JP2021157693A - Operation assisting device, operation assisting program, operation assisting method and polishing device - Google Patents

Abstract

To further surely detect an abnormality occurring on a semiconductor manufacturing device.SOLUTION: An operation assisting device 6 includes: a determination object data obtaining unit 61 that obtains determination object data which are data on a semiconductor manufacturing device and which are subjected to a determination on whether or not a predetermined standard is satisfied; a data determining unit 62 that determines whether or not the determination object data satisfy the predetermined standard using an already-learnt model created by inputting, into a machine learning device, learning data that contain first data which are data on the semiconductor manufacturing device and which satisfy the predetermined standard, and second data which are on the semiconductor manufacturing device and which do not satisfy the predetermined standard; and a determination result data output unit 63 that outputs determination result data representing a result determined by the data determining unit 62.SELECTED DRAWING: Figure 1

Description

The present invention relates to an operation support device, an operation support program, an operation support method, and a polishing device.

Conventionally, a control system for a semiconductor manufacturing apparatus described in Patent Document 1 below has been known. The control system creates a prediction formula for QC value data using PLS regression based on the EES parameter group and the QC value data group. The control system then inputs into the prediction formula the first EES parameters corresponding to the wafers in the first lot processed newly after the lot last processed by the semiconductor manufacturing equipment. Predict the first QC value data of the wafer in the first lot. Then, the control system orders the semiconductor manufacturing apparatus to process the wafer based on the predicted first QC value data.

JP-A-2007-242809

However, for example, when a defect occurs in the semiconductor manufacturing apparatus, or when the parts of the semiconductor manufacturing apparatus are worn out, at least one of the EES parameter and the QC value data may contain an abnormal value. Therefore, the control system described above may not be able to create a suitable prediction formula for the QC value data, and may not be able to instruct the semiconductor manufacturing apparatus to perform suitable processing of the wafer.

The present invention has been made in view of the above problems, and an object of the present invention is to more reliably detect an abnormality occurring in a semiconductor manufacturing apparatus.

The operation support device according to one aspect of the present invention is data related to a semiconductor manufacturing device, and is a judgment target data acquisition unit that acquires judgment target data to be determined whether or not it satisfies a predetermined criterion, and a semiconductor manufacturing unit. Data related to the device, the first data that meets the predetermined criteria and the data related to the semiconductor manufacturing device, which are generated by inputting the second data that does not meet the predetermined criteria into the machine learning device as training data. A determination result that outputs a determination result data indicating the result of determination by the data determination unit and a data determination unit that determines whether or not the determination target data satisfies the predetermined criteria using the trained model. It is equipped with a data output unit.

Further, in the operation support device, when the data determination unit determines that the determination target data does not satisfy the predetermined criteria, the first data, the second data, and the second data are acquired. The factor for which the determination target data has been acquired may be determined using a trained model generated by inputting factor data indicating the factor into the machine learning device as training data.

Further, in the operation support device, when the data determination unit determines that the determination target data does not satisfy the predetermined criteria, the first data, the second data, and the second data are acquired. Judgment of the necessary correspondence when the judgment target data is acquired using the trained model generated by inputting at least one of the correspondence data indicating the necessary correspondence in the case into the machine learning device as training data. You may.

Further, in the operation support device, the determination target data acquisition unit may acquire at least one of the still image data and the moving image data that visualize the data related to the semiconductor manufacturing apparatus as the determination target data.

Further, in the operation support device, when the ratio of the determination target data matching the first data is less than a predetermined threshold value, the data determination unit does not satisfy the predetermined criterion. May be determined.

Further, in the operation support device, when the ratio of the determination target data matching the second data exceeds a predetermined threshold value, the data determination unit satisfies the determination target data. It may be determined that there is no such thing.

The operation support program according to one aspect of the present invention is data related to a semiconductor manufacturing apparatus, and has a judgment target data acquisition function for acquiring judgment target data to be judged whether or not it satisfies a predetermined standard, and semiconductor manufacturing. Data related to the device, the first data that meets the predetermined criteria and the data related to the semiconductor manufacturing device, which are generated by inputting the second data that does not meet the predetermined criteria into the machine learning device as training data. A data judgment function for judging whether or not the judgment target data satisfies the predetermined criteria using the trained model, and a judgment result for outputting judgment result data indicating the judgment result by the data judgment function. Realize the data output function.

The operation support method according to one aspect of the present invention is data related to a semiconductor manufacturing apparatus, and includes a determination target data acquisition step for acquiring determination target data to be determined whether or not a predetermined criterion is satisfied, and a semiconductor manufacturing. Data related to the device, the first data that meets the predetermined criteria and the data related to the semiconductor manufacturing device, which are generated by inputting the second data that does not meet the predetermined criteria into the machine learning device as training data. A data determination step for determining whether or not the determination target data satisfies the predetermined criteria using the trained model, and a determination for outputting determination result data indicating the result determined in the data determination step. Includes a result data output step.

The polishing device according to one aspect of the present invention includes any one of the above-mentioned operation support devices.

According to the above aspect of the present invention, it is possible to more reliably detect an abnormality that has occurred in a semiconductor manufacturing apparatus.

