WO2024009902A1 - 情報処理方法、コンピュータプログラム及び情報処理装置 - Google Patents

情報処理方法、コンピュータプログラム及び情報処理装置 Download PDF

Info

Publication number
WO2024009902A1
WO2024009902A1 PCT/JP2023/024372 JP2023024372W WO2024009902A1 WO 2024009902 A1 WO2024009902 A1 WO 2024009902A1 JP 2023024372 W JP2023024372 W JP 2023024372W WO 2024009902 A1 WO2024009902 A1 WO 2024009902A1
Authority
WO
WIPO (PCT)
Prior art keywords
semiconductor manufacturing
data
sensor
information processing
sensor data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2023/024372
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
穣二 高良
晃一 吉田
ツォン サイ
隆男 舟久保
祐太 紀野國
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tokyo Electron Ltd
Original Assignee
Tokyo Electron Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tokyo Electron Ltd filed Critical Tokyo Electron Ltd
Priority to JP2024532095A priority Critical patent/JPWO2024009902A1/ja
Priority to KR1020257002178A priority patent/KR20250033233A/ko
Priority to CN202380049967.6A priority patent/CN119487613A/zh
Priority to TW112124884A priority patent/TW202420403A/zh
Publication of WO2024009902A1 publication Critical patent/WO2024009902A1/ja
Priority to US19/002,811 priority patent/US20250124357A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Program-control systems
    • G05B19/02—Program-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P72/00—Handling or holding of wafers, substrates or devices during manufacture or treatment thereof
    • H10P72/06—Apparatus for monitoring, sorting, marking, testing or measuring
    • H10P72/0604—Process monitoring, e.g. flow or thickness monitoring
    • H—ELECTRICITY
    • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10P—GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10
    • H10P95/00—Generic processes or apparatus for manufacture or treatments not covered by the other groups of this subclass
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Definitions

