SUBSTRATE PROCESSING APPARATUS, SUBSTRATE PROCESSING SYSTEM, AND DATA PROCESSING METHOD

Abstract

To provide a substrate processing device, a substrate processing system, and a data processing method that can guarantee the reliability of a processing condition output from a learned model.SOLUTION: A substrate processing device 100 includes a thickness measurement unit 8, a control unit 102, and a storage unit 103. The thickness measurement unit 8 measures the thickness of a target object TG included in a substrate W. Due to input of input data to a learned model LM, the control unit 102 causes a processing condition to be output from the learned model. The storage unit 103 stores reference data RE obtained on the basis of a plurality of pieces of learning data LD used to build the trained model LM. The control unit 102 acquires pre-processing measurement data indicating the thickness of the target object TG before executing substrate processing, and generates input data on the basis of the target data indicating the target value of the thickness of the target object TG and the pre-processing measurement data. The control unit 102 compares the input data with the reference data RE to determine whether to perform substrate processing.SELECTED DRAWING: Figure 12

Description

[Technical field]

[0001] The present invention relates to a substrate processing apparatus, a substrate processing system, and a data processing method. [Background technology]

[0002] A substrate processing apparatus that implements a trained model has been proposed (see, for example, Patent Document 1). Such a substrate processing apparatus operates based on output data output from the trained model.

[0003] The substrate processing apparatus of Patent Document 1 includes a discharge nozzle, a camera, and a controller. The camera captures an image of the tip of the discharge nozzle and the processing liquid discharged from the tip. The camera captures images during the period in which the processing liquid is discharged onto the substrate, and acquires multiple image data. The controller has a machine-learned classifier (trained model). The classifier classifies the image data into one of multiple categories (classes).

[0004] The categories are set according to the discharge state of the treatment liquid. Specifically, the multiple categories include a category indicating a normal discharge state in which the treatment liquid flows down as a continuous flow from the discharge nozzle, a category indicating a dripping state in which the treatment liquid falls as droplets when the discharge of the treatment liquid is stopped, and a category indicating a discharge stop state in which the treatment liquid is not discharged.

[0005] The control unit determines the discharge state of the treatment liquid based on the result of classification (output data) by the classifier. When the control unit determines that the discharge state of the treatment liquid is a dripping state, it notifies an operator of an error. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2020-32409 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, it is difficult to guarantee the reliability of the output data output from the trained model. That is, the output data of the trained model may not be optimal data. For example, when the trained model outputs processing conditions (setting values) that are conditions for substrate processing, non-optimal processing conditions (non-optimal setting values) may be set. When non-optimal processing conditions are set, the state of the substrate after the substrate processing may differ from the target state. Therefore, there is room for further improvement in the substrate processing apparatus that implements the trained model.

[0008] The present invention has been made in consideration of the above-mentioned problems, and its object is to provide a substrate processing apparatus, a substrate processing system, and a data processing method that can guarantee the reliability of processing conditions output from a trained model. [Means for solving the problem]

[0009] According to one aspect of the present invention, a substrate processing apparatus performs a substrate processing, which is a process for a substrate. The substrate processing apparatus includes a thickness measuring unit, a control unit, and a storage unit. The thickness measuring unit measures a thickness of an object included in the substrate. The control unit inputs input data indicating a target value of a processing amount by the substrate processing to a trained model that outputs processing conditions at the time of performing the substrate processing, thereby causing the trained model to output the processing conditions. The storage unit stores reference data acquired based on a plurality of learning data used to construct the trained model. The learning data indicates a processing amount by the substrate processing at the time of learning. Before performing the substrate processing, the control unit causes the thickness measuring unit to measure a thickness of the object, and acquires pre-processing measurement data indicating a thickness of the object before performing the substrate processing. The control unit generates the input data based on the pre-processing measurement data and target data indicating a target value of the thickness of the object. The control unit compares the input data with the reference data to determine whether or not to perform the substrate processing.

[0010] In one embodiment, the storage unit stores at least one threshold value for a comparison result between the input data and the reference data, and the control unit compares the input data with the reference data to obtain the comparison result, and determines whether or not to perform the substrate processing based on the comparison result and the at least one threshold value.

[0011] In one embodiment, the control unit selects one of a plurality of decision items based on the comparison result and the at least one threshold value. The plurality of decision items include a first decision item, a second decision item, and a third decision item. The first decision item is an item for determining not to perform the substrate processing. The second decision item is an item for determining to perform the substrate processing based on the processing conditions output from the trained model and to store information indicating that the substrate processing has been performed in the storage unit. The third decision item is an item for determining to perform the substrate processing based on the processing conditions output from the trained model.

[0012] In one embodiment, the at least one threshold value includes a first threshold value and a second threshold value smaller than the first threshold value. The control unit selects the first decision item when the comparison result is greater than the first threshold value. The control unit selects the second decision item when the comparison result is greater than the second threshold value and equal to or less than the first threshold value. The control unit selects the third decision item when the comparison result is equal to or less than the second threshold value.

[0013] In one embodiment, the storage unit further stores predefined default conditions as the processing conditions. The plurality of decision items further includes a fourth decision item. The fourth decision item is an item for deciding to perform the substrate processing based on the predefined conditions and to store information indicating that the substrate processing has been performed in the storage unit.

[0014] In one embodiment, the at least one threshold value includes a first threshold value and a second threshold value that is smaller than the first threshold value. The control unit selects the first decision item when the comparison result is greater than the first threshold value. The control unit selects the second decision item or the fourth decision item when the comparison result is greater than the second threshold value and equal to or smaller than the first threshold value. The control unit selects the third decision item when the comparison result is equal to or smaller than the second threshold value.

[0015] In one embodiment, the substrate processing apparatus further includes a display unit for displaying a setting screen, the setting screen including a setting field for setting the at least one threshold value.

[0016] In one embodiment, the substrate processing apparatus further includes a display unit that displays a setting screen including a first setting field for setting one of the plurality of decision items for the at least one threshold value.

[0017] In one embodiment, the setting screen further includes a second setting field for setting the at least one threshold value.

[0018] In one embodiment, the setting screen further includes a graph display section that displays a graph in which the plurality of learning data are quantified.

[0019] In one embodiment, the setting screen displays the at least one threshold value in the graph display field.

[0020] In one embodiment, the substrate processing apparatus further includes a nozzle configured to eject a processing liquid toward the substrate, the substrate processing including ejecting the processing liquid from the nozzle toward the substrate.

[0021] In one embodiment, the substrate processing apparatus further includes a nozzle moving mechanism that moves the nozzle when the substrate processing is performed.

[0022] In an embodiment, the processing conditions include a movement speed of the nozzle.

[0023] In one embodiment, the processing liquid includes an etching liquid that etches the object.

