Information processing method, information processing device, and computer program

Abstract

Provided are an information processing method and the like that, when an abnormality occurs in a substrate processing device, facilitate identification of a sensor that has produced abnormal measurement data. An information processing device according to the present embodiment acquires measurement data measured in time series by a plurality of sensors provided in a substrate processing device. Furthermore, the information processing device calculates a degree of abnormality for each time for each sensor on the basis of the acquired measurement data, and identifies an abnormality ranking of each sensor on the basis of the degree of abnormality for each time for each sensor. Then, the information processing device generates a chart indicating a variation over time in the degree of abnormality of each sensor on the basis of the identified abnormality ranking.

Description

Information Processing Method, Information Processing Apparatus, and Computer Program

[0001] The present disclosure relates to an information processing method, an information processing apparatus, and a computer program.

[0002] In Patent Document 1, a graph display system that visualizes multidimensional time-series data using a heat map has been proposed.

[0003] Japanese Unexamined Patent Application Publication No. 2017-16545

[0004] The present disclosure provides an information processing method and the like that facilitate identification of a sensor whose measurement data was abnormal when an abnormality occurred in a substrate processing apparatus.

[0005] An information processing method according to an embodiment acquires measurement data measured in time series by a plurality of sensors provided in a substrate processing apparatus, calculates an abnormality degree for each sensor for each time based on the acquired measurement data, identifies an abnormality ranking for each sensor based on the abnormality degree for each sensor for each time, and generates a chart showing a time change in the abnormality degree for each sensor based on the identified abnormality ranking. A computer executes the processing.

[0006] According to the present disclosure, it becomes easy to identify a sensor whose measurement data was abnormal when an abnormality occurred in a substrate processing apparatus.

[0007] It is an explanatory diagram showing a configuration example of a substrate processing apparatus according to this embodiment. It is a block diagram showing a configuration example of a higher-level module. It is an explanatory diagram showing an example of a sensor value DB and a past case DB. It is a block diagram showing a configuration example of a lower-level module. It is a flowchart showing an example of a processing procedure performed by the higher-level module of the substrate processing apparatus according to this embodiment. It is an explanatory diagram showing an example of a screen displaying a heat map. It is an explanatory diagram showing an example of a group DB. It is a flowchart showing an example of a procedure for heat map generation processing in Embodiment 2. It is an explanatory diagram showing an example of a screen displaying a heat map.

[0008] The information processing methods, information processing devices, and computer programs described herein will be specifically explained with reference to drawings illustrating embodiments thereof. However, this disclosure is not limited to the following examples, and is intended to include all modifications within the meaning and scope of the claims, as indicated by the claims.

[0009] (Embodiment 1) Figure 1 is an explanatory diagram showing an example configuration of a substrate processing apparatus according to this embodiment. The substrate processing apparatus 1 according to this embodiment is an apparatus that performs various substrate processing on a semiconductor substrate (wafer), such as film deposition by CVD (Chemical Vapor Deposition), etching, cleaning, ashing, coating and development, or sputtering. The substrate processing performed by the substrate processing apparatus 1 may target semiconductor substrates, as well as other types of substrates such as glass substrates, mask substrates, FPD (Flat Panel Display) substrates, etc. The substrate processing apparatus 1 is composed of a combination of multiple modules. In this embodiment, a module is a component, part, or set of devices realized by dividing hardware resources that perform multiple functions necessary for the substrate processing apparatus 1, such as temperature control, gas flow rate control, and pressure control, into a manner that allows for independent replacement or substitution for each function. A module may also be called by names such as unit, component, or assembly.

[0010] The substrate processing apparatus 1 according to this embodiment comprises, for example, one upper-level module 2 and a plurality of lower-level modules 3. The upper-level module 2 and the plurality of lower-level modules 3 are connected via communication lines or the like within the substrate processing apparatus 1, and can send and receive data from each other. In this figure, the upper-level module 2 and the plurality of lower-level modules 3 are connected in a bus-type network configuration, but this is not the only option, and any network configuration may be adopted. For example, the plurality of lower-level modules 3 may be connected in a star-type network configuration with the upper-level module 2 at the center. In addition to the configuration included in the substrate processing apparatus 1, the upper-level module 2 may be an external device different from the substrate processing apparatus 1, such as a server device.

[0011] The lower-level module 3 includes, for example, an ESC (Electric Static Chuck) temperature control module, a gas flow rate control module, and a pressure control module. The substrate processing apparatus 1 also has at least one substrate processing chamber for processing one or more substrates, and the ESC temperature, the gas flow rate supplied to the substrate processing chamber, the pressure inside the substrate processing chamber, etc., are controlled by these lower-level modules 3.

[0012] The lower module 3 is equipped with various sensors 4, or the sensors 4 are connected via signal lines, etc. The sensors 4 measure various values ​​such as temperature or pressure related to substrate processing, and the lower module 3 acquires the values ​​measured by the sensors 4 (hereinafter referred to as sensor values). In addition to temperature sensors and pressure sensors, the sensors 4 may also be various other sensors such as sensors that measure humidity, type and concentration of gas, fluid flow rate, voltage, current, charge, and light including spectral data. The lower module 3 is equipped with various controlled devices 5, or the controlled devices 5 are connected via signal lines, etc. The controlled devices 5 are, for example, devices installed in substrate processing such as actuators, valves, and edge rings. The lower module 3 is a module that controls the controlled devices 5 based on the sensor values ​​acquired from the sensors 4. However, the multiple lower modules 3 provided in the substrate processing apparatus 1 may include modules that acquire sensor values ​​from the sensors 4 but do not control the controlled devices 5, or modules that control the controlled devices 5 without acquiring sensor values ​​from the sensors 4, etc.