  • the present disclosure relates to an information processing method, a computer program, and an information processing device.
  • Patent Document 1 various information collected from substrate processing apparatuses is accumulated, and the accumulated data is used to display information necessary for energy saving of each substrate processing apparatus installed in a semiconductor manufacturing factory.
  • An integrated management system is proposed. This integrated management system accumulates various information including power consumption information, gas consumption information, or operation information of the substrate processing equipment, acquires information that satisfies predetermined conditions from the accumulated information, and uses the information consumed by the substrate processing equipment. At least one of power consumption, inert gas consumption, and device operating rate of the substrate processing apparatus is calculated and displayed.
  • the present disclosure provides an information processing method, a computer program, and an information processing device that can be expected to estimate the operating state of a semiconductor manufacturing device.
  • An information processing method includes, when operating a first semiconductor manufacturing apparatus provided with a plurality of sensors, acquiring operational data of the first semiconductor manufacturing apparatus and sensor data output from the sensors. , based on the acquired operating data and the sensor data, a learning model that outputs the sensor data in accordance with the input of the operating data is generated by machine learning, and the generated learning model is applied to the first semiconductor manufacturing apparatus.
  • the semiconductor device is placed in a second semiconductor manufacturing apparatus different from the first semiconductor manufacturing device.
  • FIG. 1 is a schematic diagram showing a configuration example of an information processing system according to the present embodiment.
  • FIG. 2 is a schematic diagram for explaining an example of a sensor for semiconductor manufacturing equipment.
  • FIG. 1 is a block diagram illustrating a configuration example of an information processing apparatus according to an embodiment.
  • FIG. 2 is a schematic diagram showing a configuration example of a process DB.
  • 2 is a flowchart illustrating an example of a procedure of predictive model generation processing performed by the information processing device according to the present embodiment.
  • FIG. 3 is a schematic diagram showing an example of information display regarding generation of a predictive model.
  • FIG. 3 is a schematic diagram showing an example of information display regarding generation of a predictive model.
  • FIG. 3 is a schematic diagram for explaining an example of how a prediction model is used in simulation.
  • FIG. 7 is a flowchart illustrating an example of a procedure for calculating a degree of contribution performed by the information processing apparatus according to the present embodiment.
  • FIG. 3 is a schematic diagram showing an example of information display regarding contribution degree.
  • FIG. 2 is a flowchart illustrating an example of a procedure of unit stop processing performed by the information processing apparatus according to the present embodiment.
  • FIG. 3 is a schematic diagram for explaining the effect of automatic stopping of a unit in a standby state.
  • FIG. 3 is a schematic diagram showing an example of display related to processing timing verification.
  • FIG. 3 is a schematic diagram showing an example of display related to processing timing verification.
  • FIG. 1 is a schematic diagram showing a configuration example of an information processing system according to the present embodiment.
  • the information processing system according to this embodiment includes an information processing device 1 and a semiconductor manufacturing device 2.
  • the semiconductor manufacturing apparatus 2 is an apparatus that performs various processes for semiconductor manufacturing, such as CVD (Chemical Vapor Deposition), sputtering, or etching.
  • the information processing device 1 is, for example, a device that acquires and collects operational data output by the semiconductor manufacturing device 2 and monitors and controls the operation of the semiconductor manufacturing device 2 .
  • the information processing device 1 may monitor and control a plurality of semiconductor manufacturing devices 2 .
  • the operation data acquired by the information processing apparatus 1 may include various data such as process log data, recipe data, wafer transfer history data, and error data.
  • a plurality of sensors are provided for measuring the detailed operating state, etc. of at least one semiconductor manufacturing device 2 that the information processing device 1 monitors and controls. 3 is provided.
  • the sensor 3 is not provided for all the semiconductor manufacturing apparatuses 2, but is provided for some (at least one) of the plurality of semiconductor manufacturing apparatuses 2. 3 may be provided. 1 and the following description, the semiconductor manufacturing apparatus 2 provided with the sensor 3 is referred to as a semiconductor manufacturing apparatus 2A, the semiconductor manufacturing apparatus 2 not provided with the sensor 3 is referred to as a semiconductor manufacturing apparatus 2B, and the semiconductor manufacturing apparatus 2 provided with the sensor 3 is referred to as a semiconductor manufacturing apparatus 2B.
  • the information processing device 1 that monitors and controls the semiconductor manufacturing device 2A in which the sensor 3 is provided is referred to as the information processing device 1A
  • the information processing device 1 that monitors and controls the semiconductor manufacturing device 2B in which the sensor 3 is not provided is referred to as the information processing device 1A
  • the information processing device 1 that performs the above processing will be referred to as an information processing device 1B, and if there is no need to distinguish between them, it will be simply referred to as an information processing device 1.
  • the plurality of sensors 3 provided in the semiconductor manufacturing apparatus 2A are sensors that perform measurements related to the environment, such as sensors that measure power consumption, sensors that measure the amount of water discharged, or sensors that measure the amount of gas discharged. Contains sensors.
  • the gases for which the sensor measures emissions may include, for example, gases such as CO2 (carbon dioxide) or NOx (nitrogen oxides).
  • the semiconductor manufacturing apparatus 2 includes, for example, an upper chiller unit, an RF (Radio Frequency) power supply unit, a DC (Direct Current) power supply unit, a vacuum pump unit, a chamber heater unit, an ESC (Electric Static Chuck) heater unit, and It is constructed by combining various units such as a lower chiller unit.
  • the plurality of sensors 3 may include sensors that measure power consumption, amount of water discharged, amount of gas discharged, etc. for each unit of the semiconductor manufacturing apparatus 2A.
  • the operation data output by the semiconductor manufacturing equipment 2A and the data of the plurality of sensors 3 are used.
  • the information processing device 1A acquires the sensor data that is the measurement result, and stores and accumulates the acquired driving data and sensor data in a database in association with each other.
  • the information processing device 1A performs so-called supervised machine learning processing based on the driving data and sensor data accumulated in the database, and creates a learning model that receives the driving data as input and outputs a predicted value of the sensor data as a prediction model 5. Generate as.
  • the information processing apparatus 1A deploys (arranges) the generated prediction model 5 to the information processing apparatus 1B of the semiconductor manufacturing apparatus 2B in which the sensor 3 is not provided.
  • the information processing device 1B uses the prediction model 5 generated by the information processing device 1 to obtain a predicted value of sensor data from the operation data output by the semiconductor manufacturing device 2B, and calculates the predicted value of the semiconductor manufacturing device 2B based on this predicted value. Processing such as control or monitoring can be performed.
  • placing the prediction model 5 in the semiconductor manufacturing equipment 2B or placing the prediction model 5 in the information processing device 1B of the semiconductor manufacturing equipment 2B means that the operation data of the semiconductor manufacturing equipment 2B is This refers to a state in which the prediction model 5 can predict sensor data.
  • the semiconductor manufacturing equipment 2A is installed at a base of a company or the like that develops and sells the semiconductor manufacturing equipment 2, for example.
  • the semiconductor manufacturing equipment 2A is operated on a trial basis, and at this time various measurements such as power consumption, amount of water discharged, and amount of gas discharged are performed by the sensor 3, and the sensor data outputted by the sensor 3 is accumulated together with the operational data. Ru. After sufficient driving data and sensor data are accumulated, a prediction model 5 is generated by the information processing device 1A.
  • the sensor 3 is not provided for each semiconductor manufacturing equipment 2B ( Alternatively, a smaller number of sensors 3 may be provided than the number of sensors 3 provided for the semiconductor manufacturing apparatus 2A).
  • a prediction model 5 generated by the information processing device 1A is installed (introduced) in the information processing device 1B that monitors and controls each semiconductor manufacturing device 2B. ), and the prediction model 5 is deployed (arranged) to each semiconductor manufacturing apparatus 2B.
  • the information processing device 1B acquires operational data during the initial operation of the semiconductor manufacturing device 2B, for example, and inputs the acquired operational data to the predictive model 5, so that the predictive model 5 outputs the data without using the sensor 3. Sensor data can be acquired.
  • FIG. 2 is a schematic diagram for explaining an example of the sensor 3 for the semiconductor manufacturing apparatus 2A.
  • the semiconductor manufacturing apparatus 2A shown in FIG. 2 includes six units 1 to 6 and a control unit that controls them. Power is supplied to the semiconductor manufacturing apparatus 2A from three power supplies 1 to 3. Power supply 1 supplies power to the control unit, unit 1 and unit 2. Power supply 2 supplies power to unit 3 and unit 4. Power supply 3 supplies power to unit 5 and unit 6. Power supplies 1 to 3 each supply power from, for example, a commercial AC power source to each unit of the semiconductor manufacturing apparatus 2A.
  • FIG. 2 individual power supply paths from the power supply to each unit are shown by thick solid lines.
  • sensors for measuring power consumption for example, are provided for each of these power supply paths.
  • the circle drawn over the thick solid line is the position where the sensor 3 is provided.
  • operating data and sensor data are acquired using the semiconductor manufacturing equipment 2A equipped with the sensor 3 during the trial operation stage, and the prediction model 5 is created based on the acquired data. generate.
