Information processing device, inference device, machine learning device, information processing method, inference method, and machine learning method

Abstract

[Problem] To provide an information processing device capable of appropriately predicting a surface state of a substrate during or after processing by a polishing process. [Solution] An information processing device 5 comprises: an information acquisition unit 500 that acquires polishing process information including device state information indicating a state of a substrate polishing device that performs a polishing process of polishing a substrate with a polishing pad; and an information generation unit 501 that generates surface state information indicating a surface state of the substrate when the polishing process has been performed by the substrate polishing device in the state indicated by the device state information included in the polishing process information, by inputting the polishing process information acquired by the information acquisition unit 500 to a learning model 10. The learning model 10 is a trained model in which a correlation between the polishing process information and the surface state information is learned by machine learning.

Description

Information processing apparatus, inference apparatus, machine learning apparatus, information processing method, inference method, and machine learning method

[0001] The present invention relates to an information processing apparatus, an inference apparatus, a machine learning apparatus, an information processing method, an inference method, and a machine learning method.

[0002] As one type of substrate polishing apparatus that performs various types of processing on a substrate such as a semiconductor wafer, a substrate polishing apparatus that performs chemical mechanical polishing (CMP: Chemical Mechanical Polishing) processing is known. In chemical mechanical polishing processing (hereinafter referred to as "polishing processing"), for example, while rotating a polishing table section having a polishing pad, with polishing liquid (slurry) supplied from a polishing fluid supply nozzle to the polishing pad, the substrate is pressed against the polishing pad by a polishing head section called a top ring, whereby the substrate is polished chemically and mechanically (see, for example, Patent Document 1).

[0003] Japanese Unexamined Patent Publication No. 2001-179613

[0004] Elements such as wiring portions, insulating portions, and transistors are formed on a substrate subjected to polishing processing. In polishing processing, if polishing of the substrate is locally insufficient or excessive, variations in the film thickness of the substrate or level differences occur on the substrate. As a result, the performance of the substrate (signal propagation, power consumption, short-circuiting, etc.) deteriorates, which is one factor that reduces product yield. On the other hand, although it is possible to measure the surface state of a substrate after polishing processing by a measuring instrument, it is difficult to measure all substrates. Therefore, means for easily and highly accurately predicting the surface state of a substrate is desired.

[0005] In view of the above problems, an object of the present invention is to provide an information processing apparatus, an inference apparatus, a machine learning apparatus, an information processing method, an inference method, and a machine learning method that enable appropriate prediction of the surface state of a substrate during or after polishing processing.

[0006] To achieve the above objective, an information processing apparatus according to one aspect of the present invention includes: an information acquisition unit that acquires polishing process information including device state information indicating the state of a substrate polishing apparatus that performs polishing processing on a substrate using a polishing pad; and an information generation unit that inputs the polishing process information acquired by the information acquisition unit into a learning model to generate surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing apparatus in the state indicated by the device state information included in the polishing process information, wherein the learning model is a trained model that has learned the correlation between the polishing process information and the surface state information by machine learning.

[0007] According to an information processing apparatus according to one aspect of the present invention, polishing process information, including apparatus status information indicating the state of the substrate polishing apparatus, is input to a learning model, thereby generating surface state information for said polishing process information. Therefore, the surface state of the substrate during or after the polishing process can be appropriately predicted.

[0008] Other issues, configurations, and effects will be clarified in the embodiments for carrying out the invention described later.

[0009] This is an overall configuration diagram showing an example of a substrate polishing system 1. This is a schematic configuration diagram showing an example of a substrate polishing apparatus 2. This is a schematic diagram showing an example of a polishing head unit 202. This is a block diagram showing an example of a substrate polishing apparatus 2. This is a hardware configuration diagram showing an example of a computer 900. This is a data configuration diagram showing an example of a database 30. This is a block diagram showing an example of a machine learning apparatus 4. This is a diagram showing an example of a learning model 10 and learning data 11. This is a diagram showing an example of film thickness state information and step state information included in surface state information. This is a flowchart showing an example of a machine learning method by the machine learning apparatus 4. This is a block diagram showing an example of an information processing apparatus 5. This is a functional explanatory diagram showing an example of an information processing apparatus 5. This is a flowchart showing an example of an information processing method by the information processing apparatus 5.

[0010] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. In the following, the scope necessary for explaining how to achieve the objectives of the present invention will be schematically shown, and the scope necessary for explaining the relevant parts of the present invention will be mainly explained, with any parts that are omitted from explanation being based on prior art.

[0011] (Substrate Polishing System 1) Figure 1 is an overall configuration diagram showing an example of a substrate polishing system 1. The substrate polishing system 1 according to this embodiment functions as a system for managing various information related to a polishing process that polishes the surface of a substrate (hereinafter referred to as "wafer") W such as a semiconductor wafer to make it flat.

[0012] The substrate polishing system 1 comprises, as its main components, a substrate polishing device 2, a database device 3, a machine learning device 4, an information processing device 5, and a user terminal device 6. Each device 2 to 6 is, for example, composed of a general-purpose or dedicated computer (see Figure 5 below) and connected to a wired or wireless network 7, enabling the mutual transmission and reception of various types of data. The number of each device 2 to 6 and the connection configuration of the network 7 are not limited to the example in Figure 1 and may be changed as appropriate.

[0013] The substrate polishing apparatus 2 is an apparatus that performs a polishing process to polish a wafer W. In doing so, the substrate polishing apparatus 2 controls the operation of each part while referring to apparatus setting information 215 consisting of multiple apparatus parameters and substrate recipe information 216 which defines the polishing conditions for the polishing process.

[0014] Furthermore, the substrate polishing apparatus 2 stores various information as operation history information 217 in accordance with the polishing process, and transmits the operation history information 217 as a report R to the database device 3. The operation history information 217 includes, for example, device status information indicating the state of the substrate polishing apparatus 2 when the polishing process was performed, consumable status information indicating the state of consumables used in the polishing process, event information detected by the substrate polishing apparatus 2, and user (operator, production manager, maintenance manager, etc.) operation information for the substrate polishing apparatus 2.

[0015] The database device 3 includes a database 30 that stores reports R (operation history information 217) transmitted from each substrate polishing device 2. In addition to the reports R transmitted from each substrate polishing device 2, the database 30 may also store information when polishing tests are performed using a dummy wafer for testing. Polishing tests may be performed by the substrate polishing device 2, by a polishing test device capable of reproducing the polishing process, or by a polishing simulation device capable of simulating the polishing process. Furthermore, the database 30 may also store device setting information 215 and substrate recipe information 216.

[0016] The machine learning device 4, for example, acquires a portion of the information stored in the database 30 as training data 11 and generates a training model 10 to be used in the substrate polishing device 2 using machine learning. The trained training model 10 (trained model) is provided to the information processing device 5 via the network 7 or a recording medium.

[0017] The information processing device 5 operates as the main component of the machine learning inference phase. Using the learning model 10 generated by the machine learning device 4, it predicts the surface state of the wafer W when the substrate polishing device 2 performs polishing on the wafer W for mass production, and transmits the predicted surface state information to the database device 3, user terminal device 6, etc. The timing of the information processing device 5 predicting the surface state information may be after the polishing process is performed (post-prediction processing), during the polishing process is performed (real-time prediction processing), or before the polishing process is performed (pre-prediction processing).