[0024] According to another aspect of the present invention, a substrate processing system includes a thickness measuring device and a substrate processing device. The thickness measuring device measures a thickness of an object included in a substrate. The substrate processing device performs a substrate processing, which is a process for the substrate, after the thickness measuring device measures the thickness of the object. The substrate processing device includes a control unit and a storage unit. The control unit inputs input data indicating a target value of a processing amount by the substrate processing to a trained model that outputs processing conditions at the time of execution of the substrate processing, to the trained model, thereby causing the trained model to output the processing conditions. The storage unit stores reference data acquired based on a plurality of learning data used to construct the trained model. The learning data indicates a processing amount by the substrate processing at the time of learning. The control unit acquires pre-processing measurement data indicating a thickness of the object before execution of the substrate processing from a measurement result of the thickness measuring device. The control unit generates the input data based on the pre-processing measurement data and target data indicating a target value of the thickness of the object. The control unit compares the input data with the reference data to determine whether or not to execute the substrate processing.

[0025] According to another aspect of the present invention, a substrate processing system includes a thickness measuring device, a determination device, and a substrate processing apparatus. The thickness measuring device measures a thickness of an object included in a substrate. The determination device determines whether or not to execute a substrate processing, which is a processing for the substrate. The substrate processing apparatus causes a trained model that outputs processing conditions at the time of execution of the substrate processing to output the processing conditions, and executes the substrate processing. The determination device includes a storage unit and a determination unit. The storage unit stores reference data acquired based on a plurality of learning data used to construct the trained model. The determination unit determines whether or not to execute the substrate processing. The learning data indicates a processing amount by the substrate processing at the time of learning. The determination unit acquires pre-processing measurement data indicating a thickness of the object before execution of the substrate processing from a measurement result of the thickness measuring device. The determination unit generates input data indicating a target value of the processing amount by the substrate processing based on target data indicating a target value of the thickness of the object and the pre-processing measurement data. The determination unit compares the input data with the reference data to determine whether or not to execute the substrate processing. The substrate processing apparatus includes a control unit. When the decision device decides to perform the substrate processing, the control unit inputs the input data to the trained model and outputs the processing conditions from the trained model.

[0026] According to another aspect of the present invention, a data processing method includes the steps of measuring a thickness of an object included in a substrate before a substrate processing operation is performed, and acquiring pre-processing measurement data indicating a measurement result of the thickness of the object, generating input data indicating a target value of a processing amount for the substrate processing based on target data indicating a target value of the thickness of the object and the pre-processing measurement data, and comparing the input data with reference data acquired based on a plurality of learning data used to construct a trained model, to determine whether or not to perform the substrate processing. The learning data indicates a processing amount for the substrate processing during learning. The trained model outputs processing conditions during the substrate processing operation when the input data is input. Effect of the Invention

[0027] The substrate processing apparatus, substrate processing system, and data processing method according to the present invention can ensure the reliability of the processing conditions output from the trained model. [Brief description of the drawings]

[0028] [Figure 1] 1 is a schematic diagram of a substrate processing apparatus according to a first embodiment of the present invention. [Diagram 2] 2 is a schematic diagram of a processing unit included in the substrate processing apparatus; FIG. [Diagram 3] 1A is a plan view showing a probe moving process, and FIG. 1B is a plan view showing a thickness measuring process. [Figure 4] FIG. 11 is a plan view showing a scanning process of a substrate by a first nozzle. [Diagram 5] FIG. 13 is a diagram showing scan speed information. [Figure 6] 13 is a graph showing an example of the moving speed of a first nozzle. [Figure 7] FIG. 4 is a schematic diagram of a chemical solution supply unit. [Figure 8] FIG. 2 is a block diagram of a control device provided in the substrate processing apparatus. [Figure 9] FIG. 13 is a diagram showing a learning data set used to construct the reference data. [Figure 10] FIG. 13 is a diagram illustrating the relationship between the variation in the amount of processing included in a plurality of learning data and a threshold value. [Figure 11] FIG. [Figure 12] 4 is a flow chart showing a process executed by a control device of the substrate processing apparatus. FIG. [Figure 13] FIG. 11 is a flow diagram showing a determination process. [Figure 14] FIG. 11 is a flow diagram showing a condition setting process. [Figure 15] FIG. 1 is a flow diagram showing a substrate processing. [Figure 16] FIG. 11 is a block diagram of a control device provided in a substrate processing apparatus according to a second embodiment of the present invention. [Figure 17]FIG. 11 is a flow diagram showing a condition setting process. [Figure 18] FIG. 11 is a flowchart showing a process executed by a control device of a substrate processing apparatus according to a third embodiment of the present invention. [Figure 19] FIG. 11 is a flow diagram showing a process of generating additional learning data. [Figure 20] FIG. 13 is a diagram showing a substrate processing system according to a fourth embodiment of the present invention. [Figure 21] FIG. 13 is a diagram showing a substrate processing system according to a fifth embodiment of the present invention. [Figure 22] FIG. 1 is a block diagram showing a configuration of an information processing device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] Hereinafter, embodiments of the substrate processing apparatus, substrate processing system, and data processing method of the present invention will be described with reference to the drawings (FIGS. 1 to 22). However, the present invention is not limited to the following embodiments, and can be embodied in various forms without departing from the gist of the present invention. Note that where explanations are repeated, they may be omitted as appropriate. Also, in the drawings, the same or corresponding parts are denoted by the same reference symbols, and explanations will not be repeated.

[0030] In the substrate processing apparatus, substrate processing system, and data processing method according to the present invention, the "substrate" to be processed can be a semiconductor wafer, a glass substrate for a photomask, a glass substrate for a liquid crystal display, a glass substrate for a plasma display, a substrate for an FED (Field Emission Display), a substrate for an optical disk, a substrate for a magnetic disk, or a substrate for a magneto-optical disk. In the following, an embodiment of the present invention will be described mainly taking as an example a case in which a disk-shaped semiconductor wafer is the substrate to be processed, but the substrate processing apparatus, substrate processing system, and data processing method according to the present invention can be similarly applied to various substrates other than the above-mentioned semiconductor wafer. The shape of the substrate is not limited to a disk shape, and the substrate processing apparatus, substrate processing system, and data processing method according to the present invention can be applied to substrates of various shapes.

[0031] [Embodiment 1] A first embodiment of the present invention will be described below with reference to Figs. 1 to 15. First, a substrate processing apparatus 100 of the present embodiment will be described with reference to Fig. 1. Fig. 1 is a schematic diagram of the substrate processing apparatus 100 of the present embodiment. More specifically, Fig. 1 is a schematic plan view of the substrate processing apparatus 100. The substrate processing apparatus 100 performs substrate processing. More specifically, the substrate processing apparatus 100 is a single-wafer type apparatus, and performs substrate processing for each substrate W. The substrate processing is processing on the substrate W.

[0032] As shown in FIG. 1, the substrate processing apparatus 100 includes a plurality of processing units 1, a fluid cabinet 100A, a plurality of fluid boxes 100B, a plurality of load ports LP, an indexer robot IR, a center robot CR, and a controller 101.

[0033] Each load port LP accommodates a stack of substrates W. For example, the load port LP accommodates substrates W after grinding processing. The indexer robot IR transports the substrates W between the load port LP and the center robot CR. The center robot CR transports the substrates W between the indexer robot IR and the processing unit 1. Note that a placement stage (path) on which the substrate W is temporarily placed may be provided between the indexer robot IR and the center robot CR, and the device may be configured to indirectly transfer the substrate W between the indexer robot IR and the center robot CR via the placement stage.