[0013] The upper module 2 communicates with multiple lower modules 3 and obtains various information from each lower module 3. Based on the obtained information, the upper module 2 controls the operation of the multiple lower modules 3 by transmitting control instructions to each lower module 3. In this way, the upper module 2 can operate the lower modules 3 according to setting information (recipe), such as procedures or conditions for board processing, which have been created in advance by the user, and perform various board processing on the board. The upper module 2 may transmit a control instruction to a lower module 3 based on information obtained from that lower module 3, or it may transmit a control instruction to another lower module 3 based on information obtained from a certain lower module 3. Furthermore, the upper module 2 may transmit control instructions to multiple lower modules 3 based on information obtained from one lower module 3, or it may transmit control instructions to one or more lower modules 3 based on information obtained from multiple lower modules 3. In this embodiment, the board processing apparatus 1 has a two-tiered configuration of an upper module 2 and lower modules 3, but it is not limited to this, and may have a three-tiered configuration of an upper module, an intermediate module and lower modules, or a configuration of four or more tiers.

[0014] While the substrate processing apparatus 1 is performing substrate processing, each lower module 3 acquires sensor values ​​by periodically sampling the measured values ​​of the sensor 4 and transmits the acquired sensor values ​​to the upper module 2. In this embodiment, instead of transmitting the periodically acquired sensor values ​​themselves to the upper module 2, the lower module 3 may generate processed values ​​obtained by performing appropriate calculations on the sensor values, for example, a summary value obtained by combining multiple sensor values ​​through appropriate calculations, and transmit the generated summary value to the upper module 2. The summary value can be various values ​​such as the average, maximum, minimum, or median of multiple sensor values. The lower module 3 transmits the sensor values, or the summary value calculated based on multiple sensor values, to the upper module 2 at predetermined intervals. Hereinafter, the sensor values ​​and summary values ​​will be collectively referred to as "sensor values".

[0015] In the substrate processing apparatus 1 according to this embodiment, the upper module 2 gives control instructions to a plurality of lower modules 3 in order to proceed with substrate processing based on settings given by the user, and acquires sensor values ​​(measurement data measured in time series by each sensor 4) transmitted from the lower modules 3 to determine the status of substrate processing. In the substrate processing apparatus 1 according to this embodiment, the upper module 2 determines the status of substrate processing, in this embodiment whether it is normal or abnormal, based on the plurality of sensor values ​​that can be acquired from the plurality of lower modules 3. The upper module 2 accumulates the sensor values ​​transmitted from each lower module 3 at predetermined intervals, and if it determines that an abnormality has occurred in the substrate processing, it identifies the sensor value (sensor 4) in which the abnormality has occurred and performs a process to generate a heat map (chart) showing the degree of abnormality in the identified sensor value.

[0016] Figure 2 is a block diagram showing one example configuration of the upper-level module 2. The upper-level module 2 (information processing device) of the substrate processing device 1 according to this embodiment is configured to include a processing unit 21 (control unit), a storage unit 22, a communication unit 23, etc. The processing unit 21 is configured to use an arithmetic processing device such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), GPU (Graphics Processing Unit), or quantum processor, ROM (Read Only Memory), and RAM (Random Access Memory), etc. The processing unit 21 reads and executes a program 22a stored in the storage unit 22 to perform substrate processing by controlling the operation of multiple lower-level modules 3, and various processing such as generating a heat map of the abnormality level of sensor values ​​when an abnormality occurs.

[0017] The storage unit 22 is configured using a large-capacity storage device such as a hard disk or an SSD (Solid State Drive). The storage unit 22 stores various programs executed by the processing unit 21, and various data necessary for the processing of the processing unit 21. In this embodiment, the storage unit 22 stores the program 22a executed by the processing unit 21, the sensor value DB 22b, and the past case DB 22c.

[0018] Figure 3 is an explanatory diagram showing an example of the sensor value DB 22b and the past case DB 22c. The sensor value DB 22b is a database that stores the sensor values ​​of each sensor 4 acquired from the lower-level module 3 during board processing. The sensor value DB 22b is prepared, for example, for each board processing and stored in the storage unit 22 in association with the board processing ID assigned to each board processing. The sensor value DB 22b shown in Figure 3 includes a sensor ID column, a sensor value column, a judgment result column, etc., and stores information of each sensor 4 in association with the sensor ID. The sensor ID column stores the identification information (sensor ID) assigned to each sensor 4. The sensor value column stores time-series data that associates the sensor value at each time with the elapsed time from the start of board processing, for example. Note that each time may be the date and time (measurement date and time) indicated by the timestamp assigned when each sensor 4 measures the sensor value, or when each lower-level module 3 acquires the sensor value from the sensor 4. The judgment result column contains information indicating the status of the substrate processing determined by the processing unit 21 based on the sensor values ​​from each sensor 4, and stores, for example, "normal" or "abnormal". The sensor values ​​stored in the sensor value DB 22b are stored each time the processing unit 21 obtains the sensor values ​​from each sensor 4 from each lower module 3, and the judgment result is stored each time the processing unit 21 performs the judgment processing.

[0019] The Past Cases DB22c is a database that stores information about abnormalities that have occurred in the past with the substrate processing apparatus 1 and apparatus of the same type as the substrate processing apparatus 1. The Past Cases DB22c shown in Figure 3 includes a Case ID column, a Sensor Value column for the 1st to Nth sensors (where N is a natural number, e.g., 10), a Type / Cause column, a Countermeasure column, etc., and stores information about each abnormal case in association with the Case ID. The Case ID column stores identification information (Case ID) assigned to each abnormal case that has occurred in the past. The Sensor Value column for the 1st to Nth sensors contains the sensor values ​​measured by each sensor 4 during the substrate processing in which each abnormal case occurred, for each sensor 4 identified as the 1st to Nth sensors. For example, it stores time-series data that associates the sensor values ​​at each time with the elapsed time from the start of the substrate processing. The 1st to Nth sensors are identified, for example, in the order in which the abnormalities occurred, or in the order of the degree of abnormality relative to the normal value in the substrate processing in which the abnormality occurred. Furthermore, the sensor values ​​of each sensor 4 may include not only the sensor values ​​measured during the substrate processing in which the abnormality occurred, but also feature quantities extracted from the sensor values. The Type / Cause column stores a name or identification information that identifies the type or cause of the abnormality. The type and cause of the abnormality may be identified by the processing unit 21 through a determination process based on the sensor values ​​of each sensor 4 (first sensor to nth sensor), or they may be identified by a user operating the substrate processing device 1 and stored in the past case DB 22c. The Countermeasure column stores the work content or control method for each module to resolve each abnormality. The countermeasure is identified, for example, by a user operating the substrate processing device 1 and stored in the past case DB 22c.

