Control system and control method for semiconductor manufacturing equipment

Abstract

The present disclosure provides a control system for semiconductor manufacturing equipment and a control method thereof. The control system includes a sensor unit configured to record a set of data indicating the operating status of the semiconductor manufacturing equipment; a sensor interface configured to receive the set of data and generate at least one input signal for a data server; and a control unit. The control unit includes a diagnostic subsystem that executes an operating status monitoring program to determine whether the semiconductor manufacturing equipment has a fault and generates a data signal.

Description

[0001] Cross-references

[0002] This disclosure claims priority to and the benefit of U.S. regular application No. 17 / 064,365, filed on October 6, 2020, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] The present disclosure relates to a control system for manufacturing equipment and a control method thereof, and more particularly to a control system for semiconductor manufacturing equipment and a control method thereof. Background Art

[0004] The increasing density of integrated circuits (ICs) is driving the rapid growth of the semiconductor industry. Among the numerous manufacturing equipment, semiconductor manufacturing equipment, such as transfer robots, are an integral part of the manufacturing process. Transfer robot failures and path deviations are often discovered only after numerous wafers have been contaminated and scrapped. Manual inspection and monitoring of transfer robot operation often results in wasted process time and product defects due to human error. Therefore, the control systems and methods used to control and monitor the transfer robot's status must be able to effectively detect faulty equipment and provide early warnings of various equipment parameters.

[0005] The above description of “prior art” only provides background technology, does not admit that the above description of “prior art” discloses the subject matter of the present disclosure, does not constitute the prior art of the present disclosure, and any description of the above “prior art” should not be regarded as any part of the present disclosure. Summary of the Invention

[0006] One embodiment of the present disclosure provides a control system for semiconductor manufacturing equipment, comprising: a sensor unit configured to obtain a set of data related to an operating status of the semiconductor manufacturing equipment; a sensor interface configured to receive the data set and generate at least one input signal for a data server; and a control unit. The control unit includes a front-end subsystem, a computing subsystem, and an information and feedback subsystem. The front-end subsystem receives the at least one input signal from the data server and executes a front-end program to generate a first data signal. The computing subsystem includes a diagnostic subsystem configuration and a decision subsystem. The diagnostic subsystem receives the first data signal from the front-end subsystem and executes an operating status monitoring program to determine whether a fault has occurred in the semiconductor manufacturing equipment and generates a second data signal. The decision subsystem executes an artificial intelligence (AI) analysis program based on the second data signal of the diagnostic subsystem to generate an output signal. The information and feedback subsystem generates an alarm signal and a feedback signal based on the output signal and sends the alarm signal to a user of the semiconductor manufacturing equipment.

[0007] In some embodiments, the front-end process executed by the front-end subsystem includes extracting, converting and / or loading the input signal by the data server.

[0008] In some embodiments, the AI ​​analysis program executed by the decision-making subsystem monitors and determines one or more trends in the second data signal of the diagnostic subsystem.

[0009] In some embodiments, the AI ​​analysis program executed by the decision subsystem also utilizes a statistical model to predict to obtain an equipment life prediction score for the semiconductor manufacturing equipment, where the equipment life prediction score corresponds to the set of data related to the operating state in the first data signal.

[0010] In some embodiments, the information and feedback subsystem sends a feedback signal according to the output signal for an automatic feedback process of the semiconductor manufacturing equipment.

[0011] In some embodiments, the warning signal includes a device operation score warning, a device life prediction warning, and a faulty component warning generated according to the output signal, wherein the device life prediction warning includes a predicted life value of the semiconductor manufacturing equipment.

[0012] In some embodiments, the semiconductor fabrication equipment includes one or more electrostatic chucks.

[0013] Another embodiment of the present disclosure provides a control system for semiconductor manufacturing equipment, comprising: a sensor unit configured to record a set of data related to an operating state of the semiconductor manufacturing equipment; a sensor interface configured to receive the set of data and generate at least one input signal for a data server; one or more processors; and one or more computer-readable non-transitory storage media. The control system is configured to be coupled to the one or more processors. The one or more computer-readable non-transitory storage media are configured to be coupled to the one or more processors and include instructions that are operable when executed by the one or more processors to cause the control system to: receive the at least one input signal from the data server and execute a front-end program to generate a data signal; execute an operating state monitoring program to determine whether a fault occurs in the semiconductor manufacturing equipment and generate a second data signal; execute an artificial intelligence (AI) analysis program to generate an output signal based on the second data signal; and generate an alarm signal and a feedback signal based on the output signal and send the alarm signal to a user of the semiconductor manufacturing equipment.

[0014] In some embodiments, the front-end process includes extracting, converting and / or loading the input signal by the data server.

[0015] In some embodiments, the one or more computer-readable non-transitory storage media further include instructions operable when executed by the one or more processors to cause the control system to perform the AI ​​analysis program by monitoring and determining one or more trends in the second data signal.

[0016] In some embodiments, the one or more computer-readable non-transitory storage media also include instructions that are operable when executed by one or more processors to enable the control system to further execute the AI ​​analysis program by utilizing a statistical model to predict and obtain an equipment life prediction score for the semiconductor manufacturing equipment, wherein the equipment life prediction score corresponds to the set of data related to the operating state in the first data signal.

[0017] In some embodiments, the one or more computer-readable non-transitory storage media further include instructions operable when executed by one or more processors to cause the control system to send the feedback signal for an automatic feedback procedure of the semiconductor manufacturing equipment based on the output signal.

[0018] In some embodiments, the warning signal includes a device operation score warning, a device life prediction warning, and a faulty component warning generated according to the output signal, wherein the device life prediction warning includes a predicted life value of the semiconductor manufacturing equipment.

[0019] In some embodiments, the semiconductor fabrication equipment includes one or more electrostatic chucks.

[0020] Another embodiment of the present disclosure provides a control method for semiconductor manufacturing equipment, including: a sensor unit recording a set of data related to an operating status of the semiconductor manufacturing equipment; a sensor interface receiving the set of data and generating at least one input signal for a data server; a front-end subsystem receiving the at least one input signal of the data server and performing a front-end program to generate a first data signal; a diagnostic subsystem receiving the first data signal of the front-end subsystem and executing an operating status monitoring program to determine whether a fault occurs in the semiconductor manufacturing equipment and generating a second data signal; a decision subsystem executing an artificial intelligence (AI) analysis program based on the second data signal of the diagnostic subsystem to generate an output signal; and an information and feedback subsystem generating a warning signal and a feedback signal based on the output signal and sending the warning signal to a user of the semiconductor manufacturing equipment.