  • the semiconductor manufacturing equipment 2B is not provided with the sensor 3 (or a smaller number of sensors 3 may be provided than in the test stage), and instead of using the sensor 3, the sensor data is calculated using the prediction model 5. Acquisition is performed.
  • FIG. 3 is a block diagram showing an example of the configuration of the information processing device 1 according to the present embodiment.
  • the information processing device 1 according to the present embodiment includes a processing section 11, a storage section 12, a communication section (transceiver) 13, an input/output section 14, a display section 15, an operation section 16, etc. .
  • a processing section 11 a storage section 12
  • a communication section (transceiver) 13 a communication section (transceiver) 13
  • an input/output section 14 a display section 15, an operation section 16, etc.
  • the present embodiment will be described assuming that the processing is performed by one information processing device 1, the processing of the information processing device 1 may be performed in a distributed manner by a plurality of devices.
  • the processing unit 11 includes an arithmetic processing unit such as a CPU (Central Processing Unit), an MPU (Micro-Processing Unit), a GPU (Graphics Processing Unit), or a quantum processor, a ROM (Read Only Memory), a RAM (Random Access Memory), etc. It is configured using The processing unit 11 reads out and executes the program 12a stored in the storage unit 12 to perform a process of acquiring operational data of the semiconductor manufacturing equipment 2, a process of acquiring sensor data measured by the sensor 3, and a process of acquiring the sensor data measured by the sensor 3. Various processes such as a process of generating a prediction model 5 based on the data are performed.
  • the storage unit 12 is configured using a large-capacity storage device such as a hard disk, for example.
  • the storage unit 12 stores various programs executed by the processing unit 11 and various data necessary for processing by the processing unit 11.
  • the storage unit 12 stores a program 12a executed by the processing unit 11.
  • the storage unit 12 also includes a process DB (database) 12b that stores operational data acquired from the semiconductor manufacturing equipment 2 and sensor data acquired from the sensor 3 in a correlated manner, and predictions generated by machine learning based on these data.
  • a predictive model storage unit 12c that stores information regarding the model 5 is provided.
  • the program (computer program, program product) 12a is provided in a form recorded on a recording medium 99 such as a memory card or an optical disk, and the information processing device 1 reads the program 12a from the recording medium 99 and stores it in the storage unit. 12.
  • the program 12a may be written into the storage unit 12, for example, during the manufacturing stage of the information processing device 1.
  • the program 12a may be distributed by a remote server device or the like, and the information processing device 1 may obtain it through communication.
  • the program 12a may be recorded on the recording medium 99 and read by a writing device and written into the storage unit 12 of the information processing device 1.
  • the program 12a may be provided in the form of distribution via a network, or may be provided in the form of being recorded on the recording medium 99.
  • the process DB 12b is a database in which the information processing device 1A stores operational data acquired from the semiconductor manufacturing device 2A and sensor data acquired from the sensor 3 in association with each other.
  • FIG. 4 is a schematic diagram showing an example of the configuration of the process DB 12b.
  • the illustrated process DB 12b stores data such as "time stamp”, “apparatus ID”, “recipe ID”, "operation data”, and “sensor data” in association with each other, for example.
  • the “time stamp” is, for example, information about the date and time when the information processing device 1A acquired the driving data and sensor data.
  • the "apparatus ID” is identification information uniquely assigned to the semiconductor manufacturing apparatus 2, and can be determined in advance by, for example, the designer or administrator of the information processing system according to the present embodiment.
  • the "recipe ID” is identification information uniquely attached to information such as the procedure or setting of the semiconductor manufacturing process performed in the semiconductor manufacturing equipment 2, so-called recipe. It can be determined in advance by a user such as a worker who performs the work.
  • “Operation data” includes, for example, “High frequency RF power [W]”, “High frequency RF power pulse duty [%]", “High frequency RF power pulse frequency [kHz]”, “Low frequency RF power [W]”, “ “Low frequency RF power pulse duty [%]”, “Low frequency RF power pulse frequency [kHz]”, “Pressure [mTorr]”, “Total gas flow rate [sccm]”, “ESC temperature [°C]” and Chiller temperature [ °C], etc. may be included. Note that the information included in the operational data illustrated in this figure is information included in so-called process log data output by the semiconductor manufacturing apparatus 2 as the semiconductor manufacturing process is performed.
  • the information included in the operation data is not limited to information on process log data, and may include various information such as recipe data, wafer transfer history data, and error data.
  • Recipe data is a collection of information such as procedures or settings of a semiconductor manufacturing process.
  • the wafer transfer history data is a collection of information related to wafer transfer, such as the ID of the transferred wafer and the date and time of transfer.
  • the error data is a collection of information such as the date and time when an error (abnormality, malfunction, etc.) was detected in the semiconductor manufacturing apparatus 2 and the type of error.
  • the “sensor data” may include information such as “power consumption 1 [W]", “power consumption 2 [W]”, . . . "power consumption N [W]”, etc.
  • the sensor data illustrated in this figure is based on the assumption that the sensor 3 individually measures the power consumption of N units included in the semiconductor manufacturing apparatus 2.
  • the information included in the sensor data is not limited to power consumption, and may include various information that can be detected by the sensor 3, such as the amount of water discharged or the amount of gas discharged. Further, the information included in the sensor data does not have to be information for each unit of the semiconductor manufacturing apparatus 2.
  • the information processing device 1A repeatedly acquires operational data while the semiconductor manufacturing device 2A is performing a semiconductor manufacturing process, and stores the acquired operational data in the process DB 12b.
  • time-series operation data is accumulated in the process DB 12b.
  • the information processing device 1B that monitors and controls the semiconductor manufacturing device 2B in which the sensor 3 is not provided does not need to have the process DB 12b.
  • the information processing apparatus 1B also has a process DB 12b, and stores data other than sensor data, such as "time stamp", "apparatus ID", "recipe ID", and "operation data".
  • the predictive model storage unit 12c of the storage unit 12 stores information regarding the predictive model 5 generated by the information processing device 1A based on the driving data and sensor data stored in the process DB 12b.
  • the information regarding the prediction model 5 may include, for example, configuration information indicating what kind of configuration the prediction model 5 has, and information such as values of internal parameters of the prediction model 5.
  • information regarding the prediction model 5 generated by the information processing apparatus 1A is stored in the prediction model storage unit 12c.
  • the predictive model storage unit 12c stores information regarding the predictive model 5 generated by another information processing device 1A. Ru.
  • the communication unit 13 of the information processing device 1 communicates with various devices via a wired or wireless network N including a LAN (Local Area Network), the Internet, or a mobile phone communication network.
  • the communication unit 13 transmits data provided from the processing unit 11 to other devices, and also provides data received from other devices to the processing unit 11.
  • the communication unit 13 sends and receives information regarding the prediction model 5 by communicating with another information processing device 1 via the network N.
  • the input/output unit 14 is connected to the semiconductor manufacturing equipment 2 via a communication line, a signal line, etc., and exchanges information with the semiconductor manufacturing equipment 2.
  • the input/output unit 14 receives operational data output from the semiconductor manufacturing apparatus 2, acquires this operational data, and provides the acquired operational data to the processing unit 11.
  • the input/output unit 14 outputs control commands and the like given from the processing unit 11 to the semiconductor manufacturing apparatus 2.
  • the input/output section 14 is connected to one or more sensors 3.
  • the input/output section 14 receives sensor data output from the sensor 3, acquires this sensor data, and provides it to the processing section 11.
  • the information processing device 1A directly acquires the sensor data of the sensor 3, but the configuration is not limited to this, and the semiconductor manufacturing device 2A acquires the sensor data output by the sensor 3. However, a configuration may be adopted in which the semiconductor manufacturing apparatus 2A transmits operation data and sensor data to the information processing apparatus 1A.
  • the display unit 15 is configured using a liquid crystal display or the like, and displays various images, characters, etc. based on the processing of the processing unit 11.
  • the display unit 15 displays various information such as, for example, operating data acquired from the semiconductor manufacturing apparatus 2, sensor data acquired from the sensor 3, and information regarding a prediction model generated from these data.
  • the operation unit 16 accepts user operations and notifies the processing unit 11 of the accepted operations.
  • the operation unit 16 receives user operations using input devices such as mechanical buttons or a touch panel provided on the surface of the display unit 15.
  • the operation unit 16 may be an input device such as a mouse and a keyboard, and these input devices may be configured to be detachable from the information processing apparatus 1.
  • the storage unit 12 may be an external storage device connected to the information processing device 1.
  • the information processing device 1 may be a multicomputer including a plurality of computers, or may be a virtual machine virtually constructed using software. Further, the information processing device 1 is not limited to the above configuration, and may not include the display section 15, the operation section 16, etc., for example.