[0034] The processing units 1 form a plurality of towers TW (four towers TW in FIG. 1). The towers TW are arranged to surround the center robot CR in a plan view. Each tower TW includes a plurality of processing units 1 (three processing units 1 in FIG. 1) stacked vertically.

[0035] The fluid cabinet 100A contains a processing liquid. Each of the fluid boxes 100B corresponds to one of the towers TW. The processing liquid in the fluid cabinet 100A is supplied to all the processing units 1 included in the tower TW corresponding to the fluid box 100B via one of the fluid boxes 100B.

[0036] Each of the processing units 1 supplies a processing liquid to the upper surface of the substrate W. The processing liquid includes a chemical liquid and a rinsing liquid. In this embodiment, each of the processing units 1 performs an etching process. The chemical liquid is an etching liquid. The upper surface of the substrate W is etched by the etching liquid. For example, each of the processing units 1 removes grinding marks generated by the grinding process by etching.

[0037] The etching solution is, for example, hydrofluoric nitric acid (a mixture of hydrofluoric acid (HF) and nitric acid (HNO3)), hydrofluoric acid, buffered hydrofluoric acid (BHF), ammonium fluoride, HFEG (a mixture of hydrofluoric acid and ethylene glycol), or phosphoric acid (H3PO4).The rinsing solution is, for example, deionized water, carbonated water, electrolytic ionized water, hydrogen water, ozone water, or hydrochloric acid water with a diluted concentration (for example, about 10 ppm to 100 ppm).

[0038] Next, the control device 101 will be described. The control device 101 controls the operation of each part of the substrate processing apparatus 100. For example, the control device 101 controls the load port LP, the indexer robot IR, and the center robot CR. The control device 101 includes a control unit 102 and a storage unit 103.

[0039] The control unit 102 includes a processor. The control unit 102 includes, for example, a central processing unit (CPU) or a micro processing unit (MPU). Alternatively, the control unit 102 may include a general-purpose computing device or a dedicated computing device. The control unit 102 may further include a neural network processing unit (NCU).

[0040] The memory unit 103 stores data and computer programs. The data includes recipe data RP (see FIG. 8). The recipe data RP indicates a recipe that specifies the processing content and processing procedure of the substrate W. In the recipe, processing conditions (setting values) when the substrate processing is performed are set. The computer programs include a control program PG (see FIG. 8) and a learned model LM (see FIG. 8).

[0041] The storage unit 103 has a main storage device. The main storage device is, for example, a semiconductor memory. The storage unit 103 may further have an auxiliary storage device. The auxiliary storage device includes, for example, at least one of a semiconductor memory and a hard disk drive. The storage unit 103 may include a removable medium. The control unit 102 controls the operation of each unit of the substrate processing apparatus 100 based on the computer programs and data stored in the storage unit 103.

[0042] Next, the substrate processing apparatus 100 of this embodiment will be further described with reference to Figures 1 and 2. Figure 2 is a schematic diagram of a processing unit 1 included in the substrate processing apparatus 100. In detail, Figure 2 is a schematic cross-sectional view of the processing unit 1.

[0043] As shown in Fig. 2, the processing unit 1 includes a chamber 2, a spin chuck 3, a spin motor section 4, a first nozzle 51, a nozzle movement mechanism 6, a second nozzle 71, a thickness measurement section 8, a probe movement mechanism 9, and a plurality of guards 10 (two guards 10 in Fig. 2). The substrate processing apparatus 100 also includes a chemical liquid supply section 5 and a rinsing liquid supply section 7. The chemical liquid supply section 5 has a first supply pipe 52, and the rinsing liquid supply section 7 has a second supply pipe 72. The control device 101 (control section 102) controls the spin chuck 3, the spin motor section 4, the chemical liquid supply section 5, the nozzle movement mechanism 6, the rinsing liquid supply section 7, the thickness measurement section 8, and the probe movement mechanism 9.

[0044] The chamber 2 has a substantially box-like shape and accommodates the substrate W, the spin chuck 3, the spin motor unit 4, the first nozzle 51, a portion of the first supply pipe 52, the nozzle movement mechanism 6, the second nozzle 71, a portion of the second supply pipe 72, a thickness measurement unit 8, a probe movement mechanism 9, and a plurality of guards 10.

[0045] The spin chuck 3 holds the substrate W horizontally. The spin chuck 3 is an example of a substrate holding section. Specifically, the spin chuck 3 has a spin base 31 and multiple chuck members 33. The spin base 31 is substantially disk-shaped and supports the multiple chuck members 33 in a horizontal position. The multiple chuck members 33 are disposed on the periphery of the spin base 31. The multiple chuck members 33 support the periphery of the substrate W. The multiple chuck members 33 hold the substrate W in a horizontal position.

[0046] The spin motor unit 4 rotates the substrate W and the spin chuck 3 together about a first rotation axis AX1. The first rotation axis AX1 extends in the up-down direction. In this embodiment, the first rotation axis AX1 extends in a substantially vertical direction. The first rotation axis AX1 is an example of a central axis, and the spin motor unit 4 is an example of a substrate rotation unit.

[0047] More specifically, the spin motor unit 4 rotates the spin base 31 about the first rotation axis AX1. Therefore, the spin base 31 rotates about the first rotation axis AX1. As a result, the substrate W held by the spin chuck 3 rotates about the first rotation axis AX1.

[0048] Specifically, the spin motor unit 4 has a motor body 41, a shaft 43, and an encoder 45. The shaft 43 is coupled to the spin base 31. The motor body 41 rotates the shaft 43. As a result, the spin base 31 rotates.

[0049] The encoder 45 detects the rotation speed of the substrate W, and outputs a signal indicating the rotation speed of the substrate W (hereinafter, referred to as a "rotation speed signal") to the control device 101 (controller 102). Specifically, the encoder 45 detects the rotation speed of the motor main body 41. The control device 101 (controller 102) controls the spin motor unit 4 based on the rotation speed signal.

[0050] The first nozzle 51 ejects a chemical liquid (etchant) toward the substrate W. In this embodiment, the substrate processing includes an ejection process in which the chemical liquid is ejected from the first nozzle 51 toward the substrate W. In detail, the first nozzle 51 ejects the chemical liquid from above the substrate W toward the rotating substrate W. The first nozzle 51 is an example of a processing liquid supply unit. For example, the first nozzle 51 ejects an etching liquid toward the substrate W after the grinding process to flatten the upper surface of the substrate W after the grinding process.

[0051] More specifically, the chemical liquid is discharged toward a target object TG (see FIG. 3) included in the substrate W. As a result, the substrate W is processed by the chemical liquid. The substrate processing apparatus 100 (processing unit 1) of this embodiment discharges the etching liquid onto the target object TG to adjust the thickness of the target object TG to a target thickness (target thickness value).