[0021] In some embodiments, the front-end program executed by the front-end subsystem further includes extracting, converting and / or loading the input signal by the data server.

[0022] In some embodiments, the AI ​​analysis program executed by the computing subsystem further includes monitoring and determining one or more trends in the second data signal of the diagnostic subsystem.

[0023] In some embodiments, the AI ​​analysis program executed by the computing subsystem also includes using a statistical model to predict to obtain an equipment life prediction score for the semiconductor manufacturing equipment, and the equipment life prediction score corresponds to the set of data related to the operating state in the first data signal.

[0024] In some embodiments, the control method further includes sending, by the information and feedback subsystem according to the output signal, the feedback signal for an automatic feedback procedure of the semiconductor manufacturing equipment.

[0025] In some embodiments, the warning signal includes an equipment operation score warning, an equipment life prediction warning, and a faulty component warning generated according to the output signal, wherein the equipment life prediction warning includes a predicted life value of the semiconductor manufacturing equipment, and wherein the semiconductor manufacturing equipment includes one or more transfer robots.

[0026] Therefore, the control system and control method for controlling semiconductor manufacturing equipment provide an automated AI subsystem and its processing, capable of monitoring and determining one or more trends in the data signal from the diagnostic subsystem. Due to the AI ​​analysis program executed by the computing subsystem, the operator of the semiconductor manufacturing equipment can obtain an equipment life prediction score for the semiconductor manufacturing equipment, the equipment life prediction score corresponding to the set of data regarding the operating state in the first data signal. Furthermore, due to the automatic feedback program of the control system and the control method, human error, process time, and the number of product defects can be reduced.

[0027] The above has been a fairly broad overview of the technical features and advantages of the present disclosure, so that the detailed description of the present disclosure below can be better understood. Other technical features and advantages that constitute the subject matter of the claims of the present disclosure will be described below. It should be understood by those skilled in the art of the present disclosure that the concepts and specific embodiments disclosed below can be used to modify or design other structures or processes to achieve the same purpose as the present disclosure. It should also be understood by those skilled in the art of the present disclosure that such equivalent constructions cannot depart from the concept and scope of the present disclosure as defined by the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] A more complete understanding of the disclosure of the present application may be obtained by referring to the embodiments and claims in conjunction with the accompanying drawings, in which like reference numerals refer to like elements.

[0029] Figure 1 A functional block diagram of a semiconductor manufacturing facility illustrating some embodiments of the present disclosure.

[0030] Figure 2 A functional block diagram of a control platform of the semiconductor manufacturing facility illustrating some embodiments of the present disclosure.

[0031] Figure 3 Schematic diagram of an electrostatic chuck illustrating some embodiments of the present disclosure.

[0032] Figure 4A Examples of some embodiments of the present disclosure Figure 3 A surface mapping of the electrostatic chuck in FIG.

[0033] Figure 4B Examples of some embodiments of the present disclosure Figure 3 Another surface of the electrostatic chuck is mapped.

[0034] Figure 5 A functional block diagram of a control system for a semiconductor manufacturing equipment illustrating some embodiments of the present disclosure.

[0035] Figure 6 A functional block diagram illustrating a sensor interface according to some embodiments of the present disclosure.

[0036] Figure 7 A functional block diagram illustrating a control unit according to some embodiments of the present disclosure.

[0037] Figure 8 A functional block diagram of a front-end subsystem illustrating some embodiments of the present disclosure.

[0038] Figure 9 A functional block diagram of a computing subsystem illustrating some embodiments of the present disclosure.

[0039] Figure 10 A state diagram illustrating an operational status monitoring program executed by a diagnostic subsystem according to some embodiments of the present disclosure.

[0040] Figure 11 A partial schematic diagram illustrating a classification process of an AI analysis program executed by a decision subsystem according to some embodiments of the present disclosure.

[0041] Figure 12 A functional block diagram illustrating the control system and a user for controlling the semiconductor manufacturing equipment according to some embodiments of the present disclosure.

[0042] Figure 13 A functional block diagram illustrating the control system and a user for controlling the semiconductor manufacturing equipment according to some embodiments of the present disclosure.

[0043] Figure 14A A diagram illustrating the relationship between the warning signal AM and its components according to some embodiments of the present disclosure.

[0044] Figure 14B A diagram illustrating the relationship between the feedback signal FB and its components according to some embodiments of the present disclosure.

[0045] Figure 15 A functional block diagram of a computer system illustrating some embodiments of the present disclosure.

[0046] Figure 16 A flowchart illustrating a control method of the semiconductor manufacturing equipment according to some embodiments of the present disclosure is provided.

[0047] Figure 17 A flowchart illustrating a step of the control method of the semiconductor manufacturing equipment according to some embodiments of the present disclosure.

[0048] Figure 18 A flowchart illustrating a step of the control method of the semiconductor manufacturing equipment according to some embodiments of the present disclosure.

[0049] Figure 19 A flowchart illustrating a step of the control method of the semiconductor manufacturing equipment according to some embodiments of the present disclosure.

[0050] Description of reference numerals:

[0051] 1: Manufacturing facilities

[0052] 2: Equipment

[0053] 3: Equipment

[0054] 4: Equipment

[0055] 5: Equipment

[0056] 6: Equipment

[0057] 7: Process sequence

[0058] 11: System

[0059] 12: System

[0060] 13: System

[0061] 14: System

[0062] 15: System

[0063] 20: Control Platform

[0064] 25: Extract, Transform, and Load (ETL) Module

[0065] 30: Electrostatic chuck

[0066] 30F: Fault

[0067] 30MF: Fault

[0068] 32: Wafer

[0069] 33: pit

[0070] 34: Electrode

[0071] 35: Heating unit

[0072] 36: Base

[0073] 40: Signal converter

[0074] 41: Filter

[0075] 50: Front-end subsystem

[0076] 51: Computing subsystem

[0077] 52: Information and Feedback Subsystem

[0078] 60: Extraction area

[0079] 61: Conversion Area

[0080] 62: Loading area

[0081] 70: Diagnostic subsystem

[0082] 71: Decision-making subsystem

[0083] 80: Computer systems

[0084] 100: System

[0085] 101: Sensor unit

[0086] 102: Sensor interface

[0087] 103: Data server

[0088] 104: First transceiver

[0089] 105: Second transceiver

[0090] 106: Control unit

[0091] 107: Network link

[0092] 108: Network link

[0093] 109: User

[0094] 110: Client system

[0095] 111: Graphical User Interface (GUI)