[0018] The user terminal device 6 is a terminal device used by the user, and may be a stationary device or a portable device. The user terminal device 6 accepts various input operations via a display screen such as an application program or a web browser, and displays various information (for example, event notifications, device setting information 215, board recipe information 216, operation history information 217, database 30, surface condition information, etc.) via the display screen. Some of the information may be editable via the user terminal device 6.

[0019] (Substrate Polishing Apparatus 2) Figure 2 is a schematic diagram showing an example of a substrate polishing apparatus 2. The substrate polishing apparatus 2 comprises a polishing unit 20 that performs polishing of the wafer W and a control unit 21 that controls the operation of the substrate polishing apparatus 2. Note that the substrate polishing apparatus 2 only needs to have at least one polishing unit 20. In addition to the polishing unit 20, the substrate polishing apparatus 2 may also include, for example, a substrate transport unit for transporting the wafer W, a cleaning unit for cleaning the wafer W, etc.

[0020] The polishing unit 20 includes a polishing table section 201 that rotatably supports a polishing pad 200a having a polishing surface, a polishing head section 202 (top ring) that presses the wafer W against the polishing surface of the polishing pad 200a to polish the wafer W, a polishing fluid supply section 203 that supplies polishing fluid to the polishing pad 200a, a dressing section 204 (dresser) that dresses the polishing surface of the polishing pad 200a, a cleaning fluid spraying section 205 (atomizer) that sprays cleaning fluid onto the polishing pad 200a, and an environmental measuring section 206 that measures the state of the processing environment in which the polishing process is carried out.

[0021] The polishing table section 201 includes a polishing table 201a to which a polishing pad 200a is removably attached, a polishing table shaft 201b that supports the polishing table 201a, and a rotational movement mechanism section 201c that rotates the polishing table 201a around the axis of the polishing table shaft 201b.

[0022] The polishing head unit 202 includes a polishing head 202a that holds the wafer W, a polishing head shaft 202b that supports the polishing head 202a, a rotational movement mechanism 202c that rotates the polishing head 202a around the axis of the polishing head shaft 202b, a vertical movement mechanism 202d that moves the polishing head 202a in the vertical direction, a support arm 202e that supports the polishing head shaft 202b, a support shaft 202f that supports the support arm 202e, and a swaying movement mechanism 202g that swivels (oscillates) the polishing head 202a around the support shaft 202f as the pivot point.

[0023] The polishing fluid supply unit 203 includes a polishing fluid supply nozzle 203a capable of supplying polishing fluid, a support shaft 203b supporting the polishing fluid supply nozzle 203a, a swinging movement mechanism 203c that swings (oscillates) the polishing fluid supply nozzle 203a around the support shaft 203b as the pivot point, a flow rate adjustment unit 203d that adjusts the flow rate of the polishing fluid, and a temperature control mechanism 203e that adjusts the temperature of the polishing fluid. The polishing fluid is a polishing liquid (slurry) or pure water, and may also contain a chemical solution, or may be a polishing liquid with a dispersant added.

[0024] The dressing section 204 includes a dressing head 204a to which a dressing pad 200b is detachably attached, a dressing shaft 204b that supports the dressing head 204a, a rotational movement mechanism 204c that rotates the dressing head 204a around the axis of the dressing shaft 204b, a vertical movement mechanism 204d that moves the dressing head 204a in the vertical direction, a support arm 204e that supports the dressing shaft 204b, a support shaft 204f that supports the support arm 204e, and a swinging movement mechanism 204g that swings (oscillates) the dressing head 204a around the support shaft 204f as the pivot point.

[0025] The cleaning fluid injection unit 205 includes a cleaning fluid injection nozzle 205a capable of injecting cleaning fluid, a support shaft 205b supporting the cleaning fluid injection nozzle 205a, a swinging movement mechanism 205c that pivots (oscillates) the cleaning fluid injection nozzle 205a around the support shaft 205b, a flow rate adjustment unit 205d that adjusts the flow rate of the cleaning fluid, and a temperature control mechanism 205e that adjusts the temperature of the cleaning fluid. The cleaning fluid is a mixed fluid of liquid (e.g., pure water) and gas (e.g., nitrogen gas) or liquid (e.g., pure water).

[0026] The environmental measurement unit 206 is equipped with various sensors to measure the state of the processing environment in which the polishing process is performed. The environmental measurement unit 206 consists of sensors that measure the temperature, humidity, and atmospheric pressure of the processing environment, but is not limited to these.

[0027] In Figure 2, the specific configurations of the rotational movement mechanisms 201c, 202c, 204c, the vertical movement mechanisms 202d, 204d, and the oscillating movement mechanisms 202g, 203c, 204g, 205c are omitted. However, they are configured by appropriately combining, for example, a module for generating driving force such as a motor or fluid pressure cylinder, a driving force transmission mechanism such as a linear guide, ball screw, gear, belt, coupling, or bearing, and sensors such as a linear sensor, encoder sensor, limit sensor, or torque sensor. In Figure 2, the specific configurations of the flow rate adjustment units 203d and 205d are omitted. However, they are configured by appropriately combining, for example, a module for adjusting fluids such as a pump, valve, or regulator, and sensors such as a flow rate sensor, pressure sensor, or liquid level sensor. Although the specific configuration of the temperature control mechanisms 203e and 205e is omitted in Figure 2, they are configured by appropriately combining, for example, temperature control modules (contact or non-contact type) such as heaters and heat exchangers with sensors such as temperature sensors.

[0028] (Polishing head section 202) Figure 3 is a schematic diagram showing an example of the polishing head section 202. The polishing head 202a includes a top ring body 2020 attached to the polishing head shaft 202b, a substantially disc-shaped carrier 2021 housed in the top ring body 2020, a membrane 2022 positioned below the carrier 2021 to press the wafer W against the polishing pad 200a, a substantially annular retainer ring 2023 positioned on the outer circumference of the carrier 2021 and the membrane 2022 to directly press the polishing pad 200a, and a retainer ring airbag 2024 positioned between the top ring body 2020 and the retainer ring 2023 to press the retainer ring 2023 against the polishing pad 200a.

[0029] The membrane 2022 is made of an elastic film and has a plurality of concentric partition walls 2022e inside, thereby having first to fourth membrane pressure chambers 2022a to 2022d arranged concentrically from the center outward toward the outer circumference of the top ring body 2020. The membrane 2022 also has a plurality of holes 2022f on its lower surface for adsorption of the wafer W and functions as a substrate holding surface for holding the wafer W. The retainer ring airbag 2024 is made of an elastic film and has a retainer ring pressure chamber 2024a inside. The configuration of the polishing head 202a may be changed as appropriate, and it may have a pressure chamber that presses the entire carrier 2021, and the number and shape of the membrane pressure chambers in the membrane 2022 may be changed as appropriate, and the number and arrangement of the adsorption holes 2022f may be changed as appropriate. The membrane 2022 may also not have the adsorption holes 2022f.

[0030] The first to fourth membrane pressure chambers 2022a to 2022d are connected to the first to fourth flow paths 2026A to 2026D, respectively, and the retainer ring pressure chamber 2024a is connected to the fifth flow path 2026E. The first to fifth flow paths 2026A to 2026E communicate with the outside via a rotary joint 2025 provided on the polishing head shaft 202b, and branch into the first branch flow paths 2027A to 2027E and the second branch flow paths 2028A to 2028E, respectively. Pressure sensors PA to PE are installed in the first to fifth flow paths 2026A to 2026E, respectively. The first branch flow paths 2027A to 2027E are connected to a gas supply source GS for pressurized fluid (air, nitrogen, etc.) via valves V1A to V1E, flow sensors FA to FE, and pressure regulators RA to RE. The second branch channels 2028A to 2028E are connected to the vacuum source VS via valves V2A to V2E, and are configured to communicate with the atmosphere via valves V3A to V3E.