[0052] The target object TG is, for example, a substrate body (e.g., a substrate body made of silicon), a substance formed on the surface of the substrate body, or the substrate W. The substance formed on the surface of the substrate body is, for example, a substance made of the same material as the substrate body (e.g., a layer made of silicon), or a substance made of a different material from the substrate body (e.g., a silicon oxide film, a silicon nitride film, or a resist). The "substance" may form a film.

[0053] The chemical liquid supply unit 5 supplies the chemical liquid to the first nozzle 51. More specifically, the chemical liquid is supplied to the first nozzle 51 via a first supply pipe 52. The first supply pipe 52 is a tubular member through which the chemical liquid flows.

[0054] The nozzle movement mechanism 6 moves the first nozzle 51 when performing substrate processing. Specifically, the nozzle movement mechanism 6 moves the first nozzle 51 in a circumferential direction centered on a second rotation axis AX2 along a substantially vertical direction. While moving, the first nozzle 51 ejects the chemical solution toward the substrate W. The first nozzle 51 is sometimes referred to as a scan nozzle.

[0055] Specifically, the nozzle movement mechanism 6 has a nozzle arm 61, a first rotating shaft 63, and a first driving unit 65. The nozzle arm 61 extends along a substantially horizontal direction. The first nozzle 51 is disposed at the tip of the nozzle arm 61. The nozzle arm 61 is coupled to the first rotating shaft 63. The first rotating shaft 63 extends along a substantially vertical direction.

[0056] The first driving unit 65 rotates the first rotating shaft 63 about the second rotation axis AX2, and swings the nozzle arm 61 about the first rotating shaft 63. As a result, the first nozzle 51 moves along a substantially horizontal plane. More specifically, the first nozzle 51 moves around the first rotating shaft 63 in the circumferential direction about the second rotation axis AX2.

[0057] The first driving unit 65 outputs a rotational position signal indicating the rotational position of the first nozzle 51 to the control device 101 (control unit 102). For example, the first driving unit 65 includes a motor body and a Hall element. The Hall element detects the rotational position of the motor body. The control device 101 (control unit 102) controls the first driving unit 65 based on the rotational position signal. The first driving unit 65 includes, for example, a stepping motor. Alternatively, the first driving unit 65 may include a motor and a reducer.

[0058] The second nozzle 71 supplies the rinsing liquid from above the substrate W to the rotating substrate W. The rinsing liquid supply unit 7 supplies the rinsing liquid to the second nozzle 71. More specifically, the rinsing liquid is supplied to the second nozzle 71 via a second supply pipe 72. The second supply pipe 72 is a tubular member through which the rinsing liquid flows. The second nozzle 71 ejects the rinsing liquid in a stationary state. The second nozzle 71 is sometimes referred to as a fixed nozzle. The second nozzle 71 may be a scan nozzle.

[0059] Each of the guards 10 has a substantially cylindrical shape. The guards 10 receive the chemical liquid and the rinsing liquid discharged from the substrate W.

[0060] The thickness measuring unit 8 measures the thickness of the target TG (see FIG. 3) included in the substrate W, and generates a measurement signal indicating the measurement result. The thickness measuring unit 8 measures the thickness of the target TG in a non-contact manner. The measurement signal is input to the control device 101 (control unit 102).

[0061] The thickness measuring unit 8 measures the thickness of the target object TG by, for example, spectroscopic interferometry. Specifically, the thickness measuring unit 8 includes an optical probe 81, a signal line 83, and a thickness measuring device 85.

[0062] The optical probe 81 has a lens. The signal line 83 optically connects the optical probe 81 and the thickness measuring device 85. The signal line 83 includes, for example, an optical fiber. The thickness measuring device 85 has a light source and a light receiving element.

[0063] Light emitted from a light source of the thickness measuring device 85 is emitted to the substrate W via the signal line 83 and the optical probe 81. The light reflected by the substrate W is received by a light receiving element of the thickness measuring device 85 via the optical probe 81 and the signal line 83.

[0064] The thickness measuring device 85 analyzes the light received by the light receiving element to calculate the thickness value of the target object TG. The thickness measuring device 85 generates a measurement signal indicating the calculated thickness value.

[0065] The probe movement mechanism 9 moves the optical probe 81. Specifically, the probe movement mechanism 9 moves the optical probe 81 in a circumferential direction about a third rotation axis AX3 that is aligned in a substantially vertical direction.

[0066] Specifically, the probe movement mechanism 9 has a probe arm 91, a second rotation shaft 93, and a second drive unit 95. The probe arm 91 extends along a substantially horizontal direction. The optical probe 81 is disposed at the tip of the probe arm 91. The probe arm 91 is coupled to the second rotation shaft 93. The second rotation shaft 93 extends along a substantially vertical direction.

[0067] The second driving unit 95 rotates the second rotation shaft 93 about the third rotation axis AX3 to turn the probe arm 91 about the second rotation shaft 93. As a result, the optical probe 81 moves along a substantially horizontal plane. More specifically, the optical probe 81 moves around the second rotation shaft 93 in the circumferential direction about the third rotation axis AX3.

[0068] The second driving unit 95 outputs a rotational position signal indicating the rotational position of the optical probe 81 to the control device 101 (control unit 102). For example, the second driving unit 95 includes a motor body and a Hall element. The Hall element detects the rotational position of the motor body. The control device 101 (control unit 102) controls the second driving unit 95 based on the rotational position signal. The second driving unit 95 includes, for example, a stepping motor. Alternatively, the second driving unit 95 may include a motor and a reducer.

[0069] Next, the probe movement process and the thickness measurement process will be described with reference to Fig. 3(a) and Fig. 3(b). The probe movement process refers to a process of moving the optical probe 81 to a measurement position P. The measurement position P refers to a position where the thickness of the target object TG is measured. Fig. 3(a) is a plan view showing the probe movement process. Fig. 3(b) is a plan view showing the thickness measurement process.

[0070] First, the probe movement process will be described with reference to Fig. 3(a). Before substrate processing is performed, the control unit 102 described with reference to Fig. 1 controls the probe movement mechanism 9 to move the optical probe 81 to the measurement position P. In this embodiment, the probe movement process is performed while the substrate W is rotating. However, the probe movement process may be performed before the substrate processing. For example, the probe movement process may be performed before the substrate W starts to rotate.

[0071] As shown in FIG. 3(a), the probe movement mechanism 9 can move the optical probe 81 along a trajectory TJ1 that is arc-shaped in a plan view. The trajectory TJ1 passes through an edge portion EG of the substrate W and a center portion CT of the substrate W. The edge portion EG indicates the peripheral portion of the substrate W. The center portion CT indicates a portion of the substrate W through which a first rotation axis AX1 passes. The probe movement mechanism 9 moves the optical probe 81 along the trajectory TJ1 to a predetermined measurement position P. The measurement position P is stored in advance in the memory unit 103 described with reference to FIG. 1.

[0072] Next, the thickness measurement process will be described with reference to Fig. 3(b). The thickness measurement process is executed before the substrate processing. Specifically, the control unit 102 described with reference to Fig. 1 causes the thickness measurement unit 8 to measure the thickness of the target object TG before the substrate processing is executed. As shown in Fig. 3(b), the optical probe 81 is placed at the measurement position P during the thickness measurement process. In other words, the position of the optical probe 81 is fixed at the measurement position P during the thickness measurement process.