[0096] 112: Automatic feedback program

[0097] 113: Standalone controller

[0098] 500: Processor

[0099] 502: Network Interface

[0100] 504: Input / output devices

[0101] 506: Storage device

[0102] 508: Memory

[0103] 510: User space

[0104] 512: Core

[0105] 514: Bus

[0106] 600: Control Methods

[0107] AI: Artificial Intelligence

[0108] AM: Warning signs

[0109] D1: first data signal

[0110] D2: Second data signal

[0111] DA: Data

[0112] DS: data signal

[0113] FB: Feedback signal

[0114] IMG: Image

[0115] IN: input signal

[0116] LS: Probability Score

[0117] OS: output signal

[0118] S601: Step

[0119] S602: Step

[0120] S603: Step

[0121] S6031: Steps

[0122] S6032: Steps

[0123] S604: Step

[0124] S6041: Steps

[0125] S6042: Steps

[0126] S6043: Steps

[0127] S605: Step

[0128] S6051: Steps

[0129] S6052: Steps

[0130] SEC: Stop device command

[0131] SMP: Statistical Model Prediction

[0132] T1: critical value

[0133] T2: critical value

[0134] THR: critical level

[0135] TR: Trend

[0136] TR_MCH: Matching trend group

[0137] TR_NMCH: Non-matching trend group

[0138] WSC: Wafer Scrap Order DETAILED DESCRIPTION

[0139] The following description of the present disclosure is accompanied by the accompanying drawings, which are incorporated in and constitute a part of the specification, and illustrates an embodiment of the present disclosure, but the present disclosure is not limited to the embodiment. In addition, the following embodiments can be appropriately integrated to complete another embodiment.

[0140] References to "one embodiment," "an embodiment," "illustrative embodiment," "another embodiment," and "another embodiment" indicate that the embodiments described herein may include particular features, structures, or characteristics, but not every embodiment is required to include such features, structures, or characteristics. Furthermore, repeated use of the phrase "in an embodiment" does not necessarily refer to the same embodiment, but may.

[0141] In order to make the present disclosure fully understandable, the following description provides detailed steps and structures. Obviously, the practice of the present disclosure is not limited to the specific details known to those skilled in the art. In addition, known structures and steps are not described in detail to avoid unnecessarily limiting the present disclosure. Preferred embodiments of the present disclosure are described below. However, in addition to the detailed description, the present disclosure can also be widely implemented in other embodiments. The scope of the present disclosure is not limited to the content of the detailed description, but is defined by the claims.

[0142] It should be understood that the following disclosure provides many different embodiments or examples for implementing different features of the present invention. Specific embodiments or examples of components and arrangements are described below to simplify the present disclosure. Of course, these are merely examples and are not intended to be limiting. For example, the dimensions of the components are not limited to the disclosed ranges or values, but may depend on the process conditions and / or the desired properties of the device. In addition, the following description of a first feature being formed "on" or "on" a second feature may include embodiments in which the first and second features are formed to be in direct contact, and may also include embodiments in which additional features may be formed between the first and second features, thereby preventing the first and second features from being in direct contact. For the sake of brevity and clarity, various features may be drawn at different scales. In the accompanying drawings, some layers / features may be omitted for simplicity.

[0143] Furthermore, for ease of description, spatially relative terms such as "beneath," "below," "lower," "above," and "upper" may be used herein to describe the relationship of one element or feature to another element or feature as illustrated in the figures. These spatially relative terms are intended to encompass different orientations of the elements in use or operation in addition to the orientation depicted in the figures. The device may be otherwise oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein should be interpreted accordingly.

[0144] To describe thin film or deposition processes, the term "deposition" will generally be used herein for consistency. For thin film removal, the term "etch" will be used, and for clean removal processes, the term "clean" will be used. For clarity or convenience, the figures may use other applicable references.

[0145] Figure 1 A functional block diagram of a manufacturing facility illustrating some embodiments of the present disclosure. Figure 1 As shown, a manufacturing facility (e.g., a semiconductor manufacturing facility) 1 may include a plurality of semiconductor manufacturing equipment, such as a deposition apparatus 2, a chemical mechanical polishing (CMP) apparatus 3, a lithography (photolithography) apparatus 4, an etching apparatus 5, and a cleaning apparatus 6, respectively including manufacturing apparatus 2 to manufacturing apparatus 6. Typically, a design of microelectronic components to be formed in a wafer is generated and a layout is performed according to the design. The layout may include a plurality of sets of patterns that are transferred to one or more stacked layers of material to be applied to the wafer during its manufacturing in a process sequence 7 to form various circuits and components on the wafer substrate. According to some embodiments, Figure 1The process sequence 7 shown is only an exemplary process, which can be used multiple times to deposit or form a thin film on the wafer substrate, and pattern the thin film using various lithography and etching techniques. Typically, the steps of this preparation may include: a deposition process (using deposition equipment 2), a planarization and / or grinding process (using CMP equipment 3), an exposure process with light wavelength patterning (using lithography equipment 4), a removal process of the exposed film (using etching equipment 5), and a cleaning process for preparing for subsequent processes (using cleaning equipment 6). It should be understood that more steps other than deposition, planarization, lithography, etching, and cleaning can be applied in the manufacturing facility 1, as understood by those skilled in the art. Moreover, the deposition, planarization, lithography, etching, and cleaning processes of each step may include various specific steps. Therefore, Figure 1 The process sequence 7 in the figure should not be construed as limiting the embodiments of the present disclosure. In some embodiments, for example, the deposition equipment 2 used in the deposition technology may include a chemical vapor deposition (CVD), an electrochemical deposition (ECD), a physical vapor deposition (PVD), a molecular beam epitaxy (MBE), an atomic layer deposition (ALD), a self-assembled monolayer (SAM), etc. In some embodiments, the deposition technology may be assisted by generating a plasma to appropriately affect the chemical reaction occurring in the substrate surface process. It should be further understood that in some embodiments, each of the manufacturing equipment 2 to the manufacturing equipment 6 may include at least one semiconductor manufacturing equipment, such as one or more transfer robots for transferring the wafer during the process sequence 7.