[0020] In the higher-level module 2 shown in Figure 2, the program (computer program, program product) 22a is provided in the form of a recording medium 2a such as a memory card or optical disc, and the substrate processing device 1 reads the program 22a from the recording medium 2a and stores it in the storage unit 22 of the higher-level module 2. However, the program 22a may also be written to the storage unit 22 during the manufacturing stage of the substrate processing device 1 or the higher-level module 2, for example. Alternatively, the program 22a may be distributed by a remote server device or the like and acquired by the substrate processing device 1 via communication. For example, the program 22a may be read from the recording medium 2a by a writing device and written to the storage unit 22 of the higher-level module 2 of the substrate processing device 1. The program 22a may be provided in the form of a distribution via a network, or it may be provided in the form of a recording medium 2a.

[0021] The communication unit 23 is connected to a plurality of lower-level modules 3 via a bus (communication line) provided on the substrate processing device 1, and transmits and receives data with these plurality of lower-level modules 3. In this embodiment, the communication unit 23 receives sensor value data transmitted from the lower-level modules 3 and provides the received sensor values ​​to the processing unit 21. The communication unit 23 also transmits control instructions and the like given by the processing unit 21 to one or more lower-level modules 3.

[0022] Furthermore, in the upper-level module 2 of the substrate processing apparatus 1 according to this embodiment, the processing unit 21 reads and executes the program 22a stored in the memory unit 22, thereby realizing the control processing unit 21a, the heat map generation unit 21b, and the past case identification unit 21c, etc., as software-based functional units. In this figure, the functional units of the processing unit 21 are shown as functional units related to substrate processing control, heat map generation, and past case identification, while functional units related to other processing are omitted from the illustration.

[0023] The control processing unit 21a communicates with the lower module 3 via the communication unit 23, receives sensor values ​​periodically transmitted from the lower module 3, and stores them in the sensor value DB 22b. Based on the multiple sensor values ​​obtained from the multiple lower module 3, the control processing unit 21a determines the state of the board processing, determines the operation that each lower module 3 should perform based on the determined state and the user's settings, and sends a control instruction to each lower module 3 to perform the determined operation. For example, the control processing unit 21a gives a control instruction to a lower module 3 that controls an actuator as a controlled device 5, along with a target value for the displacement amount of the actuator, to operate the actuator. In response to this operation instruction, the lower module 3 performs control to operate the actuator so that the displacement amount becomes the target value.

[0024] In this embodiment, if the control processing unit 21a determines that an abnormality has occurred in the substrate processing based on the sensor values ​​obtained from the lower module 3, it identifies the sensor value (sensor 4) where the abnormality has occurred and causes the heat map generation unit 21b to generate a heat map of the degree of abnormality of the sensor value. At this time, the control processing unit 21a may, for example, temporarily or forcibly stop the substrate processing before causing the heat map generation unit 21b to generate the heat map. Also, if the control processing unit 21a determines that an abnormality has occurred in the substrate processing, it causes the past case identification unit 21c to identify past abnormality cases similar to the abnormality that has occurred.

[0025] The heatmap generation unit 21b, when the control processing unit 21a determines that an abnormality has occurred in the substrate processing based on the sensor values ​​obtained from the lower module 3, identifies the sensor value (sensor 4) where the abnormality has occurred and generates a heatmap of the abnormal value for the identified sensor value. In this embodiment, the heatmap generation unit 21b calculates the degree of abnormality relative to the normal value by comparing the sensor value of each sensor 4 obtained during the substrate processing determined to be abnormal with the sensor value of each sensor 4 during the substrate processing determined to be normal (hereinafter referred to as the normal value). The heatmap generation unit 21b calculates the degree of abnormality at a predetermined interval based on the sensor values ​​periodically obtained from the lower module 3. The calculation period for the degree of abnormality may be, for example, the longest period among the periods for obtaining sensor values ​​from each lower module 3, in which case the load on the calculation process can be reduced by decreasing the calculation frequency. Note that the calculation period for the degree of abnormality does not have to be the longest period. Based on the degree of abnormality relative to the sensor value of each sensor 4 calculated at the predetermined interval, the heatmap generation unit 21b determines whether or not an abnormality has occurred in each sensor value (each sensor 4). The criteria for determining whether or not there is an abnormality are, for example, predetermined, and different criteria values ​​may be set for each sensor 4. Based on the determination results for whether or not there is an abnormality for abnormal values ​​calculated at a predetermined period, the heat map generation unit 21b identifies whether or not an abnormality has occurred in each sensor 4, and if an abnormality has occurred, the time (occurrence timing) at which it occurred. Based on the degree of abnormality of the sensor value of each sensor 4 calculated at a predetermined period, or the occurrence timing of the abnormality of each sensor 4, the heat map generation unit 21b identifies an abnormality ranking for each sensor 4. Based on the abnormality ranking, the heat map generation unit 21b identifies a predetermined number (for example, 10) of sensors 4 that are at the top of the ranking and generates a heat map of the degree of abnormality of the sensor values ​​of the identified sensors 4. The heat map generation unit 21b displays the generated heat map on the display unit of the substrate processing device 1, etc.

[0026] The past case identification unit 21c identifies similar past cases from past cases stored in the past case DB 22c based on the sensor values ​​of each sensor 4 in the heat map generated by the heat map generation unit 21b. For example, the past case identification unit 21c identifies an unusually similar past case represented in the heat map generated by the heat map generation unit 21b based on each sensor 4 identified as the first to the Nth sensor, the sensor value of each sensor 4, or the feature quantity of the sensor value. The past case identification unit 21c displays the information of the identified past case on the display unit of the substrate processing device 1.