[0031] The wafer W is held by suction on the lower surface of the polishing head 202a and moved to a predetermined polishing position on the polishing table section 201. After that, it is polished by being pressed against the polishing surface of the polishing pad 200a, which is supplied with polishing fluid from the polishing fluid supply section 203, by the polishing head 202a. At this time, the polishing head 202a independently controls the pressure regulators RA to RE to adjust the pressing force that presses the wafer W against the polishing pad 200a using the pressure fluid supplied to the first to fourth membrane pressure chambers 2022a to 2022d for each region of the wafer W, and also adjusts the pressing force that presses the retainer ring 2023 against the polishing pad 200a using the pressure fluid supplied to the retainer ring pressure chamber 2024a. The pressure of the pressurized fluid supplied to the first to fourth membrane pressure chambers 2022a to 2022d and the retainer ring pressure chamber 2024a is measured by pressure sensors PA to PE, respectively, and the flow rate of the pressurized gas is measured by flow sensors FA to FE, respectively.

[0032] In the example shown in Figure 3, the retainer ring 2023 is pressed by the supply of pressurized fluid to the retainer ring airbag 2024. However, it may also be pressed by a drive force generating module such as a motor or a fluid pressure cylinder. In that case, the pressing force of the retainer ring 2023 is adjusted based on the command values ​​for the motor or fluid pressure cylinder.

[0033] Figure 4 is a block diagram showing an example of a substrate polishing apparatus 2. The control unit 21 is electrically connected to each part 201 to 206 of the polishing unit 20 and comprehensively controls the polishing unit 20.

[0034] The polishing unit 20 includes a plurality of modules 20a which are to be controlled and are arranged in each of the parts 201 to 206 of the polishing unit 20, a plurality of sensors 20b which are arranged in each of the plurality of modules 20a which detect data (detected values ​​and measured values) necessary for controlling each module 20a, and a sequencer 20c which controls the operation of each module 20a based on the detected values ​​and measured values ​​of each sensor 20b.

[0035] The sensors 20b of the polishing unit 20 include, for example, sensors for detecting the rotational speed, rotational torque, and surface temperature of the polishing table 201, a sensor for detecting the surface roughness of the polishing pad 200a, sensors for detecting the rotational speed, rotational torque, oscillation speed, oscillation torque, and height of the polishing head 202, a sensor for detecting the oscillation position of the polishing head 202 which can be converted into the polishing position of the polishing head 202, a sensor for detecting the lifting torque of the polishing head 202 which can be converted into the pressing load of the polishing head 202, sensors for detecting the pressure and flow rate of the pressurized fluid in the first to fourth membrane pressure chambers 2022a to 202d which can be converted into the pressing force of the membrane 2022, a sensor for detecting the pressure and flow rate of the pressurized fluid in the retainer ring pressure chamber 2024a which can be converted into the pressing force of the retainer ring 2023, and polishing This includes sensors for detecting the flow rate, pressure, and temperature of the polishing fluid supplied from the fluid supply unit 203; sensors for detecting the oscillation position of the polishing fluid supply unit 203 which can be converted into a dropping position for the polishing fluid; sensors for detecting the rotation speed, rotation torque, oscillation torque, and height of the dressing unit 204; sensors for detecting the oscillation position of the dressing unit 204 which can be converted into a dressing position for the dressing unit 204; sensors for detecting the lifting torque of the dressing unit 204 which can be converted into a pressing load for the dressing unit 204; sensors for detecting the surface roughness of the dressing pad 200b; sensors for detecting the flow rate, pressure, and temperature of the cleaning fluid supplied from the cleaning fluid injection unit 205; sensors for detecting the oscillation position of the cleaning fluid injection unit 205 which can be converted into a dropping position for the cleaning fluid; and sensors in the environmental measurement unit 206.

[0036] The control unit 21 comprises a control unit 210, a communication unit 211, an input unit 212, an output unit 213, and a storage unit 214. The control unit 21 is composed of, for example, a general-purpose or dedicated computer (see Figure 5, described later).

[0037] The communication unit 211 is connected to the network 7 and functions as a communication interface for sending and receiving various types of data. The input unit 212 accepts various input operations, and the output unit 213 functions as a user interface by outputting various types of information via a display screen, signal tower illumination, and buzzer sound.

[0038] The memory unit 214 stores various programs (operating system (OS), application programs, web browser, etc.) and data (device setting information 215, circuit board recipe information 216, operation history information 217, etc.) used in the operation of the circuit board polishing apparatus 2.

[0039] The control unit 210 acquires detection values ​​and measurement values ​​from multiple sensors 20b (hereinafter referred to as the "sensor group") via the sequencer 20c, and performs polishing processing on the wafer W by coordinating the operation of multiple modules 20a (hereinafter referred to as the "module group"). The control unit 210 also stores the detection values ​​and measurement values ​​of each sensor 20b and the command values ​​of each module 20a as operation history information 217 in the storage unit 214.

[0040] (Hardware configuration of each device) Figure 5 is a hardware configuration diagram showing an example of a computer 900. Each of the substrate polishing device 2 (especially the control unit 21), database device 3, machine learning device 4, information processing device 5, and user terminal device 6 is configured with a general-purpose or dedicated computer 900.

[0041] As shown in Figure 5, the computer 900 comprises, as its main components, a bus 910, a processor 912, a memory 914, an input device 916, an output device 917, a display device 918, a storage device 920, a communication interface unit 922, an external device interface unit 924, an I / O device interface unit 926, and a media input / output unit 928. Note that the above components may be omitted as appropriate depending on the intended use of the computer 900.

[0042] The processor 912 consists of one or more arithmetic processing units (CPU (Central Processing Unit), MPU (Micro-Processing Unit), DSP (Digital Signal Processor), GPU (Graphics Processing Unit), NPU (Neural Processing Unit), etc.) and operates as a control unit that oversees the entire computer 900. The memory 914 stores various data and programs 930 and consists of volatile memory (DRAM, SRAM, etc.) that functions as main memory, and non-volatile memory (ROM), flash memory, etc.

[0043] The input device 916 consists of, for example, a keyboard, mouse, numeric keypad, or electronic pen, and functions as an input unit. The output device 917 consists of, for example, a sound (voice) output device or a vibration device, and functions as an output unit. The display device 918 consists of, for example, a liquid crystal display, an organic EL display, electronic paper, or a projector, and functions as an output unit. The input device 916 and the display device 918 may be configured as an integrated unit, such as a touch panel display. The storage device 920 consists of, for example, an HDD or SSD, and functions as a storage unit. The storage device 920 stores various data necessary for the execution of the operating system and the program 930.

[0044] The communication I / F unit 922 is connected via a wired or wireless connection to a network 940 such as the Internet or an intranet (which may be the same as the network 7 in FIG. 1), and functions as a communication unit that transmits and receives data to and from other computers in accordance with a predetermined communication standard. The external device I / F unit 924 is connected via a wired or wireless connection to an external device 950 such as a camera, a printer, a scanner, or a reader / writer, and functions as a communication unit that transmits and receives data to and from the external device 950 in accordance with a predetermined communication standard. The I / O device I / F unit 926 is connected to an I / O device 960 such as various sensors or actuators, and functions as a communication unit that transmits and receives various signals and data such as detection signals from sensors and control signals to actuators to and from the I / O device 960. The media input / output unit 928 is composed of, for example, a drive device such as a DVD drive or a CD drive, a memory card slot, and a USB connector, and reads and writes data from and to a medium (non-transitory storage medium) 970 such as a DVD, a CD, a memory card, or a USB memory.