[0146] Figure 2 A functional block diagram of a control platform 20 of a manufacturing facility 1 illustrating some embodiments of the present disclosure. Figure 1 as well as Figure 2 , the control platform 20 includes multiple systems for controlling and monitoring manufacturing equipment of the manufacturing facility 1. The control platform 20 may include a system 11 for controlling the deposition equipment 2, a system 12 for controlling the CMP equipment 3, a system 13 for controlling the lithography equipment 4, a system 14 for controlling the etching equipment 5, and a system 15 for controlling the cleaning equipment 6. In some embodiments, such as the embodiments described later in the present disclosure, the systems 11 to 15 may include another system for monitoring the operation of an electrostatic chuck in the manufacturing equipment 2 to the manufacturing equipment 6 of the manufacturing facility 1. In addition, measurement or metrology data may be recorded to control and optimize the process performed by the manufacturing equipment of the manufacturing facility 1. For example, the electrostatic chuck for monitoring and controlling the manufacturing equipment 2 to the manufacturing equipment 6 may include multiple subsystems to control and monitor various parameters and configurations to optimize Figure 1 The effectiveness of the electrostatic chucks in the manufacturing apparatuses 2 to 6 will be described in later embodiments of the present disclosure.

[0147] Figure 3 Schematic diagram illustrating an electrostatic chuck 30 according to some embodiments of the present disclosure. Figure 4A as well as Figure 4B Surface mapping of an electrostatic chuck 30 according to some embodiments of the present disclosure. In some embodiments, the electrostatic chuck 30 may be mounted on Figure 1 The electrostatic chuck 30 may be installed in any one of the manufacturing tools 2 to 6 of the manufacturing facility 1 shown, or in other manufacturing equipment including an electrostatic chuck. For example, the electrostatic chuck 30 may be a semiconductor manufacturing tool that can be combined with a transfer robot (not shown) to handle wafers 32 during a semiconductor process in the manufacturing tools 2 to 6. Figure 3 The electrostatic chuck 30 may include a plurality of dimples 33, one or more electrodes 34, a heating portion 35, and a base portion 36. In some embodiments, the electrostatic chuck 30 can hold or release a wafer 32 by manipulating electrostatic energy. For example, a direct current (DC) voltage can be applied to the electrostatic chuck 30 through the base portion 36, causing half of the surface of the electrostatic chuck 30 to be positively charged and the other half to be negatively charged. Furthermore, in some embodiments, an argon gas can be supplied to the wafer 32 through the base portion 36 of the electrostatic chuck 30 to achieve more uniform heat distribution across the wafer 32. After the required semiconductor process is completed, the wafer 32 can be released from the electrostatic chuck 30 by applying a DC voltage of opposite polarity to neutralize the static charge between the wafer 32 and the electrostatic chuck 30. In some embodiments, for example, the released wafer 32 can then be transferred to another manufacturing facility by a transfer robot.

[0148] When the process of holding and releasing the wafer 32 is repeated many times, the warping of the wafer 32 may lead to unfavorable surface contact with the electrostatic chuck 30, causing wear of the dimples 33 that serve as the surface pad of the electrostatic chuck 30. As a result, the electrostatic chuck 30 becomes unbalanced, and in some cases, the wafer 32 may be broken during manufacturing. Therefore, in some embodiments, a sensor unit 101 may be added to record a set of data related to the operating status of the electrostatic chuck 30. Figure 4A as well as Figure 4B A surface mapping of the electrostatic chuck 30 according to some embodiments of the present disclosure is illustrated. Figure 4A as well as Figure 4B For example, the sensor unit 101 may record a map of the surface of the electrostatic chuck 30 to depict the degree of wear of the pits 33. Figure 4A Describe a sample surface map of a new or properly functioning electrostatic chuck 30, and Figure 4BDepicting another sample surface mapping of an abnormal electrostatic chuck 30. It should be understood that the sensor unit 101 can record other parameters or data related to the operating state of the electrostatic chuck 30, such as the temperature of the heating portion 35, voltage changes, chamber pressure, air pressure on the wafer 32, and similar parameters or data, so that a control system of the electrostatic chuck 30 can use the data to diagnose, build models, and predict the operating state and life of the electrostatic chuck 30. In addition, the sensor unit 101 may also include a camera element to record the operating state related to each surface of the electrostatic chuck 30. In some embodiments, the camera element may be an area scan camera or a line scan camera, a complementary metal oxide semiconductor (CMOS) camera, a charge coupled device (CCD) camera, a video camera, or other suitable type of camera or sensor for recording the required image of the electrostatic chuck 30. It should be understood that, Figure 3 The position of the sensor unit 101 shown is for illustration purposes only, and the sensor unit 101 may be placed in any location that facilitates data collection. Additionally, the components of the sensor unit 101 may be integrated into the sensor unit 101, or the components may be part of a separate kit that is coupled to the sensor unit 101.

[0149] Figure 5 A functional block diagram illustrating a control system 100 for controlling the semiconductor manufacturing equipment according to some embodiments of the present disclosure. Figure 6 A functional block diagram illustrating a sensor interface 102 according to some embodiments of the present disclosure. Figure 7 A functional block diagram of a control unit 106 is shown in FIG. Figures 5 to 7 The control system 100 for controlling the electrostatic chuck 30 includes a sensor unit 101, a sensor interface 102, a data server 103, a first transceiver 104, a second transceiver 105, and a control unit 106. In some embodiments, the semiconductor manufacturing equipment may include, for example Figure 31 , although the general operating principles of system 100 can be applied to other control systems within control platform 20 of fabrication facility 1. Sensor unit 101 records a set of data DA related to the operating status of electrostatic chuck 30. In some embodiments, sensor interface 102 receives data DA from sensor unit 101 and generates at least one input signal IN for data server 103, which may be, for example, a data server in a data warehouse. The at least one input signal IN is transmitted from data server 103 via first transceiver 104 to second transceiver 105 via network link 107. In some embodiments, the at least one input signal IN is modulated when transmitted by first transceiver 104 and demodulated when received by second transceiver 105. Network link 107 can be any suitable type of network link based on wired or wireless technology known in the art, including but not limited to infrared, optical, or radio communication devices. In some embodiments, the control unit 106 performs an operating status monitoring process to determine whether the electrostatic chuck 30 has failed based on the first data signal D1. The control unit 106 then executes an artificial intelligence (AI) analysis program to build a model and predict the operating status and lifespan of the electrostatic chuck 30. A feedback signal FB and an alarm signal AM are generated based on the output signal OS. In some embodiments, the feedback signal FB is transmitted from the control unit 106 to the electrostatic chuck 30 via a network link 107 between the second transceiver 105 and the first transceiver 104. In some embodiments, the feedback signal FB is modulated when transmitted by the second transceiver 105 and demodulated when received by the first transceiver 104. In some embodiments, operation of the electrostatic chuck 30 and / or the process equipment including the electrostatic chuck 30 can be terminated based on the feedback signal FB.