[0027] Figure 4 is a block diagram showing an example configuration of the lower module 3. The lower module 3 of the substrate processing apparatus 1 according to this embodiment is configured to include a processing unit 31, a storage unit 32, a communication unit 33, an input unit 34, and an output unit 35, etc. The processing unit 31 is configured to use a processing unit such as a CPU or MPU, ROM and RAM, etc. The processing unit 31 reads and executes a program 32a stored in the storage unit 32, and performs various processes such as transmitting sensor values ​​measured by the sensor 4 to the upper module 2, and controlling the operation of the controlled device 5 in response to instructions from the upper module 2.

[0028] The storage unit 32 is configured using, for example, non-volatile memory such as flash memory or EEPROM (Electrically Erasable Programmable Read Only Memory). The storage unit 32 stores various programs and data, such as the program 32a executed by the processing unit 31. In this embodiment, the program (computer program, program product) 32a is provided in the form of being recorded on a recording medium 3a such as a memory card or optical disc, and the substrate processing device 1 reads the program 32a from the recording medium 3a and stores it in the storage unit 32 of the lower module 3. However, the program 32a may also be written to the storage unit 32 during the manufacturing stage of the substrate processing device 1 or the lower module 3, for example. Alternatively, the program 32a may be distributed by a remote server device or the like and acquired by the substrate processing device 1 via communication. For example, the program 32a may be read from the recording medium 3a by a writing device and written to the storage unit 32 of the lower module 3 of the substrate processing device 1. The program 32a may be provided in the form of distribution via a network, or it may be provided in the form of being recorded on the recording medium 3a.

[0029] The communication unit 33 is connected to the upper module 2 via a bus (communication line) provided on the substrate processing device 1, and performs data transmission and reception with the upper module 2. In this embodiment, the communication unit 33 receives control instructions and the like transmitted from the upper module 2 and provides the received control instructions and the like to the processing unit 31. The communication unit 33 also transmits sensor values ​​and the like provided by the processing unit 31 to the upper module 2. In this embodiment, multiple lower modules 3 are connected to the bus provided on the substrate processing device 1, and data transmission and reception may occur between multiple lower modules 3.

[0030] The input unit 34 is connected to the sensor 4 via signal lines or the like. The input unit 34 samples and acquires the signal (measurement result by the sensor 4) output by the sensor 4 at a predetermined sampling period and provides the acquired sensor value to the processing unit 31. Multiple sensors 4 may be connected to the input unit 34, and the input unit 34 may acquire sensor values ​​from multiple sensors 4 at individual sampling periods. The output unit 25 is connected to the controlled device 5 via signal lines or the like. The output unit 35 outputs a control signal to the controlled device 5 based on the control value provided by the processing unit 31, and operates the controlled device 5. Multiple controlled devices 5 may be connected to the output unit 35, and the output unit 35 may output individual control signals to multiple controlled devices 5, and individually control the operation of multiple controlled devices 5.

[0031] Furthermore, in the lower module 3 of the substrate processing apparatus 1 according to this embodiment, the processing unit 31 reads and executes the program 32a stored in the storage unit 32, thereby realizing the control processing unit 31a and the sensor value processing unit 31b, etc., as software-based functional units in the processing unit 31. In this figure, the functional units of the processing unit 31 that relate to control of substrate processing and transmission of sensor values ​​are shown, while functional units related to other processing are omitted from the illustration.

[0032] The control processing unit 31a communicates with the higher-level module 2 via the communication unit 33 and performs processing to control the operation of the controlled device 5 connected to the output unit 35 in accordance with the control instructions transmitted from the higher-level module 2. The control processing unit 31a determines the state of the controlled device 5 based on the sensor value measured by the sensor 4, and calculates a control value for the controlled device 5 based on the target value given as a control instruction from the higher-level module 2 and the current state of the controlled device 5. The control value may be, for example, a voltage value or current value applied to the controlled device 5, but any value is acceptable.

[0033] The sensor value processing unit 31b performs the process of transmitting the sensor values ​​of the sensor 4 acquired by the input unit 34 to the higher-level module 2 from the communication unit 33 at a predetermined transmission cycle. For example, the sensor value processing unit 31b samples and acquires the measurement results of the sensor 4 at a predetermined cycle (e.g., a 10ms cycle or a 100ms cycle) and transmits the acquired sensor values ​​to the higher-level module 2. Alternatively, the sensor value processing unit 31b may acquire sensor values ​​at a 10ms cycle, calculate the average, maximum, minimum, or median of 10 sensor values ​​as a summary value at a 100ms cycle, and transmit the calculated summary value to the higher-level module 2 at a 100ms cycle. Note that the values ​​of 10ms, 100ms, 10 times, etc. when calculating the summary value are just examples and are not limited to these.

[0034] The following describes the process of generating a heat map showing the degree of abnormality of sensor values ​​that are abnormal (sensor values ​​that are not normal) based on the sensor values ​​measured by each sensor 4 during substrate processing in which an abnormality has been determined to have occurred, and the process of identifying past cases similar to the abnormality that has occurred. Figure 5 is a flowchart showing an example of the processing procedure performed by the upper module 2 of the substrate processing apparatus 1 according to this embodiment. The upper module 2 of the substrate processing apparatus 1 according to this embodiment stores in the sensor value DB 22b the sensor values ​​measured by each sensor 4 designated as a monitoring target during the execution of substrate processing, and the state of the substrate processing (normal or abnormal) determined by the processing unit 21 based on the sensor values ​​of each sensor 4. Specifically, the processing unit 21 of the upper module 2 stores the sensor values ​​of each sensor 4 transmitted from the plurality of lower modules 3 in the sensor value DB 22b, and also stores in the sensor value DB 22b the result of determining whether or not an abnormality has occurred in the substrate processing based on the sensor values ​​of each sensor 4.

[0035] In the upper-level module 2 of the substrate processing apparatus 1 according to this embodiment, the processing unit 21 processes the sensor values ​​of each sensor 4 stored in the sensor value DB 22b, which has "abnormal" as the determination result, as the processing target. The processing unit 21 identifies the calculation timing time based on the calculation period of the abnormality degree of the sensor value (S11). The calculation period of the abnormality degree may be, for example, the longest period among the measurement periods of the sensor values ​​of each sensor 4, and the processing unit 21 identifies the calculation timing time of the abnormality degree based on the calculation period. Specifically, the processing unit 21 identifies the time elapsed from the start of substrate processing as the first calculation timing time, and identifies the time elapsed from the previous calculation timing time as subsequent calculation timing times.