It is a figure which shows an example of the substrate processing system which concerns on embodiment. It is a figure which shows an example of the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the pressure of the pressurization / depressurization area of the top ring provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the case where the graph of the pressure of each pressurization / depressurization area of the top ring is overlapped. It is a figure which shows an example of the graph which shows the time change of the rotation speed of the top ring provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the rotational torque of the top ring provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the flow rate of the polishing liquid supplied from the polishing liquid supply nozzle provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the case where the graphs shown in FIGS. 3 to 7 are superimposed. It is a figure which shows an example of the graph which shows the time change of the rotational torque of the pencil sponge type cleaning module provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the flow rate of the dressing liquid supplied from the polishing liquid supply nozzle provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the case where the graph shown in FIG. 9, the graph shown in FIG. 10 and the like are superposed. An example of a graph showing the time change of the rotation torque of the top ring immediately after the consumables of the substrate processing device are replaced, and a graph showing the time change of the rotation torque of the top ring when the consumables of the board processing device are consumed. It is a figure which shows an example. It is a figure which shows an example of the graph which shows the time change of the pressure in the normal state of each regulator provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the pressure at the time of abnormality of each regulator provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the load in the normal state of the pencil sponge type cleaning module provided in the substrate processing apparatus which concerns on embodiment. It is a figure which shows an example of the graph which shows the time change of the load at the time of abnormality of the pencil sponge type cleaning module provided in the substrate processing apparatus which concerns on embodiment. It is a flowchart which shows an example of the process executed by the operation support apparatus which concerns on embodiment.

An example of the substrate processing system 100 according to the embodiment will be described with reference to FIGS. 1 to 16. FIG. 1 is a diagram showing an example of a substrate processing system according to an embodiment. As shown in FIG. 1, the substrate processing system 100 includes a substrate processing device 1 and an operation support device 6. The substrate processing device 1 and the operation support device 6 may be formed as an integrated device.

FIG. 2 is a diagram showing an example of the substrate processing apparatus according to the embodiment. The substrate processing apparatus 1 shown in FIG. 2 is a chemical mechanical polishing (CMP) apparatus for flatly polishing the surface of a substrate W such as a silicon wafer.

The substrate processing device 1 includes a rectangular box-shaped housing 2. The housing 2 is formed in a substantially rectangular shape in a plan view. The housing 2 includes a substrate transport path 3 extending in the longitudinal direction in the center thereof. A load / unload portion 10 is provided at one end of the substrate transport path 3 in the longitudinal direction. The polishing portion 20 is disposed on one side of the substrate transport path 3, that is, in the direction orthogonal to the longitudinal direction in a plan view, and the cleaning portion 30 is disposed on the other side. The substrate transport path 3 is provided with a substrate transport unit 40 that transports the substrate W. Further, the substrate processing device 1 includes a control unit 50 that comprehensively controls the operations of the load / unload unit 10, the polishing unit 20, the cleaning unit 30, and the substrate transport unit 40.

The load / unload unit 10 includes a front load unit 11 that accommodates the substrate W. A plurality of front load portions 11 are provided on one side surface of the housing 2 in the longitudinal direction. The plurality of front load portions 11 are arranged in the width direction of the housing 2. The front load unit 11 is equipped with, for example, an open cassette, a SMIF (Standard Manufacturing Interface) pod, or a FOUP (Front Opening Unified Pod). Both SMIF and FOUP are airtight containers in which the cassette of the substrate W is housed and covered with a partition wall, and can maintain an environment independent of the external space.

Further, the load / unload unit 10 includes two transfer robots 12 that move the board W in and out from the front load unit 11, and a traveling mechanism 13 that causes each transfer robot 12 to travel along the line of the front load unit 11. Each transfer robot 12 is provided with two hands on the upper and lower sides, and these two bands are used properly before the processing of the substrate W and after the processing of the substrate W. For example, each transfer robot 12 uses the upper hand when returning the board W to the front load portion 11, and uses the lower hand when taking out the unprocessed board W from the front load portion 11. do.

The polishing unit 20 includes a plurality of polishing devices 21 (21A, 21B, 21C, 21D) for polishing the substrate W. The plurality of polishing devices 21 are arranged in the longitudinal direction of the substrate transport path 3. The polishing device 21 holds a polishing table 23 for rotating the polishing pad 22 having a polishing surface and a top ring 24 for polishing while holding the substrate W and pressing the substrate W against the polishing pad 22 on the polishing table 23. A polishing liquid supply nozzle 25 for supplying a polishing liquid, a dressing liquid, etc. to the polishing pad 22, a dresser 26 for dressing the polishing surface of the polishing pad 22, a liquid such as pure water, and nitrogen gas. It is provided with an atomizer 27 that atomizes a mixed fluid of gas or a liquid such as pure water and injects it onto the polished surface.

The polishing apparatus 21 presses the substrate W against the polishing pad 22 by the top ring 24 while supplying the polishing liquid onto the polishing pad 22 from the polishing liquid supply nozzle 25, and relatively moves the top ring 24 and the polishing table 23. As a result, the substrate W is polished to flatten the surface of the substrate W. Further, the top ring 24 includes a plurality of pressurizing / depressurizing areas arranged concentrically. The top ring 24 adjusts the degree of pressing of the substrate W against the polishing pad 22 by adjusting the pressure in these plurality of pressurizing / depressurizing areas.

In the dresser 26, hard particles such as diamond particles and ceramic particles are fixed to the rotating portion at the tip in contact with the polishing pad 22, and the rotating portion is swung while rotating to make the entire polished surface of the polishing pad 22. Dress evenly to form a flat polished surface.

The atomizer 27 purifies the polished surface and sharpens the polished surface by the dresser 26, that is, regenerates the polished surface by washing away the polishing debris, abrasive grains, etc. remaining on the polishing surface of the polishing pad 22 with a high-pressure fluid. ..

The cleaning unit 30 includes a plurality of cleaning devices 31 (31A, 31B) for cleaning the substrate W, and a substrate drying device 32 for drying the cleaned substrate W. The plurality of cleaning devices 31 and the substrate drying device 32 are arranged in the longitudinal direction of the substrate transport path 3. A first transfer chamber 33 is provided between the cleaning device 31A and the cleaning device 31B. The first transfer chamber 33 is provided with a transfer robot 35 that transfers the substrate W between the substrate transfer unit 40, the cleaning device 31A, and the cleaning device 31B. A second transfer chamber 34 is provided between the cleaning device 31B and the substrate drying device 32. The second transfer chamber 34 is provided with a transfer robot 36 that transfers the substrate W between the cleaning device 31B and the substrate drying device 32.