[0150] refer to Figure 6 The sensor interface 102 includes one or more signal converters 40 and one or more filters 41. In some embodiments, the signal converter 40 may include an image file converter (e.g., an image file compression converter), an analog-to-digital (A / D) converter, a digital-to-analog (D / A) converter, or other suitable signal converters. When generating an input signal IN for the data server 103, the signal converter 40 of the sensor interface 102 may convert the data set DA to another file format, compress or decompress the data set DA (if necessary), or perform other suitable types of conversions. The filter 41 may include a photographic filter, an anti-aliasing filter, a low-pass filter, a high-pass filter, a bandpass filter, or other suitable filter for a particular application of the control system 100. The filter 41 may be used to improve the signal-to-noise ratio of the input signal IN.

[0151] refer to Figure 7The control unit 106 includes a front-end subsystem 50, a computing subsystem 51, and an information and feedback subsystem 52. In some embodiments, the front-end subsystem 50 receives at least one input signal IN from the data server 103 and executes a front-end program to generate a first data signal D1. The computing subsystem 51 receives the first data signal D1 from the front-end subsystem 51. The computing subsystem 51 executes an operating status monitoring program based on the first data signal D1 to determine whether the electrostatic chuck 30 has failed, and executes an artificial intelligence (AI) analysis program to build a model and predict the operating status and lifespan of the electrostatic chuck 30, thereby generating an output signal OS. The information and feedback subsystem 52 generates an alarm signal AM and a feedback signal FB based on the output signal OS, and the information and feedback subsystem 52 transmits the alarm signal AM to a user of the electrostatic chuck 30.

[0152] Figure 8 A functional block diagram illustrating a front-end subsystem 50 according to some embodiments of the present disclosure is shown. Figure 8 , the front-end subsystem 50 may include an extract, transform and load (ETL) module 25 for executing the front-end program, the ETL including extracting, transforming and / or loading the input signal IN by the data server 103. In some embodiments, as Figure 8 As shown, the ETL module 25 includes an extraction area 60, a transformation area 61, and a loading area 62. In some embodiments, the extraction area 60 can unpack and extract at least one input signal IN from the data server 103. The transformation area 61 can perform appropriate data transformation, integrity, and validation procedures. The loading area 62 can generate and load a first data signal D1 based on the transformation procedures executed by the transformation area 61, allowing the computing subsystem 51 to use the first data signal D1 for the AI ​​analysis process.

[0153] Figure 9 A functional block diagram illustrating the computing subsystem 51 of some embodiments of the present disclosure. Figure 10 A state diagram illustrating the operating status monitoring routine executed by the diagnostic subsystem 70 in accordance with some embodiments of the present disclosure. Figure 11 A partial schematic diagram illustrating a classification process of the decision subsystem 71 executing the AI ​​analysis process in some embodiments of the present disclosure. Figures 9 to 11 The computing subsystem 51 includes a diagnostic subsystem 70 and a decision-making subsystem 71. In some embodiments, the diagnostic subsystem 70 performs the operation status monitoring process, determines whether the electrostatic chuck 30 has failed based on the first data signal D1, and generates a second data signal D2. In some embodiments, the decision-making subsystem 71 executes the AI ​​analysis program to establish a model and predict the operation and life of the electrostatic chuck 30, and generates an output signal OS.

[0154] In some embodiments, Figure 10The state diagram depicts the index score of the operating state of the electrostatic chuck 30 after multiple operations. For example, the index score can be included in various parameters of the electrostatic chuck 30 collected in the first data signal D1. Faults can be detected by setting the index score of various critical standards. For example, Figure 10 In the state diagram shown, the operating state of the electrostatic chuck 30 is considered normal when it is below the threshold value T1. Between the threshold value T1 and the threshold value T2, the operating state of the electrostatic chuck 30 is in a warning zone. In addition, a fault is detected when it is above the threshold value T2.

[0155] In some embodiments, the AI ​​analysis program executed by the decision subsystem 70 monitors and determines one or more trends TR in the second data signal D2 from the diagnostic subsystem 71. In some embodiments, the AI ​​analysis program executed by the decision subsystem 71 further utilizes a statistical model prediction (SMP) to obtain a device life prediction score for the electrostatic chuck 30, the device life prediction score corresponding to the set of data DA related to the operating state in the first data signal D1. Figure 11 As shown, the statistical model prediction SMP used by the decision subsystem 71 of the computing subsystem 51 can identify and classify the second data signal D2 into a matching trend group TR_MCH and a non-matching trend group TR_NMCH. It should be understood that the statistical model prediction SMP is a deep learning statistical model based on a neural network or a pseudo-neural network, although other suitable models may also be used.