[0036] The processing unit 21 reads the sensor value measured at a specified calculation timing time for one sensor 4 from the sensor value DB 22b (S12), and calculates an abnormality score indicating how far the read sensor value is from the normal value (S13). The normal value used to calculate the abnormality score can be, for example, the sensor value of the same sensor 4 measured during substrate processing that was determined to be normal (no abnormality occurred). In this case, the processing unit 21 reads the sensor value of the same sensor 4 measured at the calculation timing time from the sensor value DB 22b where "normal" is stored as the determination result, and calculates the abnormality score relative to the read normal value. The abnormality score can be, for example, the F-value obtained by ANOVA (Analysis of Variance). Note that the abnormality score is not limited to the F-value, as long as it is an index that can represent the distance from the normal value (information indicating how far it is) to that standard.

[0037] The processing unit 21 determines whether the sensor value of the sensor 4 is abnormal based on the calculated abnormality level (S14). For example, a reference value for determining whether a sensor 4 is abnormal is stored in the storage unit 22, and the processing unit 21 reads the reference value of the sensor 4 to be processed from the storage unit 22 and determines whether the calculated abnormality level is abnormal or normal depending on whether it is greater than or equal to the read reference value. The reference value may be different for each sensor 4, or it may be the same value for multiple sensors 4. When calculating the F value by ANOVA as the abnormality level, the processing unit 21 may calculate a P value based on the F value and determine whether the sensor value of the sensor 4 is abnormal based on the P value. In this case, a significance level for the P value is set in advance and stored in the storage unit 22, and the processing unit 21 determines whether the calculated P value is abnormal or normal depending on whether it satisfies the significance level (is less than the significance level). The significance level for the P value may also be different for each sensor 4, or it may be the same value for multiple sensors 4. The processing unit 21 stores the calculated abnormality level and judgment result (whether it is an abnormal value or a normal value) in the storage unit 22 in association with the calculation timing time (S15). The abnormality level and judgment result may also be stored in the sensor value DB 22b of the sensor to be processed.

[0038] The processing unit 21 determines in step S16 whether it has calculated the abnormality level of the sensor values ​​for all sensors 4 whose sensor values ​​are stored in the sensor value DB 22b as monitored targets, based on the calculation timing time identified in step S11. If it determines that it has not calculated the abnormality level of the sensor values ​​(S16: NO), it returns to step S12 and performs the processing in steps S12 to S15 for the unprocessed sensors 4. Note that the processing in steps S12 to S15 may be performed not only for all sensors 4 whose sensor values ​​are stored in the sensor value DB 22b as monitored targets, but also only for sensors 4 that are designated as processing targets among the monitored sensors 4. In this case, in step S16, the processing unit 21 may determine whether it has processed all sensors 4 that are designated as processing targets among the sensors 4 whose sensor values ​​are stored in the sensor value DB 22b. If it determines that it has calculated the abnormality level of the sensor values ​​for all sensors 4 (S16: YES), the processing unit 21 determines whether it has calculated the abnormality level of all sensor values ​​stored in the sensor value DB 22b (S17). If the processing unit 21 determines that it has not calculated the degree of abnormality for all sensor values ​​(S17: NO), it returns to step S11, identifies the next calculation timing time (S11), and performs the processing in steps S12 to S16 for the identified calculation timing time. The processing unit 21 repeats the processing in steps S11 to S17 until it determines that it has calculated the degree of abnormality for all sensor values, thereby enabling the calculation of the degree of abnormality for each sensor value of each sensor 4 at each calculation timing time.

[0039] If the processing unit 21 determines that it has calculated the degree of abnormality for all sensor values ​​(S17: YES), it identifies the abnormality ranking of each sensor 4 in order of the earliest time it was determined to be an abnormal value (i.e., in order of abnormality occurrence) based on the degree of abnormality and the judgment result at each calculation timing time for each sensor 4 stored in the storage unit 22 or the sensor value DB 22b (S18). For each sensor 4, the time at which it was first determined to be an abnormal value is defined as the abnormality occurrence time. Based on the identified abnormality ranking, the processing unit 21 identifies a predetermined number of top-ranking sensors 4 (for example, 10) and generates a heat map representing the degree of abnormality at each calculation timing time calculated for the predetermined number of identified sensors 4 (S19). The sensors 4 for which the heat map is generated may be the sensors 4 at the top of the abnormality ranking, or sensors 4 selected by the user from among the sensors 4 at the top of the abnormality ranking. For example, the information (e.g., sensor name, etc.) of a predetermined number of sensors 4 at the top of the abnormality ranking may be displayed in a list, and a heat map may be generated for the sensors 4 selected by the user from among them. The number of sensors 4 for which the heat map is generated may be arbitrarily changed by the user.

[0040] Figure 6 is an explanatory diagram showing an example of a screen displaying a heatmap. In the heatmap shown in Figure 6, the horizontal axis (first axis) represents time, and the vertical axis (second axis) represents information (e.g., sensor names) of the first to tenth sensors identified in the order of anomaly occurrence. The time shown on the horizontal axis may be, for example, the elapsed time from the start of substrate processing, or it may be the date and time of measurement of the sensor value by each sensor 4. The heatmap represents the degree of anomaly at each calculation timing time for each sensor 4 in a display manner corresponding to the magnitude of each anomaly, showing the change in the degree of anomaly of each sensor 4 over time. In Figure 6, each degree of anomaly is represented by shades of gray, with the darkest gray (black) representing an anomaly above the judgment criterion value for determining whether or not it is an anomaly, but each degree of anomaly may be represented by a different color. Furthermore, if the degree of anomaly of the sensor value of each sensor 4 is an F value, it may be represented in a display manner (shade or color) corresponding to the logarithm (log scalar) of the P value calculated from the F value. In the heatmap shown in Figure 6, the sensor values ​​of each sensor 4 are determined to be abnormal from the time indicated by the arrow, and it can be determined that the time indicated by the arrow is the timing of the abnormality. In this embodiment, the time change in the abnormality level of each sensor is represented using a heatmap, but other charts may also be used. For example, a three-dimensional graph may be used to represent the time change in the abnormality level of each sensor, with the first axis showing time, the second axis showing information of each sensor 4, and the third axis showing the abnormal value of each sensor 4. In this case, the abnormal value of each sensor 4 can be identified based on the abnormal value shown in the third axis direction.