The cleaning device 31 includes a roll sponge type cleaning module, and uses this cleaning module to clean the substrate W. The cleaning device 31A and the cleaning device 31B may be of the same type or may be different types of cleaning modules. Further, the cleaning device 31A and the cleaning device 31B may include, for example, a pencil sponge type cleaning module and a two-fluid jet type cleaning module instead of the roll sponge type cleaning module.

The substrate drying device 32 includes, for example, a drying module that performs rotagoni drying using isopropyl alcohol (IPA: Iso-Propyl Alcohol). After the rotagoni drying is executed, the substrate W is carried out from the substrate drying device 32 by the transfer robot 12 after the shutter 1a provided on the partition wall between the substrate drying device 32 and the load / unload portion 10 is opened. ..

The substrate transport unit 40 includes a lifter 41, a first linear transporter 42, a second linear transporter 43, and a swing transporter 44. In the board transport path 3, the first transport position TP1, the second transport position TP2, the third transport position TP3, the fourth transport position TP4, the fifth transport position TP5, and the sixth transport are in order from the load / unload portion 10 side. The position TP6 and the seventh transport position TP7 are set.

The lifter 41 is a mechanism for transporting the substrate W up and down at the first transport position TP1. The lifter 41 receives the substrate W from the transfer robot 12 of the load / unload unit 10 at the first transfer position TP1. Then, the lifter 41 delivers the substrate W received from the transfer robot 12 to the first linear transporter 42. A shutter 1b is provided on the partition wall between the first transport position TP1 and the load / unload portion 10. When the substrate W is transported, the shutter 1b is opened and the substrate W is received from the transport robot 12 to the lifter 41. Passed.

The first linear transporter 42 is a mechanism for transporting the substrate W between two of the first transport position TP1, the second transport position TP2, the third transport position TP3, and the fourth transport position TP4. The first linear transporter 42 includes a plurality of transport hands 45 (45A, 45B, 45C, 45D) and a linear guide mechanism 46 that moves each transport hand 45 at a plurality of heights in the horizontal direction.

The transport hand 45A moves between the first transport position TP1 and the fourth transport position TP4 by the linear guide mechanism 46. The transport hand 45A is a pass hand for receiving the substrate W from the lifter 41 and delivering the substrate W to the second linear transporter 43.

The transport hand 45B moves between the first transport position TP1 and the second transport position TP2 by the linear guide mechanism 46. The transport hand 45B receives the substrate W from the lifter 41 at the first transport position TP1 and delivers the substrate W to the polishing apparatus 21A at the second transport position TP2. The transport hand 45B is provided with an elevating drive unit, and rises when the substrate W is delivered to the top ring 24 of the polishing apparatus 21A, and descends after the substrate W is delivered to the top ring 24. The transport hand 45C and the transport hand 45D are also provided with a similar elevating drive unit.

The transport hand 45C moves between the first transport position TP1 and the third transport position TP3 by the linear guide mechanism 46. The transport hand 45C receives the substrate W from the lifter 41 at the first transport position TP1 and delivers the substrate W to the polishing apparatus 21B at the third transport position TP3. The transport hand 45C also functions as an access hand that receives the substrate W from the top ring 24 of the polishing device 21A at the second transport position TP2 and delivers the substrate W to the polishing device 21B at the third transport position TP3.

The transport hand 45D moves between the second transport position TP2 and the fourth transport position TP4 by the linear guide mechanism 46. The transfer hand 45D receives the substrate W from the top ring 24 of the polishing device 21A or the polishing device 21B at the second transfer position TP2 or the third transfer position TP3, and receives the substrate W from the swing transporter 44 at the fourth transfer position TP4. Functions as an access hand for passing.

The swing transporter 44 has a hand that can move between the fourth transport position TP4 and the fifth transport position TP5, and delivers the substrate W from the first linear transporter 42 to the second linear transporter 43. .. Further, the swing transporter 44 delivers the substrate W polished by the polishing unit 20 to the cleaning unit 30. A temporary storage base 47 for the substrate W is provided on the side of the swing transporter 44. The swing transporter 44 flips the substrate W received at the fourth transport position TP4 or the fifth transport position TP5 upside down and places it on the temporary storage table 47. The substrate W placed on the temporary storage table 47 is conveyed to the first transfer chamber 33 by the transfer robot 35 of the cleaning unit 30.

The second linear transporter 43 is a mechanism for transporting the substrate W between two of the fifth transport position TP5, the sixth transport position TP6, and the seventh transport position TP7. The second linear transporter 43 includes a plurality of transport hands 48 (48A, 48B, 48C) and a linear guide mechanism 49 for horizontally moving each transport hand 45 at a plurality of heights. The transport hand 48A moves between the fifth transport position TP5 and the sixth transport position TP6 by the linear guide mechanism 49. The transport hand 45A functions as an access hand that receives the substrate W from the swing transporter 44 and delivers the substrate W to the polishing apparatus 21C.

The transport hand 48B moves between the sixth transport position TP6 and the seventh transport position TP7. The transport hand 48B functions as an access hand for receiving the substrate W from the polishing apparatus 21C and delivering the substrate W to the polishing apparatus 21D. The transport hand 48C moves between the seventh transport position TP7 and the fifth transport position TP5. The transfer hand 48C receives the substrate W from the top ring 24 of the polishing device 21C or the polishing device 21D at the sixth transfer position TP6 or the seventh transfer position TP7, and delivers the substrate W to the swing transporter 44 at the fifth transfer position TP5. Acts as an access hand for. Although the description is omitted, the operation of the transfer hand 48 at the time of delivery of the substrate W is the same as the operation of the first linear transporter 42 described above.