[0045] In the computer 900 having the above configuration, the processor 912 calls a program 930 stored in the storage device 920 to the memory 914 and executes the program, and controls each part of the computer 900 via the bus 910. Note that the program 930 may be stored in the memory 914 instead of the storage device 920. The program 930 may be recorded on the medium 970 in an installable file format or an executable file format, and provided to the computer 900 via the media input / output unit 928. The program 930 may be provided to the computer 900 by being downloaded via the network 940 through the communication I / F unit 922. Further, the computer 900 may implement various functions realized by the processor 912 executing the program 930 by hardware such as an FPGA (Field-Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit), for example.

[0046] The computer 900 is configured by, for example, a stationary computer or a portable computer, and is an electronic device of any form. The computer 900 may be a client-type computer, a server-type computer, or a cloud-type computer, or may be an embedded computer called, for example, a control panel, a controller (including a microcomputer, a programmable logic controller, and a sequencer), or the like. The computer 900 may also be applied to devices other than the respective devices 2 to 6.

[0047] (Database 30) FIG. 6 is a data configuration diagram showing an example of the database 30. In the database 30, reports R (operation history information 217) acquired when a polishing process for a wafer W is performed in the substrate polishing apparatus 2 are classified and registered. The database 30 includes, for example, a wafer history table 300 for each wafer W, an apparatus history table 301 for apparatus status information in polishing processing, a consumable item history table 302 for consumable item status information in polishing processing, and a surface status history table 303 for the surface status of the wafer W. In addition to the above, the database 30 includes an event history table for event information, an operation history table for operation information, and the like, but detailed description thereof will be omitted.

[0048] In each record of the wafer history table 300, for example, a wafer ID for identifying the wafer W, a start time of polishing processing, an end time, and the like are registered.

[0049] In each record of the apparatus history table 301, for example, a wafer ID, polishing table status information, polishing head status information, polishing fluid supply status information, dressing status information, cleaning fluid jet status information, environmental information, and the like are registered.

[0050] Each piece of information registered as device status information in the device history table 301 is the detected or measured value of each sensor 20b sampled at predetermined time intervals by each sensor 20b of each part 201 to 206 of the polishing unit 20, or the command value for each module 20a of each part 201 to 206. In this case, the command value for each module 20a is set based on the device setting information 215 and the substrate recipe information 216, but the setting values ​​set in the device setting information 215 and the substrate recipe information 216 may also be registered in the device history table 301. Furthermore, each piece of information registered as device status information in the device history table 301 may be a parameter converted from the detected value of the sensor 20b or the command value for the module 20a, or it may be a parameter calculated based on the detected values ​​of multiple sensors 20b. Details of each piece of information will be described later.

[0051] Each record in the consumables history table 302 registers, for example, a wafer ID and consumables status information. The consumables status information indicates the status of the consumables used in the polishing process. Examples of consumables include the polishing pad 200a, membrane 2022, retainer ring 2023, and dressing pad 200b. The consumables status information may include, but is not limited to, the detected or measured values ​​of each sensor 20b in each unit 201 to 206, or the command values ​​for each module 20a in each unit 201 to 206. Further details of the consumables status information will be described later.

[0052] Each record in the surface condition history table 303 registers, for example, a wafer ID and surface condition information. The surface condition information indicates the surface condition of the wafer W that has undergone polishing. The surface condition information is obtained, for example, by measuring the surface of the wafer W using an external measuring instrument installed outside the substrate polishing apparatus 2. In this case, not all wafers W are measured by the external measuring instrument; only some wafers W are subject to measurement. For some wafers W that are subject to measurement by the external measuring instrument, the measurement results of the surface condition by the external measuring instrument are registered as surface condition information in the surface condition history table 303. For the remaining wafers W that are not subject to measurement by the external measuring instrument, the prediction results of the surface condition by the information processing device 5 are registered as surface condition information in the surface condition history table 303. The surface condition information may also indicate the surface condition of a wafer W that has undergone partial polishing. Details of the surface condition information will be described later.

[0053] (Machine Learning Device 4) Figure 7 is a block diagram showing an example of a machine learning device 4. The machine learning device 4 comprises a control unit 40, a communication unit 41, a training data storage unit 42, and a trained model storage unit 43.

[0054] The control unit 40 functions as a learning data acquisition unit 400 and a machine learning unit 401. The communication unit 41 is connected to external devices (for example, a substrate polishing device 2, a database device 3, an information processing device 5, and a user terminal device 6, a polishing test device (not shown), a polishing simulation device (not shown), etc.) via the network 7 and functions as a communication interface for sending and receiving various types of data.

[0055] The learning data acquisition unit 400 is connected to an external device via the communication unit 41 and the network 7, and acquires learning data 11 consisting of polishing process information as input data and surface state information as output data. The learning data 11 is acquired, for example, by referring to the database 30. The learning data 11 is used as training data, validation data, and test data in supervised learning. The surface state information is used as correct labels in supervised learning.

[0056] The learning data storage unit 42 is a database that stores multiple sets of learning data 11 acquired by the learning data acquisition unit 400. The specific configuration of the database constituting the learning data storage unit 42 can be designed as appropriate.

[0057] The machine learning unit 401 performs machine learning using multiple sets of training data 11. Specifically, the machine learning unit 401 inputs multiple sets of training data 11 into the learning model 10 and generates a trained learning model 10 by having the learning model 10 learn the correlation between polishing processing information included as input data and surface state information included as output data in the training data 11. The machine learning unit 401 may perform predetermined pre-processing on the input data (polishing processing information) input to the learning model 10, or it may perform predetermined post-processing on the output data (surface state information) output from the learning model 10.

[0058] The trained model storage unit 43 is a database that stores the trained model 10 (specifically, the adjusted weight parameter group) generated by the machine learning unit 401. The trained model 10 stored in the trained model storage unit 43 is provided to the actual system (for example, the information processing device 5) via the network 7 or a recording medium. In Figure 7, the training data storage unit 42 and the trained model storage unit 43 are shown as separate storage units, but they may be composed of a single storage unit.

[0059] The number of learning models 10 stored in the learned model storage unit 43 is not limited to one. For example, multiple learning models 10 with different conditions may be stored and used selectively or in parallel, such as differences in machine learning methods, wafer types (size, thickness, film type, etc.), configuration of the polishing unit 20, types of consumables (polishing pads 200a and dressing pads 200b, etc.), types of polishing fluids, types of cleaning fluids, types of data included in polishing process information, and types of data included in surface condition information. In this case, the learning data storage unit 42 should store multiple types of learning data 11, each having a data configuration corresponding to the multiple learning models 10 with different conditions.

[0060] Figure 8 shows an example of a learning model 10 and training data 11. The training data 11 used for machine learning of the learning model 10 consists of polishing process information as input data and surface state information as output data.

[0061] The polishing process information that constitutes the input data for the learning data 11 includes device status information indicating the state of the substrate polishing apparatus 2.

[0062] The device status information includes at least one of the following: polishing table status information indicating the status of the polishing table section 201, polishing head status information indicating the status of the polishing head section 202, and polishing fluid supply status information indicating the status of the polishing fluid supply section 203.