  • the information processing device 1 has a data acquisition section 11a, a predictive model generation section 11b, a predictive model placement section, etc., by the processing section 11 reading out and executing the program 12a stored in the storage section 12.
  • 11c, a control processing section 11d, a display processing section 11e, and the like are implemented in the processing section 11 as software-like functional sections.
  • the information processing apparatus 1B that monitors and controls the semiconductor manufacturing apparatus 2B in which the sensor 3 is not provided does not have to include the data acquisition section 11a, the predictive model generation section 11b, and the predictive model placement section 11c.
  • the data acquisition unit 11a performs a process of acquiring operational data output by the semiconductor manufacturing apparatus 2A, for example, while or after the semiconductor manufacturing apparatus 2A is performing a semiconductor manufacturing process. Further, the data acquisition unit 11a performs a process of acquiring sensor data output by one or more sensors 3 at the same timing as acquiring operation data from the semiconductor manufacturing apparatus 2A, for example.
  • the data acquisition unit 11a attaches information such as a "time stamp”, "apparatus ID”, and “recipe ID" to the acquired operation data and sensor data, and stores these information in the process DB 12b in association with each other.
  • the predictive model generation unit 11b performs processing to generate the predictive model 5 by performing machine learning processing using the driving data and sensor data acquired by the data acquisition unit 11a and accumulated in the process DB 12b.
  • the prediction model 5 is a learning model that receives driving data as input and outputs a predicted value of sensor data.
  • the predictive model generation unit 11b uses the operation data stored in the process DB 12b as input data (explanatory variables), and generates learning data (so-called teacher data) that is associated with sensor data corresponding to this operation data as a correct value (objective variable). ) and performs machine learning (so-called supervised learning) processing using this learning data to generate the prediction model 5.
  • the predictive model generation unit 11b stores information regarding the generated predictive model 5 in the predictive model storage unit 12c. Further, the prediction model generation unit 11b may repeatedly generate (update) the prediction model 5 at a predetermined period, such as once a week or once a month.
  • the prediction model 5 may be, for example, a PLS (Partial Least Squares) regression model.
  • various learning models such as an ARX (Auto-Regressive with eXogenous) model, an SVM (Support Vector Machine), a random forest, or a neural network may be adopted as the prediction model 5.
  • the prediction model 5 may be configured to receive time-series driving data as input and output time-series sensor data.
  • the prediction model 5 may be a learning model such as RNN (Recurrent Neural Network), LSTM (Long Short Term Memory), or Transformer. Note that the structure of these learning models, the method of generating a learning model by machine learning, etc. are existing technologies, and therefore detailed explanations will be omitted in this embodiment.
  • the predictive model placement unit 11c places (deploys) the predictive model 5 generated by the predictive model generation unit 11b on the information processing device 1B that monitors and controls the semiconductor manufacturing device 2B in which the sensor 3 is not installed. Perform processing.
  • the predictive model placement unit 11c receives, for example, information input from the user such as the device ID of the information processing device 1B as the placement destination, reads out information regarding the predictive model 5 from the predictive model storage unit 12c, and stores information on the placement destination for which the input has been received. By transmitting information regarding the prediction model 5 to the processing device 1B, the prediction model 5 can be placed in the information processing device 1B.
  • the information processing device 1B which has received the information regarding the prediction model 5 from the information processing device 1A, stores the received information in the prediction model storage unit 12c of its own storage unit 12, and can use the prediction model 5 in subsequent processing. can.
  • the predictive model generation unit 11b repeatedly generates (updates) the predictive model 5
  • the predictive model placement unit 11c updates the information processing device 1B in which the predictive model 5 was previously placed each time the predictive model 5 is updated.
  • information regarding the updated prediction model 5 may be transmitted (relocated).
  • the control processing unit 11d performs control processing of the semiconductor manufacturing device 2 based on the operation data acquired from the semiconductor manufacturing device 2 and the sensor data acquired from the sensor 3 or predicted by the prediction model 5. For example, the control processing unit 11d monitors changes in the power consumption (power consumption amount) of each unit of the semiconductor manufacturing apparatus 2 based on sensor data, determines the operating state of each unit based on the power consumption, and determines the standby state ( By stopping the operation (stopping the power supply) of the unit (in a state where no processing is being performed), control can be performed to reduce the power consumption of the entire semiconductor manufacturing apparatus 2.
  • control processing unit 11d may determine whether or not there is an abnormality in each unit based on the operation data or sensor data, and if it is determined that there is an abnormality, perform control to stop the operation of the unit or the semiconductor manufacturing apparatus 2. Can be done.
  • control processing of the semiconductor manufacturing apparatus 2 performed by the control processing section 11d may be any kind of processing.
  • the display processing unit 11e performs processing to display various information on the display unit 15.
  • the display processing unit 11e can graph and display temporal changes in various types of information included in the driving data and sensor data acquired by the data acquisition unit 11a.
  • the display processing unit 11e also includes information regarding the prediction model 5 generated by the prediction model generation unit 11b, such as comparison information between the measurement value measured by the sensor 3 and the predicted value by the prediction model 5, and the prediction accuracy by the prediction model 5. It is possible to display various information such as.
  • the display processing unit 11e may generate predicted values such as power consumption, water volume, or gas emissions of the semiconductor manufacturing equipment 2 based on the sensor data output by the prediction model 5 based on the operation data acquired by the data acquisition unit 11a. can be displayed on the display section 15.
  • the display processing unit 11e converts the power consumption into information such as carbon dioxide emissions by performing a predetermined calculation on the predicted value of the power consumption of the semiconductor manufacturing apparatus 2, and displays the information. Good too.
  • one or more semiconductor manufacturing apparatuses 2A are appropriately selected from among the plurality of semiconductor manufacturing apparatuses 2 by a user such as a designer or administrator of the system.
  • a sensor 3 is provided in the semiconductor manufacturing apparatus 2A.
  • a user performs a semiconductor manufacturing process using the semiconductor manufacturing apparatus 2A, and at this time, the information processing apparatus 1A acquires operating data of the semiconductor manufacturing apparatus 2A and sensor data of the sensor 3, and stores and accumulates them in the process DB 12b. After sufficient data is accumulated in the process DB 12b, the information processing device 1A performs prediction by performing machine learning using the accumulated operational data and sensor data, for example, in response to user operations or repeatedly at a predetermined cycle. Generate model 5.
  • the information processing device 1A sends information regarding the generated prediction model 5 to the information processing device 1B, thereby making the prediction model 5 usable for monitoring and controlling the semiconductor manufacturing device 2B in which the sensor 3 is not installed. do.
  • FIG. 5 is a flowchart illustrating an example of the procedure of the predictive model generation process performed by the information processing device 1A according to the present embodiment.
  • the data acquisition unit 11a of the processing unit 11 of the information processing device 1A according to the present embodiment is configured to perform an operation that the semiconductor manufacturing device 2A outputs at a predetermined period when the semiconductor manufacturing process of the semiconductor manufacturing device 2A is performed. Data and sensor data output by the sensor 3 are acquired (step S1).
  • the data acquisition unit 11a stores the operation data and sensor data acquired in step S1 in the process DB 12b of the storage unit 12 (step S2).
  • the processing unit 11 determines whether the timing to generate the prediction model 5 has arrived (step S3). At this time, the processing unit 11 updates the prediction model 5 repeatedly at a predetermined cycle, receives an instruction to generate the prediction model 5 from the user, or stores a predetermined amount of data in the process DB 12b. When driving data and sensor data are accumulated, it can be determined that the timing to generate the prediction model 5 has arrived. If the timing to generate the prediction model 5 has not yet arrived (S3: NO), the processing unit 11 returns the process to step S1 and continues collecting data in steps S1 and S2.
  • the prediction model generation unit 11b of the processing unit 11 reads out the operation data and sensor data stored in the process DB 12b (step S4).
  • the predictive model generation unit 11b uses the driving data read out in step S4 as input (explanatory variable) and generates learning data in which the sensor data is associated with the correct value (objective variable) (step S5).
  • the predictive model generation unit 11b may perform processing such as normalization or regularization of numerical values included in the driving data and sensor data as appropriate.
  • the predictive model generation unit 11b uses the learning data generated in step S5 to perform supervised machine learning processing on the learning model with a predetermined configuration (step S6), and calculates the internal parameters of the learning model. By determining , a prediction model 5 is generated.
  • the predictive model generation unit 11b calculates prediction accuracy by performing prediction on the predictive model 5 generated in step S6 using verification data generated separately from the learning data (step S7).
  • the verification data is data in the same format as the learning data.
  • part of all the data stored in the process DB 12b may be used as verification data, and the rest may be used as learning data.