As shown in FIG. 1, the operation support device 6 includes a determination target data acquisition unit 61, a data determination unit 62, and a determination result data output unit 63.

The determination target data acquisition unit 61 acquires determination target data to be determined whether or not it satisfies a predetermined criterion via a storage device or a network. The storage device referred to here is, for example, a USB (Universal Serial Bus) memory or a solid state drive (SSD). Further, the network referred to here is, for example, an intranet or a LAN (Local Area Network). The determination target data acquisition unit 61 uses a file transfer or a web browser when acquiring determination target data via the network.

The determination target data includes data relating to the substrate processing device 1, for example, data indicating a physical quantity related to each component constituting the substrate processing apparatus 1, data indicating a physical quantity relating to the operation of the substrate processing apparatus 1, and the substrate processing apparatus 1 having a substrate W. It is the data which shows the amount of the chemicals used when polishing. Further, the determination target data includes not only numerical data indicating these physical quantities but also still image data and moving image data that visualize the data related to the substrate processing device 1. Further, the determination target data confirms the performance of the substrate processing device 1 at the sales destination or the lease destination, for example, when the substrate processing device 1 is started up at the manufacturer, or when the substrate processing device 1 is started up at the sales destination or the lease destination. Occasionally, it is acquired when the substrate W is processed as usual by using the substrate processing apparatus 1 at the sales destination or the lease destination. Alternatively, the determination target data is changed, for example, when a transfer occurs during processing of the substrate W using the substrate processing device 1 at the sales destination or the lease destination, or when the substrate processing device 1 is maintained. , Acquired when the operation method of the substrate processing device 1 is changed. The data to be determined may be expressed in physical system units such as millimeters (mm), Newton (N), pascal (Pa), minutes (min), ampere (A), volt (V), and so on. It may be represented by an electrical system unit such as a bit. The predetermined standard is, for example, an upper limit, a lower limit, a threshold value set for the determination target data, and a ratio that matches the first data or the second data described later.

The data determination unit 62 uses the trained model generated by inputting the first data and the second data as training data into the machine learning device to determine whether or not the determination target data satisfies the above-mentioned predetermined criteria. To judge. The machine learning device includes, for example, a neural network (NN). Further, the neural network referred to here is a convolutional neural network (CNN: Convolutional Neural Network) and a recurrent neural network (RNN: Recurrent Neural Network). The machine learning device generates a trained model by executing, for example, deep learning, reinforcement learning, and the like.

The first data is data that satisfies the above-mentioned predetermined criteria, for example, data that is acquired when the substrate processing apparatus 1 is operating normally. On the other hand, the second data is data that does not satisfy the above-mentioned predetermined criteria, for example, when the substrate processing apparatus 1 is not operating normally, or when the substrate processing apparatus 1 needs to be maintained, the substrate processing apparatus 1 is used. This is data acquired when parts are worn out.

Further, when the data determination unit 62 determines that the determination target data does not satisfy the predetermined criteria, the data determination unit 62 may determine the factor from which the determination target data was acquired. For this determination, a trained model generated by inputting the first data, the second data, and the factor data indicating the factor from which the second data was acquired into the machine learning device as training data is used. The machine learning device referred to here may be the same as or different from the machine learning device described above. Further, the trained model referred to here may be the same as or different from the trained model described above.

Further, when the data determination unit 62 determines that the determination target data does not satisfy the predetermined criteria, the data determination unit 62 may determine the necessary measures when the determination target data is acquired. For this determination, the first data, the second data, and the learned data generated by inputting at least one of the correspondence data indicating the correspondence required when the second data is acquired into the machine learning device as the training data have been completed. The model is used.

Further, the correspondence data is, for example, data indicating the correspondence described in the operation manual of the board processing device 1, and is data indicating the correspondence that has been performed in the past by a person who is proficient in the operation of the board processing device 1. It is preferable to have. Specifically, the corresponding data is data indicating the correspondence described in the operation manual of the board processing device 1, and is sold to a person who is proficient in the operation of the board processing device 1, for example, a device of the same model. It is preferable that the data is supervised by a person who is familiar with the operation of the substrate processing apparatus 1 at the destination or the lease destination. For example, when the maximum value of the pressure in the pressurizing / depressurizing area shown in FIG. 3 is less than a predetermined threshold value, the substrate W is less likely to be peeled off from the top ring 24. Further, when the substrate W is not peeled off from the top ring 24, the substrate processing device 1 detects the phenomenon that the substrate W is not peeled off from the top ring 24 from the above-mentioned still image data or moving image data, and this phenomenon occurs. Outputs an alarm indicating that it has been done.

The determination result data output unit 63 outputs determination result data indicating the result of determination by the data determination unit 62. The judgment result data is data indicating whether or not the judgment target data satisfies a predetermined criterion when the factor for which the judgment target data is acquired and the necessary response when the judgment target data is acquired are not determined. Become. Further, in the judgment result data, when the factor from which the judgment target data is acquired is determined and the necessary response when the judgment target data is acquired is not determined, the judgment target data satisfies a predetermined criterion. It is data showing whether or not it is present and the factors concerned. Further, in the judgment result data, when the factor from which the judgment target data is acquired is not determined and the necessary response when the judgment target data is acquired is determined, the judgment target data satisfies a predetermined criterion. The data indicates whether or not the data is provided and the necessary measures are taken.