[0156] In some embodiments, the information and feedback subsystem 52 generates an alert signal AM based on the output signal OS of the computing subsystem 51. In some embodiments, the alert signal AM may be sent to a user of the electrostatic chuck 30. Figure 12 as well as Figure 13 A functional block diagram illustrating a system 100 for controlling the semiconductor manufacturing equipment and a user according to some embodiments of the present disclosure. Figure 14A A diagram illustrating the relationship between the warning signal AM and its components according to some embodiments of the present disclosure. Figure 14B The following diagram illustrates the relationship between the feedback signal FB and its components in some embodiments of the present disclosure. Figure 12 as well as Figure 13 In some embodiments, the information and feedback subsystem 52 may send an alert signal AM to the user 109 via the network link 108 between the second transceiver 105 and the client system 110. Figure 12 As shown, for example, the client system 110 is a desktop computer, a laptop computer, a smart phone, a tablet computer, or a similar display device capable of displaying the alert signal AM to the user 109. Figure 14A, the content of the alert signal AM (including the data contained in the output signal OS) can be presented on the user graphical interface (GUI) 111 in the client system 110. The network link 108 can be any suitable type of network link based on wired or wireless technology existing in the art, including but not limited to infrared, optical, or radio communication devices. The alert signal AM can be sent to the client system 110 by email, real-time messaging application, or similar means. The network link 107 and the network link 108 can be part of the same network or different networks. The network link 107 and the network link 108 can be an internal network, an external network, an ad hoc network, a virtual private network (VPN), a local area network (LAN), a wireless local area network (WLAN), a wireless wide area WAN (WWAN), a wide area network (WAN), a metropolitan area network (MAN), part of the Internet, a public switched telephone network (PSTN), a cellular telephone network, or a combination of two or more thereof. In some embodiments, such as Figure 13 As shown, the warning signal AM may also be provided directly by the control unit 106 and displayed to the user 109. In some embodiments, the warning signal AM provides the user 109 with early warnings regarding equipment operation score warnings, equipment life prediction warnings, and faulty component warnings.

[0157] refer to Figure 14B In some embodiments, when the feedback signal FB is sent to the electrostatic chuck 30 for the automatic feedback process 112, the electrostatic chuck 30 is automatically adjusted or shut down based on the output signal OS. For example, the feedback signal FB may include a stop equipment command SEC and / or a wafer scrap command WSC based on the output signal OS. In some embodiments, the feedback signal FB may also include other commands such as alignment adjustment commands, power adjustment commands, or other process feedback commands. In some embodiments, the electrostatic chuck 30 may automatically adjust or shut down a run (run-to-run) based on the feedback signal FB. It should be understood that the feedback signal FB may be used directly by the electrostatic chuck 30 for the automatic feedback process 112, or the feedback signal FB may be sent to an independent controller 113 to control the automatic feedback process 112 of the electrostatic chuck 30.

[0158] It should be understood that one or more of the devices, subsystems, methods, or operations described in this disclosure may be implemented by a computer including instructions that are operable when executed by one or more processors. For example, the control unit 106 and the control method 600 described later in this disclosure may be implemented by Figure 15 Computer system implementation in . Figure 15 A functional block diagram of a computer system 80 illustrating some embodiments of the present disclosure. Figure 15The computer system 80 may include one or more processors 500, a network interface (I / F) 502, a storage device 506, a memory 508, and an input / output (I / O) device 504, and the above devices or components are communicatively coupled by a bus 514 or other interconnection communication mechanism. The memory 508 includes, in some embodiments, a random access memory (RAM), other dynamic storage device, read-only memory (ROM), or other static storage device, coupled to the bus 514 for storing data or instructions to be executed by one or more processors 500, and the memory 508 may also include a core 512, user space 510, a portion of the core or user space, and components thereof. The memory 508 is also used, in some embodiments, to store temporary variables or other intermediate information during the execution of instructions by one or more processors 500.

[0159] In some embodiments, the storage device 506 is coupled to the bus 514 for transmitting data or instructions to, for example, the kernel 512 and the user space 510. In some embodiments, operations and functions are implemented as functions of programs stored in the storage device 506, where the storage device 506 may include one or more computer-readable non-transitory storage media coupled to the processor 500. Computer-readable non-transitory storage media include, but are not limited to, external / removable or internal / built-in storage or memory units, such as one or more optical disks (e.g., DVDs), magnetic disks (e.g., hard disks), semiconductor memories (e.g., ROM, RAM, memory cards), and the like. In some embodiments, the computer-readable non-transitory storage media of the storage device 506 include instructions that, when executed by one or more processors 500, are operable to cause the control system 100 to receive at least one input signal IN from the data server 103, execute the front-end program to generate a data signal DS, execute the AI ​​analysis program based on the data signal DS to determine whether a fault 30F has occurred in the semiconductor manufacturing equipment and generate an output signal OS, and generate an alert signal AM and a feedback signal FB based on the output signal OS, and transmit the alert signal AM to a user 109 of the semiconductor manufacturing equipment. In some embodiments, the front-end processing includes extracting, converting, and / or loading the input signal IN by the data server 103. In some embodiments, the one or more computer-readable non-transitory storage media further include instructions operable when executed by the one or more processors 500 to cause the control system 100 to execute the AI ​​analysis program by monitoring and determining one or more trends in the data signal of the diagnostic subsystem. In some embodiments, the one or more computer-readable non-transitory storage media further include instructions operable when executed by the one or more processors 500 to cause the control system 100 to further execute the AI ​​analysis program by using a statistical model to predict and obtain a predicted equipment life score for the semiconductor manufacturing equipment, the predicted equipment life score corresponding to a set of data related to the operating state in the first data signal. In some embodiments, the one or more computer-readable non-transitory storage media further include instructions operable when executed by the one or more processors 500 to cause the control system 100 to further execute the AI ​​analysis program by using a statistical model to predict SMP to obtain a probability score LS for the semiconductor manufacturing equipment, the probability score LS corresponding to the image IMG of the data signal DS. In some embodiments, the one or more computer-readable non-transitory storage media further include instructions operable, when executed by the one or more processors 500, to cause the control system 100 to send a feedback signal FB based on the output signal OS for an automatic feedback process of the semiconductor manufacturing equipment. In some embodiments, the semiconductor manufacturing equipment includes one or more transfer robots 30.

[0160] In some embodiments, the input / output device 604 includes an input device, an output device, or a combination of input / output devices to enable user interaction with the control unit 106. The input device includes, for example, a computer keyboard, a keyboard, a mouse, a trackball, a trackpad, or a cursor arrow key for transmitting information and commands to the processor 500. The output device includes, for example, a display, a printer, and a speech synthesizer for transmitting information to the user. In some embodiments, one or more operations or functions of the apparatus, subsystems, and methods described in this disclosure are implemented by one or more processors 500 of the computer system 80, which are programmed to perform such operations or functions. One or more of the memory 508, the network interface (I / F) 502, the storage device 506, the I / O device 504, and the bus 514 are operable to receive instructions, data, design specifications, netlists, layouts, models, and other parameters for processing by the processor 500. In some embodiments, one or more operations or functions of the devices, subsystems, and methods described in this disclosure may be implemented by specifically configured hardware (e.g., by one or more application specific integrated circuits (ASICs)) separate from or in place of processor 500. Some embodiments include more than one operation or function in a single ASIC.