  • the prediction model generation unit 11b determines whether the prediction accuracy calculated in step S7 exceeds a threshold (step S8). Note that this threshold value is the prediction accuracy required for the prediction model 5, and is determined in advance by a user such as a designer or administrator of this system, for example. If the prediction accuracy does not exceed the threshold (S8: NO), the prediction model generation unit 11b returns the process to step S6 and repeatedly performs the machine learning process to increase the prediction accuracy.
  • the prediction model generation unit 11b stores information regarding the generated prediction model in the prediction model storage unit 12c of the storage unit 12 (Step S9).
  • the display processing unit 11e of the processing unit 11 displays information such as prediction accuracy of the generated prediction model 5 on the display unit 15 (step S10).
  • the predictive model placement unit 11c of the processing unit 11 transmits information regarding the generated predictive model 5 to the information processing device 1B that monitors and controls the semiconductor manufacturing device 2B in which the sensor 3 is not installed (step S11). As a result, the prediction model 5 is placed in the semiconductor manufacturing apparatus 2B, and the process ends.
  • FIG. 6 and 7 are schematic diagrams showing an example of information display regarding generation of the prediction model 5. This figure is an example of information displayed in step S10 of the flowchart of FIG. 5, for example.
  • the information processing device 1A displays a table showing the prediction error of the power consumption of each unit included in the semiconductor manufacturing device 2A, and a graph showing temporal changes in the actual measured value and predicted value of the total power consumption. are displayed vertically on the screen of the display section 15.
  • the table showing the prediction error shows, for example, the unit name of the unit of the semiconductor manufacturing equipment 2A, the range of the actual measured value of power consumption, and the maximum prediction error value [kW] and percentage [%] in association with each other. There is.
  • the graph of total power consumption has time on the horizontal axis and power consumption on the vertical axis, with the actual measured value of the total power consumption shown by a light gray line, and the predicted value shown by a dark gray line.
  • the actual measured values oscillate up and down, and the predicted values are close to the median or average value of the actual measured values.
  • the information processing device 1A can calculate the illustrated prediction error, etc. by verifying the generated prediction model 5 using verification data prepared in advance.
  • the verification data is generated using the operation data and sensor data stored in the process DB 12b and has the same configuration as the learning data.
  • the information processing device 1A inputs the driving data included in the verification data to the prediction model 5, and acquires the predicted value of the sensor data output by the prediction model 5.
  • the information processing device 1A can calculate the prediction error by calculating the difference between the actual measured value of the sensor data included in the verification data and the predicted value of the sensor data by the prediction model 5.
  • the information processing apparatus 1A also displays the total amount of power consumed in each process for each process condition (process recipe) when the semiconductor manufacturing process is performed in the semiconductor manufacturing apparatus 2A. Information is displayed for comparison.
  • the information processing device 1A displays a table showing prediction errors for each processing condition and a bar graph showing the total power consumption for each processing condition, side by side on the top and bottom of the screen of the display unit 15.
  • the table showing prediction errors for each processing condition includes, for example, the actual measured value [kWh] and predicted value [kWh] of total power consumption, and the value [kWh] and percentage of the prediction error for three processing conditions A to C. [%] is shown in correspondence.
  • the bar graph of the total power consumption shows the actual measured value of the total power consumption in light gray and the predicted value in dark gray for the three processing conditions A to C, for example.
  • the prediction model 5 generated by the information processing device 1A can be used for various processes. Below, some usage examples of the prediction model 5 will be explained.
  • FIG. 8 is a schematic diagram for explaining an example of use of the prediction model 5 in simulation.
  • the upper part of FIG. 8 shows the flow when the prediction model 5 predicts sensor data based on the operation data output by the semiconductor manufacturing apparatus 2.
  • a user such as a designer or administrator of this system creates recipe data for a semiconductor manufacturing process and inputs it to the semiconductor manufacturing equipment 2B, and the semiconductor manufacturing equipment 2B executes the semiconductor manufacturing process based on the input recipe data. and outputs operational data such as process log data.
  • the information processing device 1B inputs the operating data acquired from the semiconductor manufacturing device 2B into the prediction model 5 generated in advance, and acquires the predicted value of the sensor data output by the prediction model 5.
  • the lower part of FIG. 8 shows the flow when the prediction model 5 is used in the simulator 7 of the semiconductor manufacturing equipment 2.
  • the simulator 7 is software that simulates the semiconductor manufacturing process performed by the semiconductor manufacturing apparatus 2.
  • the simulator 7 receives as input recipe data that includes information such as settings for the semiconductor manufacturing process performed by the semiconductor manufacturing equipment 2, and simulates the processing performed by the semiconductor manufacturing equipment 2 according to this recipe data.
  • the operation data output by the manufacturing apparatus 2 is generated.
  • the user inputs the recipe data he/she created into the simulator 7 to perform a simulation, and inputs the operation data output by the simulator 7 to the prediction model 5, thereby obtaining the predicted value of the sensor data output by the prediction model 5. can be obtained.
  • the user can verify the power consumption, wastewater volume, gas emissions, etc. by the semiconductor manufacturing equipment 2 when processing is performed based on the recipe data created by the user. can.
  • the user can repeatedly perform simulations while modifying the recipe data in order to improve, for example, power consumption, drainage volume, or gas discharge volume.
  • the user can calculate the power consumption and wastewater amount of the semiconductor manufacturing device 2 under various conditions without actually operating the semiconductor manufacturing device 2A. Or gas emissions etc. can be verified.
  • the simulator 7 and the prediction model 5 are shown as separate blocks in FIG. 8, the prediction model 5 may be incorporated into the simulator 7 and used. In this case, the simulator 7 can output the predicted value of the sensor data along with the driving data.
  • processing such as the simulation by the simulator 7 and the acquisition of predicted values of sensor data using the prediction model 5 is performed by a device different from the information processing device 1 that monitors and controls the semiconductor manufacturing device 2, for example, information for simulation. It may be performed in a processing device.
  • the learning model generated by machine learning uses output data ( It is possible to calculate the degree of contribution (degree of importance, degree of influence, etc.) of input data (explanatory variable) to objective variable).
  • the information processing device 1 calculates the degree of contribution of various items included in the driving data input to the prediction model 5 to the sensor data output by the prediction model 5, and calculates the calculation results. can be displayed on the display unit 15 and presented to the user.
  • the information processing device 1 may display information regarding the degree of contribution, for example, based on a user's operation, or when an abnormality or the like is detected in the sensor data output by the prediction model 5, for example.
  • FIG. 9 is a flowchart illustrating an example of a procedure for calculating a degree of contribution performed by the information processing device 1 according to the present embodiment.
  • the processing unit 11 of the information processing device 1 according to the present embodiment receives from the user a designation of, for example, a range of operation data to be verified for contribution from among the operation data stored in the process DB 12b, and Obtain target driving data (step S21).
  • the processing unit 11 inputs the driving data acquired in step S21 to the trained prediction model 5 whose structure, parameters, and other information are stored in the prediction model storage unit 12c (step S22).
  • the processing unit 11 acquires sensor data output by the prediction model 5 in response to the driving data input in step S22 (step S23).
  • the processing unit 11 also calculates the degree of contribution of each item included in the driving data input to the prediction model 5 (step S24). At this time, the processing unit 11 calculates the SHAP value (Shapley value) of each item included in each driving data, and calculates the average value or total value of the SHAP value for each item, and calculates the SHAP value for each item of the driving data. It can be the contribution of an item.
  • the SHAP value is a numerical value indicating how much influence each item of input data to the learning model has on output data. Since calculation of the SHAP value is an existing technique, a detailed explanation of the calculation method will be omitted.
  • the processing unit 11 displays the degree of contribution calculated in step S24 on the display unit 15 (step S25), and ends the process.
  • the contribution degree of each item of the driving data is calculated using the SHAP value, but a method other than using the SHAP value may be adopted as a method for calculating the contribution degree.
  • a learning model such as RNN (Recurrent Neural Network), LSTM (Long Short Term Memory), or Transformer
  • the degree of contribution may be calculated using an attention mechanism.
  • a tree-structured learning model such as a decision tree, random forest, XGBoost (eXtreme Gradient Boosting), or LightGBM (Light Gradient Boosting Machine) is adopted as the prediction model 5
  • Contribution degree calculation may also be performed.
  • the degree of contribution may be calculated based on the following calculation formula, for example.
  • FIG. 10 is a schematic diagram showing an example of information display regarding the degree of contribution. This figure is an example of information displayed in step S25 of the flowchart of FIG. 9, for example.
  • the information processing device 1 displays a table showing the degree of contribution for each item included in the driving data and a pie chart showing the proportion of the degree of contribution for a plurality of items included in the driving data. They are displayed side by side on the top and bottom of the screen of section 15.