The judgment result data is output to the display, for example, whether or not the judgment target data satisfies a predetermined criterion, the factor from which the judgment target data is acquired, and at least one of the measures required when the judgment target data is acquired. Used to display one. Further, the determination result data may be transferred to another device.

After the determination result data is acquired via the storage device or the network, the determination result data is taken into the operation support device 6 in which a dedicated viewer for referring to the physical quantity indicated by the determination result data is installed. Next, the determination result data is analyzed by the operation support device 6 and used to output the result of determination as to whether or not the predetermined criteria are satisfied. This result is, for example, "There is something wrong with XX, so please replace the part.", "Since the value of XX is abnormal, change the set value of △△ to XX.", "Work XX. Please adjust according to the procedure. " These messages may be output as characters or images displayed on the display, or may be output as audio output from the speaker. Then, this result is output to the display via the storage device or the network, and is displayed on the display. After visually recognizing this display result, the user performs operations such as replacement of parts, change of set values, and mechanical adjustment of the substrate processing device 1. After that, the process from capturing the determination result display to outputting the above-mentioned message is repeated, and it is confirmed that the abnormality or the like has been resolved.

Next, examples of the above-mentioned determination target data, the first data, and the second data will be described with reference to FIGS. 3 to 16.

FIG. 3 shows an example of the above-mentioned determination target data and the first data. FIG. 3 is a diagram showing an example of a graph showing the time change of the pressure in the pressurizing / depressurizing area of the top ring included in the substrate processing apparatus according to the embodiment. In FIG. 3, the horizontal axis represents time, and the vertical axis represents the pressure in the pressurizing / depressurizing area of the top ring 24. Further, the range from 0 seconds to 80 seconds in FIG. 3 is the time required for polishing one substrate W.

FIG. 4 is a diagram showing an example in which graphs of pressure in each pressurization / depressurization area of the top ring are superimposed. In FIG. 4, the horizontal axis represents time, and the vertical axis represents the pressure in any of the plurality of pressurization / depressurization areas. The pressure in the pressurization / depressurization area shown in FIGS. 3 to 7 is measured by, for example, a diffusion type semiconductor pressure transducer attached to the substrate processing apparatus 1. As shown in FIG. 4, at least one of the determination target data and the first data may be still image data showing a still image or moving image data showing a moving image drawing a graph of a plurality of physical quantities.

FIG. 5 shows an example of the above-mentioned determination target data and the first data. FIG. 5 is a diagram showing an example of a graph showing a time change of the rotation speed of the top ring included in the substrate processing apparatus according to the embodiment. In FIG. 5, the horizontal axis represents time and the vertical axis represents the rotation speed of the top ring 24. Further, the range from 0 seconds to 65 seconds in FIG. 7 is the time required for polishing one substrate W. The rotation speed shown in FIG. 5 is measured by, for example, a sensor mounted on an AC servomotor that drives the top ring 24. Alternatively, the rotation speed shown in FIG. 5 is calculated from, for example, the measured value of the current or the measured value of the voltage of the AC servomotor.

FIG. 6 shows an example of the above-mentioned determination target data and the first data. FIG. 6 is a diagram showing an example of a graph showing a time change of the rotational torque of the top ring included in the substrate processing apparatus according to the embodiment. In FIG. 6, the horizontal axis represents time, and the vertical axis represents the rotational torque of the top ring 24. Further, the range from 0 seconds to 60 seconds in FIG. 6 is the time required for polishing one substrate W. The rotational torque shown in FIG. 6 is measured by, for example, a sensor mounted on an AC servomotor that drives the top ring 24. Alternatively, the rotational torque shown in FIG. 6 is calculated from, for example, the measured value of the current or the measured value of the voltage of the AC servomotor.

FIG. 7 shows an example of the above-mentioned determination target data and the first data. FIG. 7 is a diagram showing an example of a graph showing a time change of the flow rate of the polishing liquid supplied from the polishing liquid supply nozzle included in the substrate processing apparatus according to the embodiment. In FIG. 7, the horizontal axis represents time and the vertical axis represents the flow rate of the polishing liquid. Further, the range from 0 seconds to 60 seconds in FIG. 7 is the time required for polishing one substrate W. The flow rate of the polishing liquid shown in FIG. 7 is measured by, for example, a flow meter attached to the polishing liquid supply nozzle 25.

FIG. 8 is a diagram showing an example in the case where the graphs shown in FIGS. 3 to 7 are superimposed. In FIG. 8, the horizontal axis indicates time, and the vertical axis of any of FIGS. 3 to 7 indicates the physical quantity. As shown in FIG. 8, at least one of the determination target data and the first data may be still image data showing a still image or moving image data showing a moving image drawing a graph of a plurality of physical quantities. Regarding the polishing unit 20, the determination target data, the first data, and the second data are, for example, the torque value of the polishing table 23, the flow rate of the fluid supplied to the top ring 24, the pressure of the fluid, and the top. The load when pressing the ring 24 against the polishing table 23, the flow rate of the abrasive liquid discharged from the polishing fluid supply nozzle 25, the pressure when pressing the dresser 26 against the polishing table 23, and the torque value when pressing the dresser 26 against the polishing table 23. At least two of the above may be superimposed in a complex manner.

FIG. 9 shows an example of the above-mentioned determination target data and the first data. FIG. 9 is a diagram showing an example of a graph showing a time change of the rotational torque of the pencil sponge type cleaning module included in the substrate processing apparatus according to the embodiment. In FIG. 9, the horizontal axis represents time, and the vertical axis represents the rotational torque of the pencil sponge type cleaning module. Further, the range from 10 seconds to 80 seconds in FIG. 9 is the time required for polishing one substrate W. The rotational torque shown in FIG. 9 is measured by, for example, a sensor mounted on an AC servomotor that drives a pencil sponge type cleaning module. Alternatively, the rotational torque shown in FIG. 9 is calculated from, for example, a measured value of the current or a measured value of the voltage of the AC servomotor.