[0161] Figure 16 A flowchart illustrating a control method 600 of the semiconductor manufacturing equipment according to some embodiments of the present disclosure. In some embodiments, the control method 600 may be Figures 5 to 9 The control system 100 shown is implemented, and the control method 600 can also be implemented by Figure 15 The computer system 80 shown is implemented. Figure 16The control method 600 of the semiconductor manufacturing equipment includes: recording a set of images IMG of the semiconductor manufacturing equipment by the sensor unit 101 (step S601); receiving the set of images IMG by the sensor interface 102, and generating at least one input IN signal for the data server 103 (step S602); receiving the at least one input signal IN of the data server 103 by the front-end subsystem 50 and executing a front-end program to generate a data signal DS (step S603); receiving the data signal DS of the front-end subsystem 50 by the computing subsystem 51, wherein the computing subsystem 51 A diagnostic subsystem performs an operation status monitoring process to determine whether a fault 30MF occurs in the semiconductor manufacturing equipment and generates a second data signal, and a decision subsystem of the computing subsystem 51 performs an artificial intelligence (AI) analysis program according to the second data signal of the diagnostic subsystem to generate an output signal (step S604); and an alarm signal AM and a feedback signal FB are generated by the information and feedback subsystem 52 according to the output signal OS, and the information and feedback subsystem 52 sends the alarm signal AM to the user 109 of the semiconductor manufacturing equipment (step S605).

[0162] Figure 17 FIG6 is a flow chart of step S603 of the control method 600 according to some embodiments of the present disclosure. In some embodiments, the front-end program executed by the front-end subsystem 50 further includes extracting, converting, and / or loading an input signal IN from the data server 103 (step S6031); and generating a data signal DS by the front-end program (step S6032).

[0163] Figure 18 A flowchart illustrating step S604 of control method 600 according to some embodiments of the present disclosure. In some embodiments, the AI ​​analysis program executed by the computing subsystem further includes monitoring and determining one or more trends in the data signal of the diagnostic subsystem (step S6041). In some embodiments, the AI ​​analysis program executed by the computing subsystem further utilizes a statistical model to predict and obtain a predicted equipment lifespan score for the semiconductor manufacturing equipment, the predicted equipment lifespan score corresponding to a set of data related to the operating state in the first data signal (step S6042).

[0164] Figure 19 A flowchart illustrating step S605 of the control method 600 of some embodiments of the present disclosure is provided. In some embodiments, the control method 600 further includes: transmitting, by the information and feedback subsystem, a feedback signal FB for the automatic feedback program 112 of the semiconductor manufacturing equipment (step S6051); and executing the automatic feedback program 112 based on the output signal OS to generate an alarm signal, wherein the alarm signal includes an equipment operation score alarm, an equipment life prediction alarm, and a faulty component alarm (step S6052).

[0165] Thus, system 100 for controlling semiconductor manufacturing equipment and control method 600 provide an automated AI subsystem and its processing, capable of monitoring and determining one or more trends in the data signal from the diagnostic subsystem. Due to the AI ​​analysis program executed by computing subsystem 51, an operator of the semiconductor manufacturing equipment can obtain a predicted equipment lifespan score for the semiconductor manufacturing equipment, corresponding to the set of data regarding the operating state in the first data signal. Furthermore, the automated feedback process of control system 100 and control method 600 can reduce human error, process time, and the number of product defects.

[0166] One embodiment of the present disclosure provides a control system for semiconductor manufacturing equipment, comprising: a sensor unit configured to obtain a set of images of the semiconductor manufacturing equipment; a sensor interface configured to receive the set of data and generate at least one input signal for a data server; and a control unit. The control unit includes a front-end subsystem, a computing subsystem, and an information and feedback subsystem. The front-end subsystem receives at least one input signal from the data server and performs a front-end processing to generate a data signal. The computing subsystem receives the data signal of the front-end subsystem, wherein the computing subsystem executes an artificial intelligence (AI) analysis program to determine whether a fault occurs in the semiconductor manufacturing equipment based on the data signal and generates an output signal. The information and feedback subsystem generates an alarm signal and a feedback signal based on the output signal, and the information and feedback subsystem sends the alarm signal to a user of the semiconductor manufacturing equipment.

[0167] Another embodiment of the present disclosure provides a control system for semiconductor manufacturing equipment, comprising: a sensor unit configured to acquire a set of images of the semiconductor manufacturing equipment; a sensor interface configured to receive the set of images and generate at least one input signal for a data server, one or more processors, and one or more computer-readable non-transitory storage media. The one or more computer-readable non-transitory storage media are coupled to one or more processors and include instructions operable, when executed by the one or more processors, to: receive the at least one input signal from the data server and perform front-end processing to generate a data signal; execute an artificial intelligence (AI) analysis program to determine whether a fault has occurred in the semiconductor manufacturing equipment based on the data signal and generate an output signal; and generate an alarm signal and a feedback signal based on the output signal and transmit the alarm signal to a user of the semiconductor manufacturing equipment.

[0168] Yet another embodiment of the present disclosure provides a control method for semiconductor manufacturing equipment, comprising: obtaining a group of images of the semiconductor manufacturing equipment by a sensor unit; receiving the group of images by a sensor interface and generating at least one input signal for a data server; receiving the at least one input signal of the data server by a front-end subsystem and executing a front-end program to generate a data signal; receiving the data signal of the front-end subsystem by a computing subsystem, the computing subsystem executing an artificial intelligence (AI) analysis program to determine whether a fault occurs in the semiconductor manufacturing equipment based on the data signal and generating an output signal; and generating an alarm signal and a feedback signal by an information and feedback subsystem based on the output signal, the information and feedback subsystem sending the alarm signal to a user of the semiconductor manufacturing equipment.

[0169] Although the present disclosure and its advantages have been described in detail, it should be understood that various changes, substitutions and replacements can be made without departing from the concept and scope of the present disclosure as defined in the claims. For example, many of the above processes can be implemented in different ways, and many of the above processes can be replaced by other processes or combinations thereof.