[0073] The thickness measuring unit 8 measures the thickness of the object TG at a measurement position P (a fixed position). The substrate W rotates while the thickness measurement process is being performed. Therefore, the thickness measuring unit 8 measures the thickness of the object TG along the circumferential direction CD of the substrate W. Therefore, the measurement signal indicates the distribution of the thickness of the object TG in the circumferential direction CD of the substrate W.

[0074] The control unit 102 described with reference to Fig. 1 acquires pre-processing measurement data based on the measurement signal. The pre-processing measurement data indicates the thickness of the target object TG before the substrate processing is performed. More specifically, the pre-processing measurement data indicates the thickness distribution of the target object TG.

[0075] Next, the scanning process of the substrate W by the first nozzle 51 will be described with reference to Fig. 4. Fig. 4 is a plan view showing the scanning process of the substrate W by the first nozzle 51.

[0076] 4, the scanning process is a process in which the first nozzle 51 ejects the chemical solution onto the target object TG while moving so that the landing position of the chemical solution on the surface of the target object TG forms an arc-shaped trajectory TJ2 in a plan view. The trajectory TJ2 passes through the center CT of the substrate W. The scanning process is performed while the substrate W is rotating.

[0077] In this embodiment, the first nozzle 51 ejects the chemical liquid toward the rotating substrate W while moving from a first position X1 to a ninth position X9. Each of the positions X1 to X9 is included in a locus TJ2.

[0078] The first position X1 indicates the ejection start position of the chemical liquid. The movement speed of the first nozzle 51 at the first position X1 is 0 mm / s. Therefore, the first position X1 is the start position of the scan process. In addition, the first position X1 is the movement start position of the first nozzle 51.

[0079] The ninth position X9 indicates a position where the discharge of the chemical liquid is stopped. The movement speed of the first nozzle 51 at the ninth position X9 is 0 mm / s. The ninth position X9 is the end position of the scanning process. In addition, the ninth position X9 is the movement end position of the first nozzle 51.

[0080] During the scanning process, the first nozzle 51 passes through each of the intermediate positions (positions X2 to X8 from the second position X2 to the eighth position X8) between the first position X1 and the ninth position X9. Each of the intermediate positions divides the movement section along which the first nozzle 51 moves during the scanning process into a plurality of sections.

[0081] Next, the scan speed information will be described with reference to Fig. 5. The scan speed information indicates a setting value of the movement speed of the first nozzle 51 during the scan process. Hereinafter, the movement speed of the first nozzle 51 during the scan process may be referred to as the "scan speed". Fig. 5 is a diagram showing the scan speed information. In detail, Fig. 5 shows the relationship between each of the positions X1 to X9 described with reference to Fig. 4 and the setting value of the scan speed.

[0082] 5, the upper column shows positions X1 to X9 included in the movement section of the first nozzle 51. In detail, the upper column shows the start position of the movement section of the first nozzle 51 (the start position of the movement of the first nozzle 51), the end position of the movement section of the first nozzle 51 (the end position of the movement of the first nozzle 51), and multiple intermediate positions (multiple positions through which the first nozzle 51 passes) between the start position and the end position of the movement section of the first nozzle 51. Each of the positions X1 to X9 is defined by a radial position on the substrate W.

[0083] In Fig. 5, the lower column indicates the set value of the scan speed. As shown in Fig. 5, the scan speed information indicates the set value of the scan speed for each of positions X1 to X9 included in the movement section of the first nozzle 51. Hereinafter, each of positions X1 to X9 included in the movement section of the first nozzle 51 may be referred to as a "speed setting position." In this embodiment, the scan speed information indicates nine speed setting positions.

[0084] The speed setting positions shown in Fig. 5 correspond to the positions X1 to X9 described with reference to Fig. 4. As described with reference to Fig. 4, the scan speed set at the start position (first position X1) of the movement section of the first nozzle 51 is 0 [mm / s], and the scan speed set at the end position (ninth position X9) of the movement section of the first nozzle 51 is 0 [mm / s].

[0085] The control unit 102 described with reference to Fig. 1 controls the first driving unit 65 of the nozzle movement mechanism 6 described with reference to Fig. 2 based on the scan speed information. As a result, the first nozzle 51 moves along the trajectory TJ2 described with reference to Fig. 4 so that the scan speed at each speed setting position becomes the scan speed defined by the scan speed information.

[0086] Next, the movement speed (scanning speed) of the first nozzle 51 during the scanning process will be described with reference to Fig. 6. Fig. 6 is a graph showing an example of the movement speed of the first nozzle 51.

[0087] In Fig. 6, the vertical axis indicates the scan speed [mm / s], and the horizontal axis indicates the radial position [mm] of the substrate W. As shown in Fig. 6, the scan speed at the start of the scan process is 0 [mm / s]. Also, the scan speed at the end of the scan process is 0 [mm / s].

[0088] As described with reference to Figures 4 and 5, the scan speed is set for each speed setting position (positions X1 to X9). As a result, as shown in Figure 6, between adjacent speed setting positions, the scan speed changes continuously from the scan speed set for one speed setting position to the scan speed set for the other speed setting position.

[0089] 5, a scan speed Y3 is set for the third position X3, and a scan speed Y4 is set for the fourth position X4. Therefore, the scan speed of the first nozzle 51 changes continuously from the scan speed Y3 to the scan speed Y4 while the first nozzle 51 moves from the third position X3 to the fourth position X4.

[0090] Next, the chemical liquid supplying unit 5 will be described with reference to Fig. 7. Fig. 7 is a schematic diagram of the chemical liquid supplying unit 5. As shown in Fig. 7, the chemical liquid supplying unit 5 further includes a temperature sensor 521, a concentration sensor 522, a valve 523, a mixing valve 524, a flow meter 525, and a heater 526.

[0091] The temperature sensor 521 measures the temperature of the chemical liquid flowing through the first supply pipe 52. Hereinafter, the temperature of the chemical liquid flowing through the first supply pipe 52 will be referred to as the "chemical liquid temperature." The temperature sensor 521 generates a temperature signal indicating the chemical liquid temperature. The temperature signal is input to the control unit 102 described with reference to FIG. 1. The control unit 102 controls the heater 526 based on the temperature signal. The heater 526 heats the chemical liquid flowing through the first supply pipe 52.

[0092] The concentration sensor 522 measures the concentration of an etching component contained in the chemical liquid (etchant) flowing through the first supply pipe 52. Hereinafter, the concentration of the etching component contained in the chemical liquid flowing through the first supply pipe 52 will be referred to as "chemical liquid concentration." The concentration sensor 522 generates a concentration signal indicating the chemical liquid concentration. The concentration signal is input to the control unit 102 described with reference to FIG. 1. The control unit 102 controls the mixing valve 524 based on the concentration signal.

[0093] The valve 523 is disposed on the first supply pipe 52. The valve 523 switches between supplying and stopping the supply of the chemical liquid to the first nozzle 51. In detail, when the valve 523 is opened, the chemical liquid is discharged from the first nozzle 51 toward the substrate W. On the other hand, when the valve 523 is closed, the discharge of the chemical liquid stops.