FIG. 10 shows an example of the above-mentioned determination target data and the first data. FIG. 10 is a diagram showing an example of a graph showing a time change of the flow rate of the dressing liquid supplied from the polishing liquid supply nozzle included in the substrate processing apparatus according to the embodiment. In FIG. 10, the horizontal axis represents time and the vertical axis represents the flow rate of the dressing liquid. Further, the range from 10 seconds to 80 seconds in FIG. 10 is the time required for polishing one substrate W. The flow rate of the dressing liquid shown in FIG. 10 is measured by a flow meter attached to the polishing liquid supply nozzle 25. The dressing liquid is, for example, pure water, and is supplied from the chemical supply unit attached to the substrate processing apparatus 1.

FIG. 11 is a diagram showing an example in the case where the graph shown in FIG. 9 and the graph shown in FIG. 10 are superimposed. In FIG. 11, the horizontal axis represents time, and the vertical axis represents the rotational torque of the pencil sponge type cleaning module, the flow rate of the dressing liquid, and the like. As shown in FIG. 11, at least one of the determination target data and the first data may be still image data showing a still image or moving image data showing a moving image drawing a graph of a plurality of physical quantities. Regarding the cleaning unit 30, the determination target data, the first data, and the second data are, for example, the rotational torque of the roll sponge type cleaning module, the pressing load of the roll sponge type cleaning module, and the rotation of the substrate W. At least two of the number and the flow rate of the chemical solution may be superposed in a complex manner. Further, regarding the cleaning unit 30, the determination target data, the first data, and the second data are, for example, the rotational torque of the pencil sponge type cleaning module, the pressing load of the pencil sponge type cleaning module, and the rotation of the substrate W. At least two of the number and the flow rate of the chemical solution may be superposed in a complex manner. Further, the data to be determined, the first data, and the second data are, for example, a composite superposition of at least two of the rotation speed of the substrate W and the flow rate of the chemical solution for the substrate drying device 32. You may.

The reason for using a graph in which a plurality of physical quantity graphs and the like are superimposed with respect to FIGS. 12 and 15 is as follows. For example, even if the abnormal signal of the torque value of the polishing table 23 is due to the increased rubbing between the top ring 24 and the polishing pad 22 on the polishing table 23, the top ring 24 has the original positional relationship. There are various reasons such as an increase in inclination from the horizontal, cracks or cracks in the components in contact with the polishing pad 2 of the top ring 24, cracks in the substrate W, and the like. Therefore, it becomes easier to identify the reason by preparing and comparing other related data up to the values before and after the abnormal signal is detected in time series. Further, the data to be superimposed may include not only the element data of each mechanism in the polishing unit 20 but also the element data of the mechanism other than the polishing unit 20 depending on the detection target. For example, as element data of the mechanism other than the polishing portion 20, the raising and lowering of each of the first linear transporter 42 and the second linear transporter 43, the motor torque of the traveling shaft, the speed of the traveling shaft, the input of the seating sensor, and the output of the seating sensor G, raising and lowering of the swing transporter 44, motor torque of the swivel shaft, speed of the swivel shaft, input of the limit sensor and output of the limit sensor. Further, these data may or may not have the same data format. Similarly, the superimposed data may include element data of each mechanism in the cleaning unit 30 depending on the detection target.

Further, as shown in FIGS. 12 and 15, when a graph in which a plurality of physical quantity graphs and the like are superimposed is used, at least one of the horizontal axis scale and the vertical axis scale may not match each other. .. In order to avoid a situation where at least one of the horizontal axis scale and the vertical axis scale does not match each other, the horizontal axis scale and the vertical axis scale may be aligned as much as possible. On the other hand, even if at least one of the horizontal axis scale and the vertical axis scale does not match each other, it can be used as it is. However, if at least one of the scale on the horizontal axis and the scale on the vertical axis do not match each other, the condition that the graph obtained by superimposing the graphs of a plurality of physical quantities in a complex manner does not affect the operation is satisfied. Only when there is, a graph in which a graph of a plurality of physical quantities is superimposed is used.

Further, a graph obtained by superimposing a graph of a plurality of physical quantities in a complex manner may not be displayed on the display, but may be simply input to the machine learning device and used for machine learning. Furthermore, in machine learning using a graph in which graphs of a plurality of physical quantities are superimposed in a complex manner, it is not always necessary to pay attention to the unit of each physical quantity, and the graphs of a plurality of physical quantities are compounded. The superimposed graph may be captured as an image.

FIG. 12 shows an example of a graph showing the time change of the rotational torque of the top ring immediately after the consumables of the substrate processing apparatus are replaced, and the time of the rotational torque of the top ring when the consumables of the substrate processing apparatus are consumed. It is a figure which shows an example of the graph which shows the change. In FIG. 12, the horizontal axis represents time and the vertical axis represents the rotational torque of the top ring.

The graph drawn by the solid line in FIG. 12 is a graph showing the time change of the rotational torque of the top ring 24 immediately after the consumables of the substrate processing apparatus 1 are replaced, and is the graph of the determination target data and the first data described above. This is an example.

On the other hand, the graph drawn by the dotted line in FIG. 12 is a graph showing the time change of the rotational torque of the top ring 24 when the consumables of the substrate processing apparatus 1 are consumed, and the above-mentioned determination target data and the first This is an example of two data. The graph drawn by the dotted line in FIG. 12 is similar in shape to the graph drawn by the solid line in FIG. 12, but the rotational torque value is about 10% lower.