[0041] Furthermore, the processing unit 21 identifies an abnormality ranking for each sensor 4 in descending order of the calculated abnormality degree, based on the abnormality degree of each sensor 4 stored in the storage unit 22 or the sensor value DB 22b (S20). For example, the abnormality ranking based on the abnormality degree may be generated by calculating representative values ​​such as the average, maximum, minimum, or median of the abnormality degree at each calculation timing time for each sensor 4, and then ranking them in descending order of the calculated representative values. In addition, the average value of the abnormality degree at each calculation timing time may be calculated using a simple average (time average), or if an F value is calculated using ANOVA as the abnormality degree, a weighted average of the F values ​​weighted according to the amount of information in the P value at each calculation timing time may be used, or a weighted average of the abnormality degree weighted according to the magnitude of each abnormality may be used. By weighting according to the amount of information in the P value, an index value that emphasizes the abnormality degree at the time of abnormality occurrence can be obtained, and an appropriate ranking can be performed using such an index value.

[0042] The processing unit 21 identifies a predetermined number of sensors 4 at the top (for example, 10) based on the identified anomaly ranking, and generates a heat map (S21) representing the degree of anomaly at each calculated timing time calculated for the identified predetermined number of sensors 4. Step S21 is the same process as step S19, although the sensors 4 to be processed are different. In this case as well, a heat map like the one shown in Figure 6 is generated. The screen in Figure 6 is configured to allow switching between displaying the heat map of sensors 4 identified in order of anomaly occurrence and the heat map of sensors 4 identified in order of decreasing degree of anomaly.

[0043] In the upper-level module 2 according to this embodiment, the processing unit 21 generates a heatmap and then identifies past cases (past anomalies) similar to the anomaly related to the generated heatmap based on the information of the generated heatmap and information on past anomalies stored in the past case DB 22c (S22). For example, the processing unit 21 calculates the similarity between the types of sensors 1 to N, the sensor values ​​or feature quantities of the sensor values ​​of each sensor 4 of the 1st to Nth sensors, as indicated by the anomaly ranking of each sensor 4 used to generate the heatmap, and the information of each case stored in the past case DB 22c (types of sensors 1 to N, sensor values ​​or feature quantities of each sensor 4), and identifies the case with the highest calculated similarity as a similar past case. Similarity can be calculated using, for example, Euclidean distance, Mahalanobis distance, correlation coefficient, cosine similarity, etc., or similarity may be estimated using a learning model constructed by machine learning. For example, a learning model can be used that is composed of a CNN (Convolutional Neural Network) or the like, and is trained to output two similarity scores when the anomaly ranking of each sensor 4 used to generate the heatmap (the type of each sensor 4), the sensor value or feature of the sensor value of each sensor 4, and information on each case stored in the past case DB 22c are input. Alternatively, a class classifier created based on the information on each case stored in the past case DB 22c may be used. In this case, by inputting the type of the first to the Nth sensor indicated by the anomaly ranking of each sensor 4 used to generate the heatmap, and the sensor value or feature of the sensor value of each sensor 4 into the class classifier, similar past cases can be identified by the class classifier.

[0044] The processing unit 21 reads out the information of the past cases (e.g., the type of abnormality, cause, and countermeasure) specified in step S22 from the past case DB 22c, and displays the read-out information of the past cases and the heat map generated in step S19 or S21 on the display unit or the like of the substrate processing apparatus 1 (S23). Thereby, the screen of FIG. 6 is displayed. In the screen of FIG. 6, the heat map in the order of occurrence of abnormalities generated in step S19 is displayed. The screen of FIG. 6 is provided with a “tab in the order of occurrence of abnormalities” tab and a “tab in the order of high abnormality degree”. For example, by switching the selection of the tab by the user, the heat map in the order of occurrence of abnormalities generated in step S19 and the heat map in the order of high abnormality degree generated in step S21 are switched and displayed. When the date and time of occurrence of each case are stored in the past case DB 22c, the processing unit 21 may also read out the date and time of occurrence of similar past cases from the past case DB 22c and display them as shown in FIG. 6. In addition to outputting and displaying on the display unit the screen for displaying the heat map and the information of the past cases, the processing unit 21 may be configured to transmit the screen to a device specified via a network, or may be configured to transmit the screen to a printer connected via a network or directly and print it.

[0045] When the countermeasure read out from the past case DB 22c is a control process for any of the lower-level modules 3, after the process of step S23 or instead of the process of step S23, a control signal related to the control process indicated by the countermeasure is output to the corresponding lower-level module 3 (processing unit). Thereby, the countermeasure by the corresponding lower-level module 3 can be implemented. When the abnormality in progress can be eliminated by the control of the lower-level module 3, it becomes possible for the processing unit 21 to eliminate the abnormality by controlling the lower-level module 3.