[0094] Furthermore, the valve 523 controls the flow rate of the chemical liquid flowing downstream of the valve 523 in the first supply pipe 52. In detail, the flow rate of the chemical liquid flowing downstream of the valve 523 is adjusted according to the opening degree of the valve 523. Therefore, the discharge flow rate of the chemical liquid is adjusted according to the opening degree of the valve 523. The valve 523 is, for example, a motor valve.

[0095] The mixing valve 524 is disposed in the first supply pipe 52. When the mixing valve 524 is opened, pure water flows into the first supply pipe 52, and the chemical solution is diluted. Therefore, the concentration of the chemical solution decreases.

[0096] The flow meter 525 measures the discharge flow rate of the chemical liquid. Specifically, the flow meter 525 measures the flow rate of the chemical liquid flowing through the first supply pipe 52. The flow meter 525 generates a discharge flow rate signal indicating the discharge flow rate of the chemical liquid. The discharge flow rate signal is input to the control unit 102 described with reference to Fig. 1. The control unit 102 controls the valve 523 based on the discharge flow rate signal.

[0097] Next, the control device 101 will be described with reference to Fig. 8. Fig. 8 is a block diagram of the control device 101 included in the substrate processing apparatus 100. As shown in Fig. 8, the control device 101 further includes a display unit 104 and an input unit 105.

[0098] The display unit 104 displays various information. For example, the display unit 104 displays various error screens and various setting screens (input screens). The display unit 104 has, for example, a liquid crystal display or an organic EL (electroluminescence) display.

[0099] The input unit 105 receives input from an operator and outputs various information to the control unit 102. The input unit 105 includes input devices such as a keyboard, a pointing device, and a touch panel. The touch panel may be disposed on the display surface of the display unit 104, for example, to configure a graphical user interface together with the display unit 104. When a setting screen is displayed on the display unit 104, the operator can operate the input unit 105 to input various processing conditions or various setting values.

[0100] Next, the control unit 102 and the storage unit 103 will be further described with reference to Fig. 8. As shown in Fig. 8, the storage unit 103 stores a control program PG and recipe data RP. Various processing conditions for performing substrate processing are set in the recipe data RP. For example, the recipe data RP sets, as processing conditions, each set value of the movement speed of the first nozzle 51 (scanning speed at each speed setting position), the rotation speed of the substrate W, the chemical temperature, the chemical concentration, and the discharge flow rate of the chemical. The control unit 102 controls each unit of the substrate processing apparatus 100 based on the control program PG and the recipe data RP.

[0101] For example, the control unit 102 controls the first driving unit 65 described with reference to FIG. 2 so that the scan speed at each speed setting position matches the set value. Similarly, the control unit 102 controls the motor main body 41 described with reference to FIG. 2 so that the rotation speed of the substrate W matches the set value. The control unit 102 controls the heater 526 described with reference to FIG. 7 so that the chemical temperature matches the set value. The control unit 102 controls the mixing valve 524 described with reference to FIG. 7 so that the chemical concentration matches the set value. The control unit 102 controls the valve 523 described with reference to FIG. 7 so that the discharge flow rate of the chemical matches the set value.

[0102] The memory unit 103 further stores the trained model LM. The trained model LM outputs processing conditions during substrate processing based on input data. Before starting substrate processing, the control unit 102 sets the processing conditions output from the trained model LM in the recipe data RP.

[0103] In this embodiment, the trained model LM outputs scan speed information based on input data. Before starting substrate processing, the control unit 102 inputs the input data to the trained model LM and causes the trained model LM to output scan speed information. Then, based on the scan speed information output from the trained model LM, the moving speed of the first nozzle 51 (scan speed at each speed setting position) is set in the recipe data RP. The scan speed information output from the trained model LM is a target variable.

[0104] In detail, the control unit 102 generates input data indicating a target value of the processing amount by the substrate processing. The target value of the processing amount input to the trained model LM is an explanatory variable. In this embodiment, the target value of the processing amount is a target value of the etching amount. The target value of the etching amount indicates the distribution of the etching amount required to make the thickness of the object TG the target thickness. The control unit 102 generates the input data based on the pre-processing measurement data (the distribution of the thickness of the object TG before the substrate processing is performed) and the target data indicating the target value of the thickness of the object TG (the target thickness). Specifically, the input data indicates the difference between the distribution of the thickness of the object TG before the substrate processing is performed and the target thickness.

[0105] Hereinafter, the thickness of the target object TG before the substrate processing is performed may be referred to as a “pre-processing thickness.” Similarly, the thickness of the target object TG after the substrate processing is performed may be referred to as a “post-processing thickness.”

[0106] The storage unit 103 further stores reference data RE. The reference data RE is acquired based on a plurality of pieces of learning data LD used to construct the trained model LM. The trained model LM is constructed by machine learning the plurality of pieces of learning data LD.

[0107] In addition, the machine learning algorithm for constructing the trained model LM is not particularly limited as long as it is supervised learning, and may be, for example, a decision tree, a nearest neighbor method, a naive Bayes classifier, a support vector machine, or a neural network. Therefore, the trained model LM includes a decision tree, a nearest neighbor method, a naive Bayes classifier, a support vector machine, or a neural network. The backpropagation method may be used for machine learning.

[0108] For example, the neural network includes an input layer, one or more intermediate layers, and an output layer. Specifically, the neural network is a deep neural network (DNN), a recurrent neural network (RNN), or a convolutional neural network (CNN), and performs deep learning. For example, the deep neural network includes an input layer, multiple intermediate layers, and an output layer.

[0109] The control unit 102 executes a determination process based on the input data and the reference data RE. The determination process is a process for determining whether or not to execute a substrate process. Specifically, the control unit 102 compares the input data with the reference data RE to determine whether or not to execute a substrate process.

[0110] When the control unit 102 determines to perform substrate processing, it inputs the input data to the trained model LM. Then, based on the scan speed information output from the trained model LM, the control unit 102 sets the moving speed of the first nozzle 51 (the scan speed at each speed setting position) in the recipe data RP, and then controls the processing unit 1 to cause the processing unit 1 to perform the substrate processing.

[0111] On the other hand, when the control unit 102 determines not to execute the substrate processing, it stops the execution of the substrate processing by the processing unit 1 without inputting the input data into the learned model LM. Then, it causes the display unit 104 to display an error screen. The error screen is a screen that notifies the operator that an error has occurred.

[0112] Next, the reference data RE will be described with reference to Fig. 9. Fig. 9 is a diagram showing a learning data set LDS used to construct the reference data RE. As shown in Fig. 9, the learning data set LDS includes a plurality of learning data LD.

[0113] Each of the learning data LD includes scan speed information and processing amount information. The scan speed information and the processing amount information are associated with each other. The learning data LD is generated by performing substrate processing on a learning target substrate. Hereinafter, the learning target substrate may be referred to as a "learning target substrate."