Therefore, when the graph drawn by the solid line in FIG. 12 is the first data and the graph drawn by the dotted line in FIG. 12 is the judgment target data, the ratio of matching between the two is 90%. .. Further, for example, when this ratio is less than 60%, the determination result data is data including data indicating that the consumables need to be replaced.

FIG. 13 is a diagram showing an example of a graph showing a time change of pressure in each pressurization / depressurization area when the top ring included in the substrate processing apparatus according to the embodiment is normal. In FIG. 13, the horizontal axis represents time, and the vertical axis represents the pressure in each pressurization / depressurization area. Specifically, the thin solid line, the thin broken line, the thin alternate long and short dash line, the thick solid line and the thick broken line indicate the time change of the first pressurization / decompression area, the second pressurization / decompression area, the third pressurization / decompression area and the fourth pressurization / decompression area, respectively. ing. The graphs shown in FIG. 13 are examples of the determination target data and the first data.

FIG. 14 is a diagram showing an example of a graph showing a time change of pressure in each pressurization / depressurization area when the top ring included in the substrate processing apparatus according to the embodiment is abnormal. In FIG. 14, the horizontal axis represents time, and the vertical axis represents the pressure in each pressurization / depressurization area. Specifically, a thin solid line, a thin broken line, a thin alternate long and short dash line, a thick solid line, and a thick broken line indicate the time change of each pressurization / depressurization area. The graph drawn by a thin solid line in FIG. 14 is an example of the determination target data and the second data. On the other hand, the other graphs drawn in FIG. 14 are examples of the determination target data and the first data.

The graph drawn with a thin solid line in FIG. 14 contains a singular point surrounded by an ellipse. In this case, the determination result data is data including data indicating that the valve needs to be replaced.

FIG. 15 shows an example of the above-mentioned determination target data and the first data. FIG. 15 is a diagram showing an example of a graph showing a time change of a load in a normal state of a pencil sponge type cleaning module included in the substrate processing apparatus according to the embodiment. In FIG. 15, the horizontal axis represents time, and the vertical axis represents the load of the pencil sponge type cleaning module.

FIG. 16 shows an example of the above-mentioned determination target data and the second data. FIG. 16 is a diagram showing an example of a graph showing a time change of a load when an abnormality occurs in a pencil sponge type cleaning module included in the substrate processing apparatus according to the embodiment. In FIG. 16, the horizontal axis represents time, and the vertical axis represents the load of the pencil sponge type cleaning module.

The graph shown in FIG. 16 contains singular points surrounded by an ellipse. In this case, the determination result data is data including data indicating that the regulator mounted on the substrate processing device 1 needs to be readjusted.

Next, an example of the process executed by the operation support device according to the embodiment will be described with reference to FIG. FIG. 17 is a flowchart showing an example of processing executed by the operation support device according to the embodiment. Further, the process shown in FIG. 17 is executed, for example, at the timing when the setup or maintenance / inspection of the substrate processing device 1 is performed.

In step S10, the determination target data acquisition unit 61 acquires the determination target data that is the data related to the semiconductor manufacturing apparatus and is the target of the determination of whether or not the predetermined criteria are satisfied.

In step S20, the data determination unit 62 determines whether or not the determination target data satisfies a predetermined criterion. When the data determination unit 62 determines that the determination target data satisfies the predetermined criteria (step S20: YES), the data determination unit 62 proceeds to the process in step S30. On the other hand, when the data determination unit 62 determines that the determination target data does not satisfy the predetermined criteria (step S20: NO), the process proceeds to step S40.

In step S30, the determination result data output unit 63 outputs the determination result data indicating the result determined in step S20.

In step S40, the data determination unit 62 determines the factor from which the determination target data was acquired.

In step S50, the data determination unit 62 determines the necessary response when the determination target data is acquired.

In step S60, the determination result data output unit 63 outputs determination result data indicating the result determined in step S40 from step S20.

The operation support device 6 may be executed by changing the order of the process executed in step S40 and the process executed in step S50. Further, the operation support device 6 does not have to execute at least one of the process executed in step S40 and the process executed in step S50.

The substrate processing system 100 according to the embodiment has been described above, focusing on the operation support device 6. The operation support device 6 determines whether or not the determination target data satisfies a predetermined criterion by using the trained model generated by inputting the first data and the second data as learning data into the machine learning device. do. As a result, the operation support device 6 determines that the determination target data does not meet the predetermined criteria, that is, that an abnormality has occurred in the board processing device 1, that the parts of the board processing device 1 are worn out, and the like. It can be detected more reliably.

Further, when the operation support device 6 determines that the determination target data does not satisfy the predetermined criteria, the operation support device 6 uses the factor data indicating the factors for which the first data, the second data, and the second data are acquired as learning data for machine learning. The factor for which the determination target data was acquired is determined using the trained model generated by inputting to the device. As a result, the operation support device 6 can notify the user of the board processing device 1 of the factor that the determination target data does not satisfy the predetermined criteria, and can more accurately grasp the state of the board processing device 1.

Further, when the operation support device 6 determines that the determination target data does not satisfy the predetermined criteria, at least the correspondence data indicating the necessary correspondence when the first data, the second data, and the second data are acquired. The required correspondence is determined when the determination target data is acquired using the trained model generated by inputting one as training data into the machine learning device. As a result, the operation support device 6 notifies the user of the board processing device 1 of the measures that should be taken when the determination target data does not meet the predetermined criteria, the measures to be taken, and the like, and causes the board processing device 1 to perform. It can be made possible to use it more accurately.