  • a table showing the degree of contribution for each item for example, multiple items included in the driving data are shown arranged from top to bottom in descending order of degree of contribution, and the percentage [%] of each item's contribution to the whole is shown as an item. They are shown in correspondence with each other.
  • the pie chart is a graph in which the proportion of the contribution of multiple items included in the driving data to the whole is mapped to the area of the sector, which is color-coded, based on the contribution shown in the table above. It is something.
  • the information processing device 1 By the information processing device 1 calculating and displaying the contribution of multiple items included in the operating data, the user can judge which item is effective to improve, for example, when power consumption is high. can do. Note that processing such as calculation and display of contribution degrees for a plurality of items included in the operation data may be performed by an information processing device different from the information processing device 1 that monitors and controls the semiconductor manufacturing device 2.
  • the information processing device 1B acquires operational data while the semiconductor manufacturing device 2B is performing a semiconductor manufacturing process, and based on the acquired operational data.
  • a predicted value of sensor data is obtained using the prediction model 5.
  • the information processing device 1B obtains a predicted value of power consumption of each unit using the prediction model 5, for example, and determines that this unit is in a standby state when the predicted value of power consumption of each unit is smaller than a threshold value.
  • the information processing device 1B can reduce power consumption by stopping the operation of a unit in a standby state and stopping power supply to this unit.
  • FIG. 11 is a flowchart illustrating an example of a procedure for unit stop processing performed by the information processing device 1B according to the present embodiment.
  • the processing unit 11 of the information processing device 1B according to the present embodiment acquires operational data output by the semiconductor manufacturing device 2B while the semiconductor manufacturing device 2B is performing a semiconductor manufacturing process (step S41).
  • the processing unit 11 inputs the driving data acquired in step S41 to the learned predictive model 5 whose structure, parameters, and other information are stored in the predictive model storage unit 12c (step S42).
  • the processing unit 11 acquires sensor data output by the prediction model 5 in response to the driving data input in step S42 (step S43).
  • the processing unit 11 selects one unit to be processed from among the plurality of units included in the semiconductor manufacturing apparatus 2B (step S44).
  • the processing unit 11 acquires a predicted power consumption value for the selected unit in step S44 from the sensor data acquired in step S43, and determines whether this power consumption is smaller than a predetermined threshold ( Step S45).
  • this threshold value is a value predetermined by a user such as a designer or administrator of the information processing system according to the present embodiment, and a different threshold value may be adopted for each unit. If the power consumption is smaller than the threshold (S45: YES), the processing unit 11 stops the operation of the target unit (step S46), and advances the process to step S47. If the power consumption is equal to or greater than the threshold (S45: NO), the processing unit 11 advances the process to step S47.
  • the processing unit 11 determines whether the determination of power consumption has been completed for all units included in the semiconductor manufacturing apparatus 2B (step S47). If the determination of power consumption has not been completed for all units (S47: NO), the processing unit 11 returns the process to step S44, selects another unit, and repeats the above process.
  • the processing unit 11 determines whether the semiconductor manufacturing process of the semiconductor manufacturing apparatus 2B has been completed (step S48). If the semiconductor manufacturing process has not been completed (S48: NO), the processing unit 11 returns the process to step S41, acquires new operating data from the semiconductor manufacturing apparatus 2B, and repeats the above-described process. If the semiconductor manufacturing process has ended (S48: YES), the processing unit 11 ends the process.
  • FIG. 12 is a schematic diagram for explaining the effect of automatic stopping of the standby unit.
  • changes in power consumption are shown in graphs for three units, an RF power supply unit, a dry pump unit, and a heater, among the plurality of units included in the semiconductor manufacturing apparatus 2B.
  • Each graph is a line graph with time on the horizontal axis and power consumption of the unit on the vertical axis.
  • the left side of FIG. 12 shows a graph when the standby unit is not automatically stopped, and the right side is a graph when the standby unit is automatically stopped.
  • each unit consumes standby power (the area shown in light gray in this figure) even if it is in a standby state where no processing is actually performed. continues to consume.
  • a value with a margin for this standby power is set in advance as a threshold, and the information processing device 1B determines whether each unit is in a standby state based on this threshold, and the power consumption is set to the threshold.
  • the information processing device 1B automatically stops the RF power supply unit and dry pump unit that are determined to be in the standby state, and the power consumption of these units is lower than the standby power amount. It has been reduced to a small value, for example 0.
  • the information processing apparatus 1B that monitors and controls the semiconductor manufacturing equipment 2B, which is not equipped with the sensor 3, uses the prediction model 5 to automatically stop the unit. Similar processing may be performed for the information processing device 1A that monitors and controls the semiconductor manufacturing device 2A. However, the information processing device 1A may perform a similar automatic stop using sensor data output from the sensor 3 without using the prediction model 5.
  • the user can verify the processing timing of a plurality of units included in the semiconductor manufacturing apparatus 2 based on the sensor data output by the prediction model 5.
  • the information processing device 1B stores, for example, operational data from the start to the end of the semiconductor manufacturing process for one wafer from among the operational data acquired when the semiconductor manufacturing device 2B performed the semiconductor manufacturing process and stored in the process DB 12b. read out.
  • the information processing device 1B inputs the operation data read from the process DB 12b into the prediction model 5, and acquires the sensor data output from the prediction model 5.
  • the semiconductor manufacturing apparatus 2B is configured with a plurality of units, and for example, the sensor data includes predicted values such as power consumption, amount of water discharged, or amount of gas discharged for each of these units. Regarding the amount of water discharged or the amount of gas discharged, it is sufficient that the sensor data includes a value only for the unit that performs drainage or gas discharge. Regarding power consumption, it is preferable that sensor data includes values for all units that consume power.
  • the information processing device 1B creates a graph showing temporal changes in the power consumption, water discharge amount, or gas emission amount of each unit from the sensor data output by the prediction model 5, and displays the plurality of graphs created for the plurality of units. They are displayed side by side in section 15.
  • FIG. 13 is a schematic diagram showing a display example related to processing timing verification.
  • the information processing device 1B creates graphs showing temporal changes in power consumption for each of the three units 1 to 3 included in the semiconductor manufacturing device 2B, and displays the three graphs vertically side by side. Further, the information processing device 1B displays a graph of the total power consumption of the three units 1 to 3 below these three graphs.
  • Each graph of Units 1 to 3 and the graph of total power consumption are graphs with time on the horizontal axis and power consumption on the vertical axis, and the scales of the vertical and horizontal axes of these four graphs are the same. is preferred.
  • the information processing device 1B uses a graph based on the sensor data output by the prediction model 5 as a graph of the current power consumption, creates a graph of power consumption when changing the processing timing of each unit, for example, and generates a graph of the current power consumption. and the graph of power consumption after the timing change are displayed side by side on the left and right sides of the display unit 15.
  • the graph of power consumption after the timing change are displayed side by side on the left and right sides of the display unit 15.
  • the maximum value of total power consumption has reached approximately 15kW.
  • the three units 1 to 3 perform processing at different timings, and the maximum value of the total power consumption is suppressed to 12 kW or less.
  • the information processing device 1B receives from the user an input of information regarding how much to change the processing of each unit with respect to the current power consumption graph, and changes the current status of each of the units 1 to 3 according to the received information.
  • the information processing device 1B creates a graph of the total power consumption based on the graph in which the processing timing of each unit 1 to 3 is changed, and displays these graphs side by side on the display unit 15.
  • the information processing device 1B may, for example, search for the timing of processing where the maximum value of total power consumption is the smallest, and create and display a graph at this timing.
  • the graph after the timing change is not created by moving the current graph in the time axis direction, but may be created using the simulator 7 shown in FIG. 8, for example.
  • FIG. 14 is a schematic diagram showing a display example related to processing timing verification.
  • the information processing device 1B displays a graph showing temporal changes in the set temperature of the semiconductor manufacturing process and a graph showing temporal changes in the power consumption of the semiconductor manufacturing device 2B side by side. There is.
  • the information processing device 1B extracts information on the set temperature from recipe data included in the operating data acquired from the semiconductor manufacturing device 2B, and creates a graph showing changes in the set temperature over time based on the extracted information.
  • the information processing device 1B inputs the driving data to the prediction model 5, acquires the sensor data output by the prediction model 5, extracts power consumption information included in the acquired sensor data, and based on the extracted information, A graph showing temporal changes in power consumption of the semiconductor manufacturing apparatus 2B is created.
