds as claim 2.
Claims 3 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Arabshahi in view of Freese further in view of Kanada et al. [U.S. Pat. 10,222,791] ("Kanada").
With regard to claim 3, the combination of Arabshahi and Freese teaches the method of claim 1. Arabshahi in the combination teaches the method further comprising:
determining a control line based on the current sensor data ([fig. 3] and "One method of detecting faults that may be used during the manufacturing process is to monitor environmental conditions in the processing chamber and compare them to desired thresholds throughout the manufacturing process. FIG. 3 illustrates graphs 300 showing various sensor outputs 306 during a stage of a manufacturing process [par. 0035];" where the model based approach still uses time series data of sensor values: "when the wafer recipe includes a time series of setpoint temperatures, the plurality of sensors may include one or more temperature sensors that are located around the processing chamber to measure the resulting temperatures in the chamber as they are controlled by the wafer recipe [par. 0039]");
determining a projected control line based on the control line and the slope value (Freese: "iteratively determining the slope of the signal SIG is going to cross the alarm limit Sa [par. 0197]" and "estimating when the alarm limit will be met by the signal SIG based on the slope of the signal SIG [par. 0204]").
Although Freese in the combination teaches display various plotted data [par. 0088], the combination does not explicitly teaches displaying the projected control line on a graphical user interface.
In the same field of endeavor (displaying sensor data), Kanada teaches displaying a projected control line on a graphical user interface ("The operator is provided with a displayed graph showing the received sensor signals, estimated sensor signals for the future, the action margin time and an action threshold value corresponding to the next action. Accordingly, the operator can readily grasp how much time is available to perform the next action before the received sensor signals are predicted to exceed the action threshold value [abstract]").
It would have been obvious to one having ordinary skill in the art at the time of filing the invention to have taken the projected sensor data as taught by Freese, and displayed the projected data as taught by Kanada, for the benefit of informing an operator whether the projected data will reach a set threshold.
With regard to claim 16, the combination above teaches claim 3. Claim 16 recites limitations having the same scope as those pertaining to claim 3; therefore, claim 16 is rejected along the same grounds as claim 3.
Allowable Subject Matter
Claims 7-12 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.
The following is a statement of reasons for the indication of allowable subject matter:
Arabshahi et al. [U.S. Pub. 2021/0116896] teaches a method of detecting and classifying anomalies during semiconductor processing including executing a wafer recipe; monitoring sensor outputs from sensors that monitor conditions associated with the semiconductor processing system; providing the sensor outputs to models trained to identify when the conditions associated with the semiconductor processing system indicate a fault in the semiconductor wafer; receiving an indication of a fault from at least one of the models; and generating a fault output in response to receiving the indication of the fault.
Freese et al. [U.S. Pub. 2023/0400508] teaches where a health monitoring, assessing and response system includes an interface and a controller. The interface is configured to receive a signal from a sensor disposed in a substrate processing system. The controller includes a health index module. The health index module is configured to perform an algorithm including: obtaining a window and a boundary threshold; monitoring the signal output from the sensor; determining whether the signal has crossed the boundary threshold; updating a health index component, where the health index component is a binary value and transitioned between HIGH and LOW values in response to the signal crossing the boundary threshold; and generating a health index value based on the health index component
Pham et al. [U.S. Pub. 2021/0066141] teaches anomaly detection and remedial recommendation techniques for improving the quality and yield of microelectronic products. In one aspect, a method for quality and yield improvement via anomaly detection includes: collecting time series sensor data during individual steps of a semiconductor manufacturing process; calculating anomaly scores for each of the individual steps using a predictive model; and implementing changes to the semiconductor manufacturing process based on the anomaly scores.
Yun et al. [U.S. Pat. 8,989,888] teaches a method for automatically detecting fault conditions and classifying the fault conditions during substrate processing. The method includes collecting processing data by a set of sensors during the substrate processing. The method also includes sending the processing data to a fault detection/classification component. The method further includes performing data manipulation of the processing data by the fault detection/classification component. The method also includes executing a comparison between the processing data and a plurality of fault models stored within a fault library. Each fault model of the plurality of fault models represents a set of data characterizing a specific fault condition.
Ho et al. [KR 20140031075 A] teaches methods of monitoring the status of a tool including a system for implementing a tool status monitoring function. The exemplary method includes: a step of receiving data related to a process performed on a wafer by an integrated circuit manufacturing process tool; and a step of monitoring the status of the integrated circuit manufacturing process tool using the data. The monitoring step includes a step of evaluating the data based on abnormality determination criteria, abnormality filtering criteria, and an abnormality threshold value, in order to determine whether the data meets a warning threshold value condition.
While the prior art of record teaches obtaining data from a plurality of different sensors, including performing aggregation and determining distribution between groups of similar sensors, different sensors, related sensors, and unrelated sensors [see Freese et al.] the prior art of record fails to teach or suggest, alone or in combination, obtaining output data associated with a sensor of the plurality of sensors; generating a first distribution based on the output data and a set of time values;
generating a second distribution based on the output data and a set of tool-life values; generating a set of coefficients of variations based on the first distribution and the second distribution; generating a set of correlation coefficients based on the first distribution and the second distribution; and
responsive to the set of coefficients of variations satisfying a first threshold criterion, and the correlation coefficients satisfying a second threshold criterion, assigning the sensor to a group, in combination with the limitations of base claim 1 and intervening claim 6.
Claims 8-12 are objected by virtue of their dependency.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to VINCENT W CHANG whose telephone number is (571)270-1214. The examiner can normally be reached (M-F) 10:00 am - 6:00 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, Mohammad Ali can be reached at 571-272-4105. 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.
/VINCENT WEN-LIANG CHANG/
Examiner
Art Unit 2119
/MOHAMMAD ALI/Supervisory Patent Examiner, Art Unit 2119