[0114] The scan speed information indicates the scan speed at each speed setting position set when performing substrate processing on the learning target substrate. The processing amount information indicates the processing amount due to the substrate processing during learning. Specifically, the processing amount indicates the difference between the pre-processing thickness and the post-processing thickness of the learning target substrate. In this embodiment, the processing amount is the etching amount. More specifically, the processing amount information indicates the difference between the distribution of the pre-processing thickness and the distribution of the post-processing thickness of the learning target substrate. Therefore, the processing amount information indicates the distribution of the etching amount due to the etching processing during learning.

[0115] The reference data RE is acquired based on the processing amount information of each learning data LD. For example, the reference data RE may be an average value of the processing amounts included in the multiple learning data LD. Alternatively, the reference data RE may be selected from the processing amounts included in the multiple learning data LD. Alternatively, the reference data RE may be created by an operator based on the processing amounts included in the multiple learning data LD.

[0116] 1 to 9, the control unit 102 compares the reference data (reference data RE) acquired based on the processing amount by the substrate processing during learning with the target value of the processing amount (input data) to determine whether or not to execute the substrate processing. Therefore, the reliability of the processing conditions output from the trained model LM can be guaranteed.

[0117] For example, when the target value of the processing amount (input data) corresponds to a processing amount not acquired at the time of learning, there is a possibility that non-optimal processing conditions (unexpected processing conditions) are output from the learned model LM. In contrast, according to the present embodiment, the control unit 102 can determine not to execute the substrate processing when the target value of the processing amount (input data) corresponds to a processing amount not acquired at the time of learning. Therefore, it is possible to prevent non-optimal processing conditions (unexpected processing conditions) from being output from the learned model LM. Furthermore, the control unit 102 can determine to execute the substrate processing when the target value of the processing amount (input data) corresponds to a processing amount acquired at the time of learning. In this case, there is a low possibility that non-optimal processing conditions (unexpected processing conditions) are output from the learned model LM. Therefore, it is possible to guarantee the reliability of the processing conditions output from the learned model LM.

[0118] In addition, the learning data LD may further include information indicating that the processing amount (etching amount) does not match the target processing amount (target etching amount), or information indicating that the processing amount is not within the acceptable range.

[0119] Next, the determination process executed by the control unit 102 will be further described with reference to Fig. 8. As shown in Fig. 8, the storage unit 103 further stores a threshold value TH. The threshold value TH is a threshold value for a comparison result between the input data and the reference data RE. The control unit 102 compares the input data with the reference data RE to obtain the comparison result, and determines whether or not to execute substrate processing based on the comparison result and the threshold value TH.

[0120] In this embodiment, the threshold value TH indicates a threshold value for the Euclidean distance between the reference data RE and the input data. The control unit 102 calculates the Euclidean distance between the reference data RE and the input data. Then, by comparing the calculated Euclidean distance with the threshold value TH, if the Euclidean distance is greater than the threshold value TH, it decides not to perform the substrate processing, and if the Euclidean distance is equal to or less than the threshold value TH, it decides to perform the substrate processing. Hereinafter, the Euclidean distance between the reference data RE and the input data may be referred to as a "first Euclidean distance."

[0121] More specifically, the reference data RE is normalized. The control unit 102 normalizes the input data and calculates the Euclidean distance (first Euclidean distance) between the reference data RE and the normalized input data. The normalized reference data RE can be obtained by normalizing the processing amount of each learning data LD by its average value or minimum value. Similarly, the control unit 102 obtains the average value or minimum value of the input data and normalizes the input data by the average value or minimum value.

[0122] According to this embodiment, it is possible to determine whether or not to perform substrate processing using the threshold value TH, which makes it easy to determine whether or not to perform substrate processing.

[0123] Next, the determination process executed by the control unit 102 will be further described with reference to Fig. 8 and Fig. 10. Fig. 10 is a diagram showing the relationship between the variation in the amount of processing contained in a plurality of pieces of learning data LD and the threshold value TH.

[0124] 10, the horizontal axis represents Euclidean distance, the vertical axis represents the number of pieces of learning data LD corresponding to each Euclidean distance included on the horizontal axis, and the graph GR represents the variation in the amount of processing included in the multiple pieces of learning data LD.

[0125] The graph GR is obtained by normalizing the processing amount of each learning data LD and calculating the Euclidean distance between the normalized reference data RE and the normalized processing amount of each learning data LD. Hereinafter, the Euclidean distance between the processing amount of the learning data LD and the reference data RE may be referred to as the "second Euclidean distance."

[0126] In this embodiment, the threshold value TH includes a first threshold value TH1 and a second threshold value TH2. The control unit 102 compares the input data with the reference data RE to obtain a comparison result, and selects one of the multiple decision items based on the comparison result and the threshold values ​​TH (the first threshold value TH1 and the second threshold value TH2).

[0127] The multiple decision items include a first decision item to a third decision item. The first decision item is an item for deciding not to perform substrate processing. The second decision item is an item for deciding to perform substrate processing based on the processing conditions output from the trained model LM, and to store flag information indicating that substrate processing has been performed in the storage unit 103. The third decision item is an item for deciding to perform substrate processing based on the processing conditions output from the trained model LM. Note that in the third decision item, flag information indicating that substrate processing has been performed is not added.

[0128] The value A of the first threshold TH1 is set to a value greater than the maximum value of the second Euclidean distance. That is, the first threshold TH1 corresponds to a processing amount not included in the processing amount of the multiple learning data LD. In other words, the first threshold TH1 corresponds to a processing amount not acquired during learning. Therefore, if the input data is input to the trained model LM when the first Euclidean distance is greater than the first threshold TH1, there is a high possibility that an unexpected processing condition will be output from the trained model LM.

[0129] In this embodiment, when the first Euclidean distance is greater than the first threshold TH1, the control unit 102 selects the first decision item and decides not to perform the substrate processing. Therefore, when there is a possibility that the processing conditions output from the trained model LM are not optimal, input data is not input to the trained model LM. This makes it possible to prevent processing conditions whose reliability cannot be guaranteed from being output from the trained model LM. Therefore, it is possible to guarantee the reliability of the processing conditions output from the trained model LM.

[0130] The value B of the second threshold TH2 is set to a value in an area where the number of learning data LD is small. The area where the number of learning data LD is small indicates an area where the amount of processing not acquired during learning is large. When the amount of processing included in the area where the amount of processing not acquired during learning is large is input data, there is a slight possibility that optimal processing conditions are not output from the trained model LM. Therefore, when the first Euclidean distance is equal to or smaller than the first threshold TH1 and is larger than the second threshold TH2, if the input data is input to the trained model LM, there is a slight possibility that non-optimal processing conditions (unexpected processing conditions) are output from the trained model LM. In other words, the reliability of the processing conditions output from the trained model LM is slightly low.

[0131] When the first Euclidean distance is equal to or smaller than the first threshold TH1 and larger than the second threshold TH2, the control unit 102 selects the second decision item to execute substrate processing based on the processing conditions output from the learned model LM, and determines to store flag information indicating that the substrate processing has been executed in the storage unit 103. Therefore, when the thickness or the state of the upper surface of the target object TG included in the substrate W after the substrate processing differs from the expected thickness or state, the operator can verify the cause.