[0063] The polishing table status information includes at least one of the rotational speed of the polishing table 201, the rotational torque of the polishing table 201, and the surface temperature of the polishing pad 200a.

[0064] The polishing head status information includes at least one of the following: the rotational speed of the top ring body 2020, the rotational torque of the top ring body 2020, the oscillation position of the top ring body 2020, the oscillation speed of the top ring body 2020, the oscillation torque of the top ring body 2020, the height of the top ring body 2020, the lifting speed of the top ring body 2020, the lifting torque of the top ring body 2020, the pressing force of the membrane 2022, and the pressing force of the retainer ring 2023.

[0065] The polishing fluid supply status information includes at least one of the flow rate of the polishing fluid, the temperature of the polishing fluid, and the dropping position of the polishing fluid. If the polishing fluid consists of multiple types of polishing fluids (e.g., polishing liquid, pure water, chemical solution, dispersant, etc.), it is sufficient to include at least one of the flow rate, temperature, and dropping position for each type. For example, if the polishing fluid consists of polishing liquid and pure water, it is sufficient to include at least one of the flow rate, temperature, dropping position of the polishing liquid, flow rate of the pure water, temperature of the pure water, and dropping position of the pure water.

[0066] Furthermore, the device status information may include at least one of dressing status information indicating the status of the dressing unit 204 and cleaning fluid injection status information indicating the status of the cleaning fluid injection unit 205.

[0067] The dressing state information includes at least one of the rotational speed of the dressing unit 204, the rotational torque of the dressing unit 204, the dressing position of the dressing unit 204, and the pressing load of the dressing unit 204.

[0068] The cleaning fluid injection state information includes at least one of the cleaning fluid flow rate, the cleaning fluid temperature, and the cleaning fluid droplet position.

[0069] Furthermore, the polishing process information that constitutes the input data for the learning data 11 may also include consumable status information indicating the condition of consumables used in the polishing process by the substrate polishing apparatus 2.

[0070] The consumable status information includes at least one of the following: the type of polishing pad 200a, the surface roughness of the polishing pad 200a, the polishing time performed by the polishing pad 200a, the type of membrane 2022, the polishing time performed by the membrane 2022, the type of retainer ring 2023, the polishing time performed by the retainer ring 2023, and the type of polishing fluid. The polishing time for the polishing pad 200a, membrane 2022, and retainer ring 2023 is the cumulative time counted when they are used and is reset when they are replaced with new ones. The types of polishing pad 200a, membrane 2022, and retainer ring 2023 are set when they are replaced with new ones. The surface roughness of the polishing pad 200a is reset when it is replaced with a new one and is set to decrease as the usage time increases.

[0071] Furthermore, the consumable status information may include at least one of the following: the type of dressing pad 200b, the surface roughness of the dressing pad 200b, the dressing time during which dressing was performed with the dressing pad 200b, and the type of cleaning fluid. The dressing time is the cumulative time counted when the dressing pad 200b is used and is reset when it is replaced with a new one. The type of dressing pad 200b is set when it is replaced with a new one. The surface roughness of the dressing pad 200b is reset when it is replaced with a new one and is set to decrease as the usage time increases.

[0072] The surface state information that constitutes the output data of the training data 11 is information indicating the surface state of the wafer W that has undergone polishing treatment according to the polishing treatment information. The surface state information may be, for example, the surface state at any point in time included in the polishing treatment period (the time required for polishing treatment per wafer) from the start to the end of the polishing treatment, or it may indicate the surface state at the time the polishing treatment is completed.

[0073] Figure 9 shows an example of film thickness state information and step state information included in surface state information. As shown in Figure 9, the surface state information of wafer W includes film thickness state information indicating the film thickness state of wafer W and step state information indicating the step state of wafer W.

[0074] The film thickness state information is film thickness distribution information that shows the film thickness state for each surface position of the wafer W. The film thickness distribution information is, for example, information that shows the film thickness at each surface position. Each surface position of the wafer W is represented, for example, as a position coordinate in a Cartesian coordinate system or a polar coordinate system, and may be arranged at predetermined intervals, or may be arranged in wiring sections, insulating sections, or their boundaries. The film thickness is represented, for example, as a numerical value normalized to a predetermined range (e.g., 0 to 1), or as a classification value selected from multiple classifications. In this case, the classification value shows the classification result of classifying the film thickness state, and may be a binary classification of normal or abnormal, or a multi-level classification such as -2, -1, 0, +1, +2 with the target film thickness value as the base (0). In the film thickness distribution information shown in Figure 9, a multi-level classification value indicating the film thickness is recorded for each of the surface positions P12 to P54.

[0075] Furthermore, if multiple electronic components are formed on the wafer W, the film thickness state information may be component-unit film thickness distribution information showing the film thickness state for each of the multiple electronic components formed on the wafer W, or it may be component-region film thickness distribution information showing the film thickness state for each surface position within the component region where the electronic components are formed on the wafer W.

[0076] When electronic components are formed in rectangular component regions when a wafer W is divided into a matrix, for example, the component-unit film thickness distribution information identifies each component region in an X-row, Y-column format and shows the film thickness of each component region as, for example, a numerical value or classification value. In the component-unit film thickness distribution information shown in Figure 9, a numerical value indicating the film thickness is recorded for each component region D12 to D54. Furthermore, the component region-internal film thickness distribution information identifies each component region and shows the film thickness at each surface position included in each component region as, for example, a numerical value or classification value. In the component region-internal film thickness distribution information shown in Figure 9, a multi-level classification value indicating the film thickness is recorded for each surface position P11 to P22 in each component region D12 to D54.

[0077] Step state information is step distribution information that shows the step state for each surface position of the wafer W. Step distribution information is, for example, information that shows the presence or absence and degree of step at each surface position. Each surface position of the wafer W is represented, for example, as a position coordinate in a Cartesian coordinate system or polar coordinate system, similar to the film thickness distribution information, and may be distributed at predetermined intervals, or may be placed in wiring sections, insulating sections, or their boundaries. Steps are represented, for example, as a numerical value normalized to a predetermined range (e.g., 0 to 1), or as a classification value selected from multiple classifications. In this case, the classification value shows the classification result of classifying the step state, and may be a binary classification of normal or abnormal, or a multi-level classification such as +1, +2, +3 with the case of no step as the base (0). In the step distribution information shown in Figure 9, a multi-level classification value indicating a step is recorded for each of the surface positions P12 to P54.

[0078] Furthermore, if multiple electronic components are formed on the wafer W, the step state information may be component-unit step distribution information showing the step state for each of the multiple electronic components formed on the wafer W, or it may be component-region step distribution information showing the step state for each surface position of the wafer W within the component region where the electronic components are formed.

[0079] When electronic components are formed in rectangular component regions when a wafer W is divided into a matrix, for example, the component-unit step distribution information identifies each component region in an X-row Y-column format and indicates the step height of each component region as, for example, a numerical value or classification value. In the component-unit step distribution information shown in Figure 9, a numerical value indicating the step height is recorded for each component region D12 to D54. Furthermore, the step distribution information within a component region identifies each component region and indicates the step height at each surface position included in each component region as, for example, a numerical value or classification value. In the component-region step distribution information shown in Figure 9, for each component region D12 to D54, a multi-level classification value indicating the step height is recorded as the step height for each surface position P11 to P22.