  • the information processing device 1 creates a graph of the set temperature and power consumption when the timing of changing the set temperature of the semiconductor manufacturing device 2B is changed, for example, with respect to the current set temperature and power consumption graph, and The graph and the graph after the timing change are displayed side by side on the left and right sides of the display section 15.
  • the graph displayed on the left side of the illustrated example shows that currently low-temperature processing and high-temperature processing are performed alternately with respect to the set temperature, and power consumption increases when switching from low-temperature processing to high-temperature processing.
  • the timing of changing the temperature setting is changed so that low-temperature treatment is first performed continuously, then switching from low-temperature treatment to high-temperature treatment is performed, and high-temperature treatment is performed continuously. are doing.
  • the information processing device 1B receives, for example, from the user an input of a new set temperature change pattern for the current set temperature change pattern, and performs a simulation using, for example, the simulator 7 according to the received content.
  • the information processing device 1B acquires operational data output by the simulator 7 as a simulation result, inputs the acquired operational data to the prediction model 5, acquires sensor data, and generates a graph of power consumption.
  • processing related to processing timing verification may be performed in a device different from the information processing device 1 that monitors and controls the semiconductor manufacturing device 2, and may be performed in an information processing device on which the simulator 7 operates, for example. suitable.
  • the processing device 1A acquires, based on the acquired driving data and sensor data, a prediction model 5 that outputs a predicted value of the sensor data according to the input of the driving data, which is generated by machine learning.
  • the information processing device 1A transmits predictions to the semiconductor manufacturing device 2A by transmitting the generated prediction model 5 to the information processing device 1B that monitors and controls the semiconductor manufacturing device 2B in which the sensor 3 is not installed. Place model 5.
  • the semiconductor manufacturing equipment 2B provided with the sensor 3 predicted values of sensor data based on operating data are acquired and used for various controls, verification, etc. be able to.
  • the information processing device 1B acquires operating data of the semiconductor manufacturing device 2B, and uses a prediction model 5 that has been subjected to machine learning so as to output sensor data in accordance with the input of the operating data. , inputs the acquired driving data, acquires the sensor data output by the prediction model 5, and outputs information regarding the sensor data. As a result, the information processing device 1B predicts sensor data using the prediction model 5 generated in advance based on the operation data obtained from the semiconductor manufacturing device 2B in which the sensor 3 is not installed, and monitors the semiconductor manufacturing device 2B. and control.
  • the plurality of sensors 3 include a sensor that measures power consumption, a sensor that measures the amount of water discharged, or a sensor that measures the amount of gas discharged.
  • the operation data also includes process log data output by the semiconductor manufacturing equipment 2, recipe data set in the semiconductor manufacturing equipment 2, wafer transfer history data by the semiconductor manufacturing equipment 2, or error data output by the semiconductor manufacturing equipment 2. It is preferable to include the following.
  • the information processing device 1B calculates the degree of contribution of each item of driving data to the sensor data output by the prediction model 5, and outputs information regarding the calculated degree of contribution.
  • the user can expect to understand which of the multiple items included in the operation data has a high degree of contribution to the sensor data, and to study recipes for semiconductor manufacturing processes, for example.
  • the information processing apparatus 1B determines the standby state of the plurality of units included in the semiconductor manufacturing apparatus 2B based on the sensor data output by the prediction model 5, and operates the units in the standby state. stop. This can be expected to further reduce the power consumption of the semiconductor manufacturing apparatus 2B.
  • the information processing system outputs information regarding temporal changes in power consumption of a plurality of units included in the semiconductor manufacturing apparatus 2B based on sensor data output by the prediction model 5. This allows the user to understand, for example, the peak power consumption of each unit, and configure the semiconductor manufacturing process of the semiconductor manufacturing equipment 2B to avoid the peak power consumption of multiple units from overlapping in time. , it can be expected that the peak power consumption of the semiconductor manufacturing apparatus 2B can be further reduced.
  • the information processing device 1A transmits the generated prediction model 5 to the information processing device 1B via the network N, but the present invention is not limited to this.
  • the prediction model 5 may be exchanged between the information processing apparatus 1A and the information processing apparatus 1B via a recording medium, or the prediction model 5 may be exchanged by a method other than these methods.
  • the prediction model 5 is not generated by the information processing device 1A that monitors and controls the semiconductor manufacturing device 2A, but by the information processing device 1A, such as a server device different from the information processing devices 1A and 1B.
  • the prediction model 5 may be generated by acquiring driving data and sensor data from the information processing apparatus 1B, and an information processing apparatus such as a server apparatus may transmit the prediction model 5 to the information processing apparatus 1B.
  • the plurality of sensors include a sensor that measures power consumption, a sensor that measures drainage amount, or a sensor that measures gas emissions.
  • the operation data includes process log data output by the first semiconductor manufacturing apparatus, processing recipe data set in the first semiconductor manufacturing apparatus, wafer transfer history data by the first semiconductor manufacturing apparatus, or the first semiconductor manufacturing apparatus.
  • 1 Contains error data output by semiconductor manufacturing equipment, The information processing method described in Supplementary note 1 or 2.
  • the number of sensors provided in the second semiconductor manufacturing device is smaller than the number of sensors provided in the first semiconductor manufacturing device.
  • the learning model outputs the power consumption, the amount of water discharged, or the amount of gas discharged by the semiconductor manufacturing device when process log data output by the semiconductor manufacturing device is input.
  • the first semiconductor manufacturing equipment that is being operated on a trial basis includes a predetermined number of sensors that measure sensor data related to the environment, including a sensor that measures power consumption, a sensor that measures water discharge, and a sensor that measures gas emissions. It is provided, When newly installing the second semiconductor manufacturing equipment at one or more bases, the generated learning model may be installed in the second semiconductor manufacturing equipment with fewer or no sensors than the predetermined number. placed in the second semiconductor manufacturing equipment, inputting operational data acquired during initial operation of the second semiconductor manufacturing apparatus into the learning model, and acquiring sensor data without using the sensor; The information processing method described in any one of Supplementary notes 1 to 5.
  • the learning model is based on operating data of the first semiconductor manufacturing apparatus obtained when operating a first semiconductor manufacturing apparatus provided with a plurality of sensors, and sensor data output from the sensors. generated by machine learning to output the sensor data according to the input of The computer program according to appendix 7 or appendix 8.
  • the plurality of sensors include a sensor that measures power consumption, a sensor that measures drainage amount, or a sensor that measures gas emissions.
  • the computer program according to any one of Supplementary notes 7 to 9.
  • the operation data includes process log data output by the first semiconductor manufacturing apparatus, processing recipe data set in the first semiconductor manufacturing apparatus, wafer transfer history data by the first semiconductor manufacturing apparatus, or the first semiconductor manufacturing apparatus.
  • 1 Contains error data output by semiconductor manufacturing equipment, The computer program according to any one of Supplementary notes 7 to 10.
  • the learning model outputs the power consumption, the amount of water discharged, or the amount of gas discharged by the semiconductor manufacturing device when process log data output by the semiconductor manufacturing device is input.
  • the computer program according to any one of Supplementary notes 7 to 11.
  • the operation data includes a plurality of items, Calculating the degree of contribution of each item of the driving data to the sensor data output by the learning model, outputting the calculated contribution degree;
  • the computer program according to any one of appendices 7 to 12.
  • the sensor data includes data on power consumption of each unit for a plurality of units included in the second semiconductor manufacturing apparatus, outputting temporal changes in power consumption of a plurality of units based on the sensor data output by the learning model;
  • the computer program according to any one of Supplementary notes 7 to 14.
  • An information processing device comprising: an output unit that outputs information related to the generated learning model as information for placement in a second semiconductor manufacturing device different from the first semiconductor manufacturing device.
  • An information processing device comprising: an output unit that outputs the acquired sensor data.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Semiconductor Integrated Circuits (AREA)
PCT/JP2023/024372 2022-07-04 2023-06-30 情報処理方法、コンピュータプログラム及び情報処理装置 Ceased WO2024009902A1 (ja)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2024532095A JPWO2024009902A1 (https=) 2022-07-04 2023-06-30
KR1020257002178A KR20250033233A (ko) 2022-07-04 2023-06-30 정보 처리 방법, 컴퓨터 프로그램 및 정보 처리 장치
CN202380049967.6A CN119487613A (zh) 2022-07-04 2023-06-30 信息处理方法、计算机程序以及信息处理装置
TW112124884A TW202420403A (zh) 2022-07-04 2023-07-04 資訊處理方法、電腦程式及資訊處理裝置
US19/002,811 US20250124357A1 (en) 2022-07-04 2024-12-27 Information processing method, computer program, and information processing apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2022-107851 2022-07-04
JP2022107851 2022-07-04