Further, the operation support device 6 acquires at least one of the still image data and the moving image data, which is a visualization of the data related to the semiconductor manufacturing device, as the determination target data. As a result, the operation support device 6 can reflect the knowledge derived for the first time by examining a plurality of parts of the graph, the knowledge derived for the first time by comprehensively examining the graph, and the like in the determination result data. Further, as a result, the operation support device 6 reads the phenomenon occurring in the substrate processing device 1 by the user looking at the graphs by overlapping a plurality of graphs or showing fine behavior of the graphs. Even if it is difficult to do so, the phenomenon can be detected and reflected in the determination result data.

In the above-described embodiment, the case where the operation support device 6 acquires the determination target data from the substrate processing device 1 has been described as an example, but the present invention is not limited to this. The operation support device 6 may acquire determination target data from a semiconductor processing device other than the substrate processing device 1, for example. Further, the semiconductor processing apparatus other than the substrate processing apparatus 1 is installed in, for example, another chemical mechanical polishing apparatus, a bevel polishing apparatus installed in the front stage or the rear stage of the substrate processing apparatus 1, and the front stage or the rear stage of the substrate processing apparatus 1. It is a plating device. Further, the determination target data may be acquired for each of a plurality of polishing modules or a plurality of cleaning modules included in the substrate processing apparatus 1, analyzed for each of these modules, and compared among these modules.

Further, at least a part of the functions of the operation support device 6 is a circuit unit such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), and a GPU (Graphics Processing Unit). It may be realized by hardware including circuits).

Further, at least a part of the functions of the operation support device 6 may be realized by the cooperation of these hardware and software. The software may be stored in, for example, a storage device including a non-transient storage medium, read by the operation support device 6, and executed. The storage device is, for example, a hard disk drive (HDD: Hard Disk Drive) or a solid state drive (SSD: Solid State Drive). Alternatively, the software may be stored in a storage device including a removable non-transient storage medium, read by the operation support device 6, and executed. The storage device is, for example, a DVD or a CD-ROM.

Further, in the above-described embodiment, the case where the substrate processing device 1 and the operation support device 6 are separate devices has been described as an example, but the present invention is not limited to this. For example, at least a part of the operation support device 6 may be included in the substrate processing device 1. Further, at least a part of the operation support device 6 referred to here is, for example, a determination target data result unit 63. Further, the operation support device 6 may be incorporated in the control unit 50 shown in FIG. Further, the operation support device 6 is installed, for example, at the manufacturer of the board processing device 1, the sales destination of the board processing device 1, or the leasing destination.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the embodiments of the present invention are not limited to the above-described embodiments, and various modifications and substitutions are made without departing from the gist of the present invention. And at least one of the design changes can be added.

100 ... Board processing system, 1 ... Board processing device, 6 ... Operation support device, 61 ... Judgment target data acquisition unit, 62 ... Data judgment unit, 63 ... Judgment result data output unit

Claims (9)

Judgment target data acquisition unit that acquires judgment target data that is data related to semiconductor manufacturing equipment and is the target of judgment as to whether or not it meets a predetermined standard.
The data relating to the semiconductor manufacturing apparatus, the first data relating to the predetermined criteria and the data relating to the semiconductor manufacturing apparatus, and the second data not satisfying the predetermined criteria are input to the machine learning apparatus as learning data. A data determination unit that determines whether or not the determination target data satisfies the predetermined criteria using the trained model generated by
A judgment result data output unit that outputs judgment result data indicating the judgment result of the data judgment unit, and
Operation support device equipped with. When the data determination unit determines that the determination target data does not satisfy the predetermined criteria, the data determination unit learns the first data, the second data, and factor data indicating the factors from which the second data was acquired. The factor for which the determination target data was acquired is determined using the trained model generated by inputting the data into the machine learning device.
The operation support device according to claim 1. When the data determination unit determines that the determination target data does not satisfy the predetermined criteria, the data determination unit indicates a response necessary when the first data, the second data, and the second data are acquired. Using the trained model generated by inputting at least one of the data as training data into the machine learning device, the necessary response when the determination target data is acquired is determined.
The operation support device according to claim 1 or 2. The determination target data acquisition unit acquires at least one of the still image data and the moving image data that visualize the data related to the semiconductor manufacturing apparatus as the determination target data.
The operation support device according to any one of claims 1 to 3. When the ratio of the determination target data matching the first data is less than a predetermined threshold value, the data determination unit determines that the determination target data does not satisfy the predetermined criteria.
The operation support device according to any one of claims 1 to 4. When the ratio of the determination target data matching the second data exceeds a predetermined threshold value, the data determination unit determines that the determination target data does not satisfy the predetermined criteria.
The operation support device according to any one of claims 1 to 5. On the computer
Judgment target data acquisition function that acquires judgment target data that is data related to semiconductor manufacturing equipment and is the target of judgment as to whether or not it meets a predetermined standard, and
The data relating to the semiconductor manufacturing apparatus, the first data relating to the predetermined criteria and the data relating to the semiconductor manufacturing apparatus, and the second data not satisfying the predetermined criteria are input to the machine learning apparatus as learning data. A data determination function for determining whether or not the determination target data satisfies the predetermined criteria using the trained model generated by
A judgment result data output function that outputs judgment result data indicating the judgment result by the data judgment function, and
Operation support program to realize. Judgment target data acquisition step to acquire the judgment target data that is the data related to the semiconductor manufacturing equipment and is the target of the judgment of whether or not the predetermined criteria are satisfied, and the judgment target data acquisition step.
The data relating to the semiconductor manufacturing apparatus, the first data relating to the predetermined criteria and the data relating to the semiconductor manufacturing apparatus, and the second data not satisfying the predetermined criteria are input to the machine learning apparatus as learning data. A data determination step for determining whether or not the determination target data satisfies the predetermined criteria using the trained model generated by
A judgment result data output step that outputs judgment result data indicating the judgment result in the data judgment step, and
Operation support methods including. A polishing device including the operation support device according to any one of claims 1 to 6.

Images