[0080] The learning data acquisition unit 400 acquires learning data 11 by referring to the database 30 and, if necessary, accepting user input operations from the user terminal device 6. For example, the learning data acquisition unit 400 acquires polishing process information and surface condition information when polishing is performed by referring to the device history table 301, the consumables history table 302, and the surface condition history table 303 of the database 30. At that time, the learning data acquisition unit 400 acquires polishing process information and surface condition information by referring to the device condition information, consumables condition information, and surface condition information associated with the wafer ID that identifies the wafer W that is the target of measurement by the external measuring instrument, based on the wafer ID.

[0081] The polishing process information may be obtained as time-series data for the entire polishing process period, as time-series data for a specific period that is part of the polishing process, or as time-series data for a specific point in time.

[0082] Furthermore, in this embodiment, the case in which the surface condition information includes film thickness condition information and step condition information, as shown in Figure 8, is described, but it may also include at least one of the film thickness condition information and step condition information. Also, if the polishing process information is acquired, for example, as time-series data for the entire polishing process period or as time-series data for a target period which is a part of the polishing process period, the surface condition information may also be time-series data for the entire polishing process period or as time-series data for the target period, or as time-series data at the end of the polishing process or as time-series data at a target time. If the polishing process information is acquired, for example, as time-series data at a specific target time, the surface condition information may also be acquired as time-series data at that specific target time.

[0083] Furthermore, as described above, if the definition of either the surface condition information or the polishing process information is changed, the data structure of the output data in the learning model 10 and the learning data 11 should be changed as appropriate.

[0084] The learning model 10 employs, for example, a neural network structure and comprises an input layer 100, a hidden layer 101, and an output layer 102. Synapses (not shown) connect each neuron between each layer, and each synapse is associated with a weight. The weight parameter set, consisting of the weights of each synapse, is adjusted by machine learning.

[0085] The input layer 100 has a number of neurons corresponding to the polishing process information as input data, and each value of the polishing process information is input to each neuron. The output layer 102 has a number of neurons corresponding to the surface state information as output data, and the inference result of the surface state information in relation to the polishing process information (inference result) is output as output data. If the learning model 10 is composed of a regression model, the surface state information is output as numerical values ​​normalized to a predetermined range (e.g., 0 to 1). If the learning model 10 is composed of a classification model, the surface state information is output as a score (accuracy) for each classification value (class), as numerical values ​​normalized to a predetermined range (e.g., 0 to 1).

[0086] (Machine Learning Method) Figure 10 is a flowchart showing an example of a machine learning method using the machine learning device 4.

[0087] First, in step S100, the training data acquisition unit 400 acquires a desired number of training data 11 from the database 30 as preparation for starting machine learning, and stores the acquired training data 11 in the training data storage unit 42. The number of training data 11 to be prepared here should be set considering the inference accuracy required for the final learning model 10.

[0088] Next, in step S110, the machine learning unit 401 prepares a pre-training model 10 in order to start machine learning. The pre-training model 10 prepared here consists of the neural network model illustrated in Figure 8, and the weights of each synapse are set to their initial values.

[0089] Next, in step S120, the machine learning unit 401 randomly selects, for example, one set of training data 11 from the multiple sets of training data 11 stored in the training data storage unit 42.

[0090] Next, in step S130, the machine learning unit 401 inputs the polishing process information (input data) contained in a set of training data 11 to the input layer 100 of the prepared pre-training (or training) learning model 10. As a result, surface state information (output data) is output from the output layer 102 of the learning model 10 as an inference result, but this output data is generated by the pre-training (or training) learning model 10. Therefore, in the pre-training (or training) state, the output data output as an inference result shows information different from the surface state information (ground truth label) contained in the training data 11.

[0091] Next, in step S140, the machine learning unit 401 compares the surface state information (ground truth labels) contained in the set of training data 11 acquired in step S120 with the surface state information (output data) output from the output layer 102 as an inference result in step S130, and performs machine learning by adjusting the weight of each synapse (backpropagation). In this way, the machine learning unit 401 allows the learning model 10 to learn the correlation between polishing process information and surface state information.

[0092] Next, in step S150, the machine learning unit 401 determines whether predetermined learning termination conditions have been met, for example, based on the evaluation value of the error function, which is based on the surface state information (correct labels) contained in the training data 11 and the surface state information output as an inference result (output data), or on the remaining number of untrained training data 11 stored in the training data storage unit 42.

[0093] In step S150, if the machine learning unit 401 determines that the learning termination condition has not been met and that machine learning should continue (No in step S150), it returns to step S120 and performs steps S120 to S140 multiple times on the learning model 10 that is currently being learned, using the untrained training data 11. On the other hand, in step S150, if the machine learning unit 401 determines that the learning termination condition has been met and that machine learning should be terminated (Yes in step S150), it proceeds to step S160.

[0094] Then, in step S160, the machine learning unit 401 stores the trained model 10 (set weight parameter group) generated by adjusting the weights associated with each synapse in the trained model storage unit 43, and the series of machine learning methods shown in Figure 10 is completed. In the machine learning method, step S100 corresponds to the training data storage step, steps S110 to S150 are the machine learning steps, and step S160 is the trained model storage step.

[0095] As described above, the machine learning apparatus 4 and machine learning method according to this embodiment provide a learning model 10 that can predict (infer) surface state information indicating the surface state of the wafer W from polishing process information including apparatus state information and consumable state information.

[0096] (Information Processing Device 5) Figure 11 is a block diagram showing an example of the information processing device 5. Figure 12 is a functional diagram showing an example of the information processing device 5. The information processing device 5 comprises a control unit 50, a communication unit 51, and a learned model storage unit 52.

[0097] The control unit 50 functions as an information acquisition unit 500, an information generation unit 501, and an output processing unit 502. The communication unit 51 is connected to external devices (for example, a substrate polishing device 2, a database device 3, a machine learning device 4, and a user terminal device 6, etc.) via the network 7 and functions as a communication interface for sending and receiving various types of data.

[0098] The information acquisition unit 500 is connected to an external device via the communication unit 51 and the network 7, and acquires polishing process information, including device status information and consumable status information.

[0099] For example, when performing a "post-processing prediction" of surface condition information for a wafer W after polishing has already been performed, the information acquisition unit 500 refers to the equipment history table 301 and the consumables history table 302 of the database 30 to acquire the equipment status information and consumables status information at the time the wafer W was polished as polishing process information. When performing a "real-time prediction" of surface condition information for a wafer W while polishing is in progress, the information acquisition unit 500 receives reports R regarding equipment status information and consumables status information from the substrate polishing apparatus 2 performing the polishing process as needed, and acquires the equipment status information and consumables status information at the time the wafer W is being polished as polishing process information. When performing a "pre-prediction process" of surface condition information for a wafer W before polishing, the information acquisition unit 500 receives substrate recipe information 216 from the substrate polishing apparatus 2 which is scheduled to perform the polishing process, and simulates the apparatus state information and consumables state information when the polishing unit 20 operates according to the substrate recipe information 216, thereby acquiring the apparatus state information and consumables state information when polishing is performed on the wafer W as polishing process information.

[0100] As described above, the information generation unit 501 inputs the polishing process information (including equipment status information and consumable status information in this embodiment) acquired by the information acquisition unit 500 as input data to the learning model 10, thereby generating surface status information (including film thickness status information and step height status information in this embodiment) that indicates the surface state of the wafer W when polishing is performed using the substrate polishing apparatus 2 in the state indicated by the equipment status information included in the polishing process information and the consumables in the state indicated by the consumable status information included in the polishing process information.