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US19/002,811 Continuation US20250124357A1 (en) 2022-07-04 2024-12-27 Information processing method, computer program, and information processing apparatus

Publications (1)

Publication Number Publication Date
WO2024009902A1 true WO2024009902A1 (ja) 2024-01-11

Family

ID=89453506

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2023/024372 Ceased WO2024009902A1 (ja) 2022-07-04 2023-06-30 情報処理方法、コンピュータプログラム及び情報処理装置

Country Status (6)

Country Link
US (1) US20250124357A1 (https=)
JP (1) JPWO2024009902A1 (https=)
KR (1) KR20250033233A (https=)
CN (1) CN119487613A (https=)
TW (1) TW202420403A (https=)
WO (1) WO2024009902A1 (https=)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021157453A1 (ja) * 2020-02-07 2021-08-12 東京エレクトロン株式会社 プロセス推定システム、プロセスデータ推定方法及びプログラム
JP2021132140A (ja) * 2020-02-20 2021-09-09 東京エレクトロン株式会社 情報処理システム及びシミュレーション方法
WO2022044335A1 (ja) * 2020-08-31 2022-03-03 富士通株式会社 モデル生成プログラム、モデル生成方法及びモデル生成装置

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150253762A1 (en) 2012-09-26 2015-09-10 Hitachi Kokusai Electric Inc. Integrated management system, management device, method of displaying information for substrate processing apparatus, and recording medium

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021157453A1 (ja) * 2020-02-07 2021-08-12 東京エレクトロン株式会社 プロセス推定システム、プロセスデータ推定方法及びプログラム
JP2021132140A (ja) * 2020-02-20 2021-09-09 東京エレクトロン株式会社 情報処理システム及びシミュレーション方法
WO2022044335A1 (ja) * 2020-08-31 2022-03-03 富士通株式会社 モデル生成プログラム、モデル生成方法及びモデル生成装置

Also Published As

Publication number Publication date
CN119487613A (zh) 2025-02-18
JPWO2024009902A1 (https=) 2024-01-11
US20250124357A1 (en) 2025-04-17
TW202420403A (zh) 2024-05-16
KR20250033233A (ko) 2025-03-07

Similar Documents

Publication Publication Date Title
JP4740142B2 (ja) 半導体製造プロセスを容易にする第１の原理シミュレーションを用いたシステム及び方法
JP6847787B2 (ja) 情報処理装置、情報処理方法及びコンピュータプログラム
TWI663510B (zh) 設備保養預測系統及其操作方法
CN118898205A (zh) 一种基于数字孪生的设备故障检测方法及系统
JP2019021186A (ja) データ処理装置、制御システム、データ処理方法及びプログラム
JP2018025932A (ja) センサと機械学習部を備えた作業管理システム
JP2016164772A (ja) プロセス監視装置、プロセス監視方法及びプログラム
Korambath et al. A smart manufacturing use case: Furnace temperature balancing in steam methane reforming process via kepler workflows
CN105793789A (zh) 用于过程单元中的全部过程区段的自动的监视和状态确定的计算机实现的方法和系统
JP2020004041A (ja) 消費電力推定装置
WO2021210353A1 (ja) 故障予兆システム
CN118497722B (zh) Mpcvd金刚石培育过程中的高温电离控制系统
JP5761662B2 (ja) ビルエネルギー管理装置及びビルエネルギー管理方法
CN110340884B (zh) 测定动作参数调整装置、机器学习装置以及系统
WO2018138973A1 (ja) 系統運用意思決定支援装置および方法
WO2024009902A1 (ja) 情報処理方法、コンピュータプログラム及び情報処理装置
US20250231535A1 (en) Information processing method, computer program, and information processing apparatus
TWI791949B (zh) 監視裝置、顯示裝置、監視方法及監視程式
WO2024024633A1 (ja) 情報処理方法、コンピュータプログラム及び情報処理装置
JP5843716B2 (ja) 電力管理システム、電力管理方法、統合管理装置及びプログラム
JP2015225637A (ja) 相関分析装置、相関分析方法、および相関分析用プログラム
JP2015064816A (ja) エネルギー削減量予測方法および装置
JP6829271B2 (ja) 測定動作パラメータ調整装置、機械学習装置及びシステム
Das et al. Segmentation analysis in human centric cyber-physical systems using graphical lasso
CN117321969A (zh) 可编程逻辑控制器及可编程逻辑控制器的动作方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23835434

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2024532095

Country of ref document: JP

WWE Wipo information: entry into national phase

Ref document number: 202380049967.6

Country of ref document: CN

ENP Entry into the national phase

Ref document number: 20257002178

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWP Wipo information: published in national office

Ref document number: 202380049967.6

Country of ref document: CN

WWP Wipo information: published in national office

Ref document number: 1020257002178

Country of ref document: KR

122 Ep: pct application non-entry in european phase

Ref document number: 23835434

Country of ref document: EP

Kind code of ref document: A1