[0170] Furthermore, the scope of the present disclosure is not limited to the specific embodiments of the processes, machines, manufactures, compositions of matter, means, methods, and steps described in the specification. Those skilled in the art will understand from the disclosure of this disclosure that existing or future developed processes, machines, manufactures, compositions of matter, means, methods, or steps that function the same as or achieve substantially the same results as the corresponding embodiments described herein may be used in accordance with the present disclosure. Accordingly, such processes, machines, manufactures, compositions of matter, means, methods, or steps are intended to be encompassed by the claims of this disclosure.

Claims

1. A control system for semiconductor manufacturing equipment, comprising: a sensor unit configured to record a set of data related to an operating state of the semiconductor manufacturing equipment; a sensor interface configured to receive the set of data and generate at least one input signal for a data server; as well as A control unit comprising: a front-end subsystem configured to receive the at least one input signal from the data server and execute a front-end program to generate a first data signal; a diagnostic subsystem that receives the first data signal from the front-end subsystem, executes an operation status monitoring program to determine whether a fault occurs in the semiconductor manufacturing equipment, and generates a second data signal; a decision subsystem configured to execute an artificial intelligence analysis program based on the second data signal of the diagnostic subsystem to generate an output signal, wherein the artificial intelligence analysis program executed by the decision subsystem further utilizes a statistical model prediction to obtain an equipment life prediction score for the semiconductor manufacturing equipment, the equipment life prediction score corresponding to the set of data related to the operating state in the first data signal; wherein the statistical model prediction utilized by the decision subsystem identifies the second data signal and classifies it into a matching trend group and a non-matching trend group; An information and feedback subsystem generates a warning signal and a feedback signal according to the output signal, and the information and feedback subsystem sends the warning signal to a user of the semiconductor manufacturing equipment. 2 . The control system of claim 1 , wherein the front-end program executed by the front-end subsystem comprises extracting, converting and / or loading the input signal by the data server.

3. The control system of claim 1 , wherein the artificial intelligence analysis program executed by the decision-making subsystem monitors and determines one or more trends in the second data signal of the diagnostic subsystem. 4 . The control system of claim 3 , wherein the information and feedback subsystem sends a feedback signal according to the output signal for an automatic feedback process of the semiconductor manufacturing equipment.

5. The control system of claim 1, wherein the warning signal comprises a device operation score warning, a device life prediction warning, and a faulty component warning generated according to the output signal, wherein the device life prediction warning comprises a predicted life value of the semiconductor manufacturing equipment.

6. The control system of claim 1, wherein the semiconductor manufacturing equipment comprises one or more electrostatic chucks.

7. A control system for semiconductor manufacturing equipment, comprising: a sensor unit configured to record a set of data related to an operating state of the semiconductor manufacturing equipment; a sensor interface configured to receive the set of data and generate at least one input signal for a data server; one or more processors; as well as One or more computer-readable non-transitory storage media configured to be coupled to the one or more processors and comprising instructions operable when executed by the one or more processors to cause the control system to: The data server receives the at least one input signal and executes a front-end program to generate a first data signal; executing an operation status monitoring program to determine whether a fault occurs in the semiconductor manufacturing equipment and generating a second data signal; executing an artificial intelligence analysis program to generate an output signal according to the second data signal; executing the artificial intelligence analysis program to obtain a predicted equipment life score for the semiconductor manufacturing equipment by using a statistical model prediction, the predicted equipment life score corresponding to the set of data related to the operating state in the first data signal; Using a statistical model to predict and identify the second data signal and classify it into a matching trend group and a non-matching trend group; as well as A warning signal and a feedback signal are generated according to the output signal, and the warning signal is sent to a user of the semiconductor manufacturing equipment. 8 . The control system of claim 7 , wherein the front-end process comprises extracting, converting and / or loading the input signal by the data server.

9. The control system of claim 7, wherein the one or more computer-readable non-transitory storage media further comprise instructions operable when executed by the one or more processors to cause the control system to execute the artificial intelligence analysis program by monitoring and determining one or more trends in the second data signal.

10. The control system of claim 7, wherein the one or more computer-readable non-transitory storage media further comprise instructions operable when executed by one or more processors to cause the control system to send the feedback signal for an automatic feedback procedure of the semiconductor manufacturing equipment based on the output signal.

11. The control system of claim 7, wherein the warning signal comprises a device operation score warning, a device life prediction warning, and a faulty component warning generated according to the output signal, wherein the device life prediction warning comprises a predicted life value of the semiconductor manufacturing equipment.

12. The control system of claim 7, wherein the semiconductor manufacturing equipment comprises one or more electrostatic chucks.

13. A method for controlling semiconductor manufacturing equipment, comprising: A sensor unit records a set of data related to an operating state of the semiconductor manufacturing equipment; receiving the set of data from a sensor interface and generating at least one input signal for a data server; A front-end subsystem receives the at least one input signal from the data server and performs a front-end process to generate a first data signal; A diagnostic subsystem receives the first data signal of the front-end subsystem and executes an operation status monitoring program to determine whether a fault occurs in the semiconductor manufacturing equipment and generates a second data signal; A decision subsystem executes an artificial intelligence analysis program according to the second data signal of the diagnosis subsystem to generate an output signal; The decision subsystem uses a statistical model to predict and obtain a device life prediction score for the semiconductor manufacturing equipment, wherein the device life prediction score corresponds to the set of data related to the operating state in the first data signal; Using the statistical model to predict and identify the second data signal and classify it into a matching trend group and a non-matching trend group; and An information and feedback subsystem generates a warning signal and a feedback signal according to the output signal and sends the warning signal to a user of the semiconductor manufacturing equipment. 14 . The control method of claim 13 , wherein the front-end program executed by the front-end subsystem further comprises extracting, converting and / or loading the input signal by the data server. 15 . The control method of claim 13 , wherein the artificial intelligence analysis program executed by the decision-making subsystem further comprises monitoring and determining one or more trends in the second data signal of the diagnosis subsystem. 16 . The control method of claim 13 , further comprising sending, by the information and feedback subsystem according to the output signal, the feedback signal for an automatic feedback process of the semiconductor manufacturing equipment.

17. The control method of claim 13, wherein the warning signal comprises an equipment operation score warning, an equipment life prediction warning, and a faulty component warning generated according to the output signal, the equipment life prediction warning comprises a predicted life value of the semiconductor manufacturing equipment, and wherein the semiconductor manufacturing equipment comprises one or more transfer robots.

Images