[0101] Furthermore, if the polishing process information acquired by the information acquisition unit 500 includes only equipment status information, the information generation unit 501 can input the polishing process information as input data to the learning model 10 to generate surface status information indicating the surface state of the wafer W when polishing is performed by the substrate polishing apparatus 2 in the state indicated by the equipment status information included in the polishing process information.

[0102] The trained model storage unit 52 is a database that stores trained models 10 used by the information generation unit 501. The number of trained models 10 stored in the trained model storage unit 52 is not limited to one. For example, multiple trained models 10 with different conditions may be stored and used selectively or in parallel, such as machine learning methods, wafer types (size, thickness, film type, etc.), differences in the configuration of the polishing unit 20, types of consumables (polishing pads 200a and dressing pads 200b, etc.), types of polishing fluids, types of cleaning fluids, types of data included in polishing process information, and types of data included in surface condition information. The trained model storage unit 52 may also be replaced by a storage unit of an external computer (for example, a server-type computer or a cloud-type computer), in which case the information generation unit 501 only needs to access the external computer.

[0103] The output processing unit 502 performs output processing to output the surface state information generated by the information generation unit 501. For example, the output processing unit 502 may transmit the surface state information to the user terminal device 6 so that a display screen based on the surface state information is displayed on the user terminal device 6, or it may transmit the surface state information to the database device 3 so that the surface state information is registered in the surface state history table 303 of the database 30.

[0104] (Information Processing Method) Figure 13 is a flowchart showing an example of an information processing method by the information processing device 5. Below, we will describe an example of operation when a user operates the user terminal device 6 to perform "post-prediction processing" of surface state information for a specific wafer W.

[0105] First, in step S200, when the user performs an input operation to input a wafer ID that identifies the wafer W to be predicted to the user terminal device 6, the user terminal device 6 transmits that wafer ID to the information processing device 5.

[0106] Next, in step S210, the information acquisition unit 500 of the information processing device 5 receives the wafer ID transmitted in step S200. In step S211, the information acquisition unit 500 uses the wafer ID received in step S210 to refer to the device history table 301 and the consumables history table 302 of the database 30 to acquire polishing process information (in this embodiment, device status information and consumables status information) when polishing is performed on the wafer W identified by that wafer ID.

[0107] Next, in step S220, the information generation unit 501 inputs the polishing process information acquired in step S211 as input data to the learning model 10, thereby generating surface state information for the polishing process information as output data and predicting the surface state of the wafer W.

[0108] Next, in step S230, the output processing unit 502 transmits the surface condition information generated in step S220 to the user terminal device 6 as an output process. In addition to the user terminal device 6, the destination for the surface condition information may also be the database device 3.

[0109] Next, in step S240, the user terminal device 6 receives the surface state information transmitted in step S230 as a response to the transmission process in step S200, and then displays a display screen based on that surface state information, thereby ending the series of information processing methods shown in Figure 13. The user can then check the surface state of the wafer W via the display screen. In the series of information processing methods shown in Figure 13, steps S210 and S211 correspond to the information acquisition process, step S220 to the information generation process, and step S230 to the output processing process.

[0110] As described above, according to the information processing apparatus 5 and information processing method of this embodiment, polishing processing information including apparatus status information and consumable status information is input to the learning model 10, and surface state information (in this embodiment, film thickness status information and step status information) is generated for said polishing processing information. Therefore, the surface state of the wafer W during or after the polishing process can be appropriately predicted.

[0111] (Other Embodiments) The present invention is not limited to the embodiments described above, and can be implemented with various modifications without departing from the spirit of the invention. All of these modifications are included in the technical concept of the present invention.

[0112] In the above embodiment, the case where the substrate to be polished by the substrate polishing apparatus 2 is a wafer W was described. However, the substrate to be polished may be a substrate other than a wafer W, for example, a CCL substrate (Copper Clad Laminate) substrate, a PCB (Printed Circuit Board) substrate, a photomask substrate, a glass substrate, a silicon substrate, a display panel, etc. Also, the shape of the substrate is not limited to a circle, but may be any shape, for example, a polygon such as a square. In that case, the configuration of the substrate polishing apparatus 2 may be appropriately changed according to the type and shape of the substrate to be polished. For example, if the shape of the substrate is square, the shape of the polishing head 202a for holding the substrate may be changed to a square.

[0113] In the above embodiment, the database device 3, the machine learning device 4, and the information processing device 5 were described as being composed of separate devices, but these three devices may be composed of a single device, or any two of these three devices may be composed of a single device. Furthermore, at least one of the machine learning device 4 and the information processing device 5 may be incorporated into the control unit 21 or user terminal device 6 of the substrate polishing device 2.

[0114] In the above embodiment, a case in which a neural network is used as the learning model 10 that realizes machine learning by the machine learning unit 401 was described, but other machine learning models may also be used. Examples of other machine learning models include tree-type models such as decision trees and regression trees, ensemble learning such as bagging and boosting, neural network types such as recurrent neural networks, convolutional neural networks, and LSTM (including deep learning), hierarchical clustering, non-hierarchical clustering, clustering types such as k-nearest neighbors and k-means, multivariate analysis such as principal component analysis, factor analysis, and logistic regression, and support vector machines.

[0115] (Machine Learning Programs and Information Processing Programs) The present invention can also be provided in the form of a program (machine learning program) that causes the computer 900 to function as each part of the machine learning apparatus 4, or a program (machine learning program) that causes the computer 900 to execute each step of the machine learning method. Furthermore, the present invention can also be provided in the form of a program (information processing program) that causes the computer 900 to function as each part of the substrate polishing apparatus 2, or a program (information processing program) that causes the computer 900 to execute each step of the information processing method according to the above embodiment.

[0116] (Inference device, inference method, and inference program) The present invention can be provided not only in the form of the information processing device 5 (information processing method or information processing program) according to the above embodiment, but also in the form of an inference device (inference method or inference program) used to infer surface state information. In that case, the inference device (inference method or inference program) may include a memory and a processor, the processor of which may execute a series of processes. The series of processes includes an information acquisition process (information acquisition step) for acquiring polishing process information, and an inference process (inference step) for inferring surface state information indicating the surface state of the wafer W when polishing is performed by the substrate polishing device 2 in the state indicated by the device state information included in the polishing process information, after acquiring the polishing process information in the information acquisition process.

[0117] By providing the inference device (inference method or inference program) in the form of an inference device, it becomes easier to apply to various devices compared to implementing the information processing device 5. It will be obvious to those skilled in the art that when the inference device (inference method or inference program) infers surface state information, it may apply the inference method performed by the information generation unit 501 using the trained learning model 10 generated by the machine learning device 4 and machine learning method according to the above embodiment.

[0118] 1...Substrate polishing system, 2...Substrate polishing device, 3...Database device, 4...Machine learning device, 5...Information processing device, 6...User terminal device, 10...Learning model, 11...Learning data, 20...Polishing unit, 21...Control unit, 30...Database, 40...Control unit, 41...Communication unit, 42...Learning data storage unit, 43...Learned model storage unit, 50...Control unit, 51...Communication unit, 52...Learned model storage unit, 200a...Polishing pad, 200b...Dressing pad, 201...Polishing table unit, 202...Polishing head unit, 203...Polishing fluid supply unit, 204...Dressing unit, 205...Cleaning fluid injection unit, 206...Environmental measurement unit, 400...Learning data acquisition unit, 401...Machine learning unit, 500...Information acquisition unit, 501...Information generation unit, 502...Output processing unit

Claims

1. An information processing device comprising: an information acquisition unit that acquires polishing process information including device status information indicating the state of a substrate polishing device that performs a polishing process in which a substrate is polished with a polishing pad; and an information generation unit that inputs the polishing process information acquired by the information acquisition unit into a learning model to generate surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing device in the state indicated by the device status information included in the polishing process information, wherein the learning model is a trained model that has learned the correlation between the polishing process information and the surface state information by machine learning.