[0046] By the above-described processing, in the present embodiment, when a trouble (abnormality) occurs during substrate processing in the substrate processing apparatus 1, the sensor 4 with an abnormal sensor value is specified, and the temporal change in the degree of abnormality of the sensor value of the specified sensor 4 can be visualized by a heat map. The temporal change in the degree of abnormality of the sensor value of each sensor 4 can be visualized, for example, by a line graph showing time on the horizontal axis and the degree of abnormality on the vertical axis. However, in the line graph, the graphs showing the degree of abnormality of each sensor 4 overlap, resulting in poor visibility. In contrast, in the present embodiment, as shown in FIG. 6, the temporal change in the degree of abnormality is displayed for each sensor 4, so the visibility is good, and even when the number of sensors 4 displayed in the heat map increases, the visibility does not deteriorate. Also, it is easy to compare the degrees of abnormality among the respective sensors 4. Abnormalities are generally considered to spread, and the sensor 4 with an earlier abnormal value often relates to the main cause of the abnormality. In the present embodiment, by generating a heat map of the sensor 4 specified by the ranking in the order of occurrence of the abnormality, the sensor 4 highly likely related to the cause of the abnormality can be specified at the top of the ranking. Since it is easy to specify the sensor 4 with an abnormal sensor value by the ranking in the order of occurrence of the abnormality or in the order of high degree of abnormality, it is expected to contribute to specifying the location and cause of the abnormality based on the specified sensor 4.

[0047] The heat map shown in FIG. 6 is configured such that, for example, the time scale shown on the horizontal axis can be changed and can be displayed enlarged (in a state stretched in the time axis direction) to the scale specified by the user. Also, the heat map may be configured to be extracted and enlarged for display for each processing step in the substrate processing.

[0048] As shown in FIG. 6, the present embodiment is configured to separately generate a heat map based on the ranking in the order of occurrence of the abnormality and a heat map based on the ranking in the order of high degree of abnormality, but it may also be configured to generate one heat map. For example, for each sensor 4, an abnormality score corresponding to the ranking in the order of occurrence of the abnormality and an abnormality score corresponding to the ranking in the order of high degree of abnormality are added, and the abnormality ranking may be specified in the order of high (or low) added abnormality score.

[0049] (Embodiment 2) This embodiment describes a configuration in which each sensor 4 provided in the substrate processing apparatus 1 is grouped, and a heat map showing the degree of abnormality of the sensor values ​​of the sensors 4 belonging to each group is generated for each group. The substrate processing apparatus of this embodiment has the same configuration as the substrate processing apparatus 1 of Embodiment 1, so the explanation of the configuration will be omitted. In addition to the configurations shown in Figures 1 to 4, the substrate processing apparatus 1 of this embodiment stores a group DB in the storage unit 22 of the upper module 2.

[0050] Figure 7 is an explanatory diagram showing an example of a group DB 22d. The group DB 22d is a database that stores information about each group into which the sensors 4 provided in the substrate processing apparatus 1 are grouped. The group DB 22d shown in Figure 7 includes a group ID column, a sensor ID column, etc., and stores the sensor IDs of the sensors 4 grouped into each group in association with the group ID. A single sensor 4 may be grouped into multiple groups. Each group may be a group based on the role or function that each sensor 4 performs, or a group based on the lower module 3 (device that performs each substrate processing) in which each sensor 4 is provided. For example, the grouping may be based on the type and function of the sensor, such as temperature sensors and pressure sensors, or it may be based on its application. In the case of grouping by application, for example, a pressure sensor that measures the pressure when discharging gas or liquid and a flow sensor that measures the flow rate are used for similar purposes, so such sensors may be grouped together. Also, sensors 4 whose sensor values ​​were abnormal in past abnormal cases may be grouped together. The grouping of each sensor 4 may be done in advance by the user, for example, by the processing unit 21 of the higher-level module 2 based on the sensor name or installation location of each sensor 4. In grouping based on sensor name or installation location, for example, a language model may be used to group together sensors 4 with similar sensor names or installation locations, or a language model may be used to identify feature quantities for sensor names or installation locations, and sensors 4 with similar feature quantities may be grouped together.

[0051] Figure 8 is a flowchart showing an example of the procedure for generating a heatmap in Embodiment 2. The process shown in Figure 8 is the same as the process shown in Figure 5, but with steps S31 to S35 added between step S17 (YES) and step S18. The same steps as in Figure 5 will not be explained.

[0052] In the upper-level module 2 of the substrate processing apparatus 1 according to this embodiment, the processing unit 21 executes the processes of steps S11 to S17. As a result, in this embodiment as well, the processing unit 21 can calculate the degree of abnormality of the sensor value for each sensor 4 at each calculation timing time. If the processing unit 21 determines that it has calculated the degree of abnormality for all sensor values ​​(S17: YES), it reads out the degree of abnormality for each calculation timing time of each sensor 4, which has been grouped into one group, from the degree of abnormality for each calculation timing time of each sensor 4 stored in the storage unit 22 or the sensor value DB 22b (S31). Based on the read degree of abnormality for each calculation timing time of each sensor 4, the processing unit 21 calculates the degree of abnormality for each calculation timing time (group degree of abnormality) for that group (S32). For example, the group degree of abnormality can be the average value of the degree of abnormality of the sensor value of each sensor 4 at each calculation timing time. Alternatively, the group degree of abnormality may be a weighted average of the abnormality values ​​obtained by weighting the degree of abnormality of the sensor value of each sensor 4 at each calculation timing time for each sensor 4. By weighting the abnormal values ​​at each sensor 4 based on the weight set for each sensor 4, an index value that emphasizes the degree of abnormality at the sensor 4 of interest can be obtained. This allows for the acquisition of time-series data of the degree of abnormality within the group being processed.

[0053] The processing unit 21 determines whether or not the abnormality level of each group has been calculated for all groups registered in the group DB 22d (S33). If it determines that it has not been calculated (S33: NO), it returns to step S31 and performs steps S31 to S32 for the unprocessed groups. If it determines that the abnormality level of each group has been calculated for all groups (S33: YES), the processing unit 21 generates a heat map (second chart) showing the change in the abnormality level of each group over time (S34). Figure 9 is an explanatory diagram showing an example of a screen displaying the heat map. The heat map in the upper part of Figure 9 is a heat map for each group, with time on the horizontal axis and information for each group (e.g., group name) on the vertical axis. The group heat map has the same configuration as the heat map of each sensor 4 shown in Figure 6 and shows the change in the abnormality level of each group over time. In the heat map in the upper part of Figure 9, the time shown on the horizontal axis may be the elapsed time from the start of substrate processing, or it may be the date and time of measurement of the sensor value.