2. The information processing apparatus according to claim 1, wherein the apparatus state information included in the polishing process information includes at least one of: polishing table state information indicating the state of a polishing table that rotatably supports the polishing pad; polishing head state information indicating the state of a polishing head that presses the substrate against the polishing pad; and polishing fluid supply state information indicating the state of a polishing fluid supply unit that supplies polishing fluid to the polishing pad.

3. The polishing table state information includes at least one of the rotational speed of the polishing table, the rotational torque of the polishing table, and the surface temperature of the polishing pad; the polishing head comprises a top ring body moved by a rotational movement mechanism, an up-and-down movement mechanism, and an oscillating movement mechanism; a membrane housed in the top ring body that presses the substrate against the polishing pad; and a retainer ring arranged on the outer circumference of the membrane that presses the polishing pad; the polishing head state information includes at least one of the rotational speed of the top ring body, the rotational torque of the top ring body, the polishing position of the top ring body, the oscillating speed of the top ring body, the oscillating torque of the top ring body, the height of the top ring body, the lifting speed of the top ring body, the lifting torque of the top ring body, the pressing force of the membrane, and the pressing force of the retainer ring; and the polishing fluid supply state information includes at least one of the flow rate of the polishing fluid, the temperature of the polishing fluid, and the dropping position of the polishing fluid.

4. The polishing process information further includes consumable status information indicating the state of consumables used in the polishing process by the substrate polishing apparatus, the consumable status information included in the polishing process information includes at least one of the type of polishing pad, the surface roughness of the polishing pad, the polishing time during which the polishing process was performed using the polishing pad, the type of membrane, the polishing time during which the polishing process was performed using the membrane, the type of retainer ring, the polishing time during which the polishing process was performed using the retainer ring, and the type of polishing fluid, the information generation unit inputs the polishing process information acquired by the information acquisition unit into the learning model to generate surface status information indicating the surface state of the substrate when the polishing process was performed using the substrate polishing apparatus in the state indicated by the apparatus status information included in the polishing process information and the consumables in the state indicated by the consumable status information included in the polishing process information, the information processing apparatus according to claim 3.

5. The information processing apparatus according to claim 2, wherein the apparatus state information included in the polishing process information further includes at least one of dressing state information indicating the state of a dressing section that dresses the polishing pad with a dressing pad, and cleaning fluid injection state information indicating the state of a cleaning fluid injection section that injects cleaning fluid onto the polishing pad.

6. The information processing apparatus according to claim 5, wherein the dressing state information includes at least one of the rotational speed of the dressing part, the rotational torque of the dressing part, the dressing position of the dressing part, and the pressing load of the dressing part, and the cleaning fluid injection state information includes at least one of the flow rate of the cleaning fluid, the temperature of the cleaning fluid, and the dropping position of the cleaning fluid.

7. The polishing process information further includes consumable status information indicating the state of consumables used in the polishing process by the substrate polishing apparatus, wherein the consumable status information included in the polishing process information includes at least one of the type of dressing pad, the surface roughness of the dressing pad, the dressing time during which the dressing was performed with the dressing pad, and the type of cleaning fluid, and the information generation unit inputs the polishing process information acquired by the information acquisition unit into the learning model to generate surface status information indicating the surface state of the substrate when the polishing process was performed using the substrate polishing apparatus in the state indicated by the apparatus status information included in the polishing process information and the consumables in the state indicated by the consumable status information included in the polishing process information, the information processing apparatus according to claim 6.

8. The information processing apparatus according to claim 1, wherein the surface state information includes film thickness state information indicating the film thickness state of the substrate as the surface state.

9. The information processing apparatus according to claim 8, wherein the film thickness state information included in the surface state information is film thickness distribution information indicating the film thickness state for each surface position of the substrate.

10. The information processing apparatus according to claim 8, wherein the film thickness state information included in the surface state information is component-unit film thickness distribution information indicating the film thickness state for each of the plurality of electronic components formed on the substrate.

11. The information processing apparatus according to claim 8, wherein the film thickness state information included in the surface state information is component region film thickness distribution information indicating the film thickness state for each surface position within the component region where the electronic components are formed, for a plurality of electronic components formed on the substrate.

12. The information processing apparatus according to claim 1, wherein the surface state information includes step state information indicating the step state of the substrate as the surface state.

13. The information processing apparatus according to claim 12, wherein the step state information included in the surface state information is step distribution information indicating the step state for each surface position of the substrate.

14. The information processing apparatus according to claim 12, wherein the step state information included in the surface state information is component-unit step distribution information indicating the step state for each of the plurality of electronic components formed on the substrate.

15. The information processing apparatus according to claim 12, wherein the step state information included in the surface state information is component region step distribution information indicating the step state for each surface position of the substrate within the component region where the electronic components are formed, for a plurality of electronic components formed on the substrate.

16. An inference device comprising memory and a processor, wherein the processor performs: an information acquisition process for acquiring polishing process information including device state information indicating the state of a substrate polishing device that performs polishing processing for polishing a substrate with a polishing pad; and an inference process for inferring surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing device in the state indicated by the device state information included in the polishing process information, after acquiring the polishing process information in the information acquisition process.

17. A machine learning device comprising: a learning data storage unit that stores multiple sets of learning data, each set consisting of polishing process information including device status information indicating the state of a substrate polishing apparatus that performs a polishing process of polishing a substrate with a polishing pad, and surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing apparatus in the state indicated by the device status information included in the polishing process information; a machine learning unit that inputs multiple sets of the learning data into a learning model to train the learning model on the correlation between the polishing process information and the surface state information; and a trained model storage unit that stores the learning model on which the machine learning unit has trained on the correlation.

18. An information processing method comprising: an information acquisition step of acquiring polishing process information including device status information indicating the state of a substrate polishing apparatus that performs a polishing process of polishing a substrate with a polishing pad; and an information generation step of inputting the polishing process information acquired in the information acquisition step into a learning model to generate surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing apparatus in the state indicated by the device status information included in the polishing process information, wherein the learning model is a trained model that has learned the correlation between the polishing process information and the surface state information by machine learning.

19. An inference method performed by an inference device comprising memory and a processor, wherein the processor performs an information acquisition step of acquiring polishing process information including device state information indicating the state of a substrate polishing device that performs polishing processing for polishing a substrate with a polishing pad, and an inference step of, upon acquiring the polishing process information in the information acquisition step, inferring surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing device in the state indicated by the device state information included in the polishing process information.

20. A machine learning method comprising: a learning data storage step of storing multiple sets of learning data in a learning data storage unit, each set consisting of polishing process information including device status information indicating the state of a substrate polishing apparatus that performs a polishing process of polishing a substrate with a polishing pad, and surface state information indicating the surface state of the substrate when the polishing process is performed by the substrate polishing apparatus in the state indicated by the device status information included in the polishing process information; a machine learning step of inputting multiple sets of the learning data into a learning model to train the learning model on the correlation between the polishing process information and the surface state information; and a trained model storage step of storing the learning model on a trained model storage unit, in which the learning model has learned the correlation by the machine learning step.

Images