[0054] The heatmap of the groups shown in the upper part of Figure 9 is configured to allow selection of any group via the group name or heatmap. The processing unit 21 of the upper module 2 accepts the selection of any group by receiving user input via the input unit (operation unit) of the substrate processing device 1 for the displayed heatmap of the groups. In Figure 9, the fourth group is selected as indicated by the white arrow. The processing unit 21 determines whether or not it has accepted the group selection (S35). If it determines that it has not accepted the selection (S35: NO), it waits or terminates processing. If the processing unit 21 determines that it has accepted the group selection (S35: YES), it proceeds to step S18 and performs the processing in steps S18 to S23 for each sensor 4 grouped into the selected group. As a result, a heatmap for the sensors 4 belonging to the selected group is generated, and a screen like the one shown in the lower part of Figure 9 is displayed.

[0055] Through the process described above, in this embodiment, when a problem (abnormality) occurs during substrate processing in the substrate processing apparatus 1, the time-dependent change in the degree of abnormality of each group can be visualized using a heat map. Furthermore, by selecting any group, the sensor 4 belonging to the selected group that has an abnormal sensor value can be identified, and the time-dependent change in the degree of abnormality of the identified sensor 4 can be visualized using a heat map. Therefore, by first checking the degree of abnormality for each group, and then drilling down to check each sensor 4 within a group, the degree of abnormality of each sensor 4 within any group can be checked. For example, if an abnormality occurs in the group related to temperature in the group heat map, by selecting the temperature group, the state of each sensor 4 (the measurement target area of ​​each sensor 4) belonging to the temperature group can be checked using the heat map of each sensor 4. Since it becomes easy to identify the sensor 4 whose sensor value is abnormal when a problem occurs, it is expected that the location and cause of the abnormality can be identified based on the identified sensor 4.

[0056] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of this disclosure is indicated by the claims, not in the sense described above, and all modifications within the sense and scope equivalent to the claims are intended.

[0057] The matters described in each embodiment can be combined with each other. Furthermore, the independent and dependent claims described in the claims can be combined with each other in any combination, regardless of the form of reference. In addition, the claims use a form in which claims referencing two or more other claims (multi-claim form), but are not limited to this. A form in which multi-claims referencing at least one multi-claim (multi-multi-claim) may also be used.

[0058] 1. Substrate processing unit 2. Upper module 3. Lower module 4. Sensor 5. Controlled device 21. Processing unit 21a. Control processing unit 21b. Heat map generation unit 21c. Past case identification unit 22. Storage unit 22a. Program (computer program) 23. Communication unit

Claims

1. An information processing method in which a computer performs the following processes: acquires measurement data measured in a time series by multiple sensors installed on a substrate processing device; calculates the degree of abnormality for each sensor over time based on the acquired measurement data; identifies an abnormality ranking for each sensor based on the degree of abnormality for each sensor over time; and generates a chart showing the change in the degree of abnormality for each sensor over time based on the identified abnormality ranking.

2. The information processing method according to claim 1, wherein the computer performs a process to determine whether or not an anomaly has occurred in the measurement data of each sensor based on the degree of anomaly of each sensor over time, and to identify an anomaly ranking of each sensor based on the time at which the anomaly occurred in each sensor.

3. The information processing method according to claim 1 or 2, wherein the computer performs a process to determine the anomaly ranking of each sensor based on the calculated weighted average of the anomalies of each sensor, based on the anomaly degree of each sensor over time.

4. The information processing method according to claim 1 or 2, wherein the figure is a heat map in which time is shown on the first axis, each sensor on the second axis, and the change in the degree of abnormality of each sensor over time is shown by the color corresponding to the degree of abnormality of each sensor.

5. The information processing method according to claim 1 or 2, wherein the computer performs a process of identifying a predetermined number of sensors based on the identified anomaly ranking and generating the chart showing the change in the degree of anomaly of the identified sensors over time.

6. The information processing method according to claim 1 or 2, wherein the computer performs a process of grouping the plurality of sensors into a plurality of groups, and generating a chart for each group that shows the change in the degree of abnormality of each sensor over time.

7. The information processing method according to claim 1 or 2, wherein the plurality of sensors are divided into a plurality of groups, and the computer is instructed to perform a process of calculating the time-dependent abnormality level for each group based on the time-dependent abnormality level of the sensors belonging to each group, and generating a second chart showing the time-dependent change in the abnormality level of each group.

8. The information processing method according to claim 7, which causes the computer to perform a process of calculating the degree of abnormality for each group by a weighted average based on the weights set for each sensor belonging to each group.

9. The information processing method according to claim 7, wherein the computer receives the selection of any group via the second chart and performs a process to generate the chart showing the time change in the abnormality of the sensors belonging to the selected group.

10. The information processing method according to claim 1 or 2, wherein the substrate processing apparatus stores characteristic quantities of measurement data measured by each sensor when the abnormality occurred, associates this information with information about abnormalities that have occurred in the past, the computer identifies past abnormalities similar to the characteristic quantities of the measurement data based on the acquired characteristic quantities of the measurement data, and outputs information about the identified past abnormalities.

11. The information processing method according to claim 1 or 2, wherein the plurality of sensors are provided in a plurality of processing units of the substrate processing apparatus, and the computer performs a process to output a control signal based on the degree of abnormality of the sensor to the processing unit corresponding to the sensor in the diagram.

12. An information processing device having a control unit, wherein the control unit acquires measurement data measured in a time series by a plurality of sensors provided on a substrate processing device, calculates the degree of abnormality for each sensor over time based on the acquired measurement data, identifies an abnormality ranking for each sensor based on the degree of abnormality for each sensor over time, and generates a chart showing the change in the degree of abnormality for each sensor over time based on the identified abnormality ranking.

13. A computer program that causes a computer to perform the following processes: acquire measurement data measured in a time series by multiple sensors installed on a substrate processing device; calculate the degree of anomaly for each sensor over time based on the acquired measurement data; identify an anomaly ranking for each sensor based on the degree of anomaly for each sensor over time; and generate a chart showing the change in the degree of anomaly for each sensor over time based on the identified anomaly ranking.

Images