[0132] When the Euclidean distance is equal to or smaller than the second threshold TH2, the control unit 102 selects the third decision item and determines to perform the substrate processing based on the processing conditions output from the trained model LM. In this case, the flag information indicating that the substrate processing has been performed is not stored in the storage unit 103.

[0133] Next, the setting screen SE will be described with reference to Fig. 8 and Fig. 11. Fig. 11 is a diagram showing the setting screen SE. As shown in Fig. 11, the display unit 104 displays the setting screen SE. The setting screen SE includes a graph display field 111 and a first setting field 112 to a fourth setting field 115.

[0134] The first setting field 112 is an input field in which the worker operates the input unit 105 to input a value A of the first threshold value TH1. The second setting field 113 is an input field in which the worker operates the input unit 105 to input a value B of the second threshold value TH2. Note that Fig. 11 illustrates an example in which the numerical values ​​input in the first setting field 112 and the second setting field 113 are displayed as a percentage. Specifically, the first setting field 112 and the second setting field 113 display a percentage equivalent to a value obtained by converting the Euclidean distance from the reference data RE into a mean square deviation.

[0135] The third setting field 114 is an input field for inputting an operation to be executed by the substrate processing apparatus 100 when the first Euclidean distance is greater than the first threshold value TH1. In other words, the third setting field 114 is an input field for inputting a decision item to be selected by the control unit 102 when the first Euclidean distance is greater than the first threshold value TH1. "Stop processing" shown in FIG. 11 corresponds to the first decision item.

[0136] The fourth setting field 115 is an input field for inputting an operation to be executed by the substrate processing apparatus 100 when the first Euclidean distance is greater than the second threshold TH2 and is equal to or smaller than the first threshold TH1. In other words, the fourth setting field 115 is an input field for inputting a decision item to be selected by the control unit 102 when the first Euclidean distance is greater than the second threshold TH2 and is equal to or smaller than the first threshold TH1. "Continue processing" shown in Fig. 11 corresponds to the second decision item.

[0137] The graph display field 111 displays a graph GR, a horizontal axis, and a vertical axis. In this embodiment, the graph GR indicates the variation in the etching amount. The horizontal axis indicates the Euclidean distance. The vertical axis indicates the number of learning data LD corresponding to each Euclidean distance included in the horizontal axis. In this embodiment, the graph display field 111 further displays a first threshold value TH1 and a second threshold value TH2 inputted in the first setting field 112 and the second setting field 113.

[0138] According to this embodiment, the user can set the values ​​of the thresholds TH (the value A of the first threshold TH1 and the value B of the second threshold TH2) to any value, thereby improving convenience.

[0139] Furthermore, the user can arbitrarily set the operation to be performed by the substrate processing apparatus 100 when the first Euclidean distance exceeds the value of the threshold TH (the value A of the first threshold TH1 or the value B of the second threshold TH2), thereby improving convenience.

[0140] Furthermore, since the graph GR is displayed in the graph display field 111, the user can easily determine the value of the threshold value TH. Similarly, since the graph GR is displayed in the graph display field 111, the user can easily determine the operation to be performed by the substrate processing apparatus 100 when the first Euclidean distance exceeds the value of the threshold value TH (the value A of the first threshold value TH1 or the value B of the second threshold value TH2).

[0141] Furthermore, since the first threshold value TH1 and the second threshold value TH2 input in the first setting field 112 and the second setting field 113 are further displayed together with the graph GR, the user can easily determine the value of the threshold value TH.

[0142] The graph display section 111 may be omitted.

[0143] Next, the process executed by the control device 101 (controller 102) of the substrate processing apparatus 100 will be described with reference to Figs. 1 to 12. Fig. 12 is a flow chart showing the process executed by the control device 101 (controller 102) of the substrate processing apparatus 100. In detail, Fig. 12 shows the process executed by the control unit 102 when performing substrate processing on one substrate W. As shown in Fig. 12, the process executed by the control unit 102 includes steps S1 to S9. Of steps S1 to S9, steps S4 and S5 are included in the data processing method.

[0144] 12 is started when an operator operates the input unit 105 to instruct the start of etching processing of the substrate W. When the operator operates the input unit 105 to instruct the start of etching processing of the substrate W, the control unit 102 controls the indexer robot IR and the center robot CR to load the substrate W into the chamber 2 of the processing unit 1 (step S1). The loaded substrate W is held by the spin chuck 3 (step S2).

[0145] When the spin chuck 3 holds the substrate W, the control unit 102 controls the spin motor unit 4 to start rotating the substrate W (step S3). Specifically, the spin motor unit 4 rotates the substrate W together with the spin chuck 3.

[0146] When the rotation speed of the substrate W becomes stable at a predetermined rotation speed, the control unit 102 controls the thickness measuring unit 8 to measure the thickness (pre-processing thickness) of the target object TG included in the substrate W (step S4).

[0147] Specifically, as described with reference to Fig. 3(a), the control unit 102 controls the probe movement mechanism 9 to move the optical probe 81 to the measurement position P (probe movement process). Then, the control unit 102 causes the thickness measurement unit 8 to measure the thickness of the target object TG (thickness measurement process). The control unit 102 acquires pre-processing measurement data based on a measurement signal output from the thickness gauge 85 of the thickness measurement unit 8. The pre-processing measurement data indicates the distribution of the thickness of the target object TG before the execution of the substrate processing (pre-processing thickness).

[0148] After acquiring the pre-processing measurement data, the control unit 102 executes the determination process described with reference to FIGS. 8 to 11 to determine whether or not to execute substrate processing (step S5).

[0149] When the control unit 102 determines to perform substrate processing (Yes in step S5), it inputs input data to the trained model LM to acquire scan speed information (processing conditions). Then, based on the scan speed information acquired from the trained model LM, it sets the scan speed for each speed setting position in the recipe data RP (step S6).

[0150] The control unit 102 sets the scan speed for each speed setting position in the recipe data RP, and then executes the substrate processing (step S7). Specifically, the control unit 102 executes the scan processing described with reference to Fig. 4. That is, the control unit 102 controls the nozzle moving mechanism 6 to move the first nozzle 51, and controls the chemical liquid supply unit 5 to discharge the chemical liquid from the first nozzle 51 toward the substrate W.

[0151] After the substrate processing is performed, the control unit 102 releases the substrate W held by the spin chuck 3. Then, the control unit 102 controls the center robot CR to unload the substrate W from the chamber 2 of the processing unit 1 (step S8). The control unit 102 also controls the center robot CR and the indexer robot IR to transport the substrate W unloaded from the chamber 2 of the processing unit 1 to one of the multiple load ports LP. As a result, the processing shown in FIG. 12 is completed.

[0152] When the control unit 102 determines not to perform substrate processing (No in step S5), it executes an alarm process (step S9) and ends the process shown in Fig. 12. The alarm process is a process for notifying that an error has occurred in the substrate processing apparatus 100. The control unit 102 notifies that an error has occurred in the substrate processing apparatus 100, for example, by causing the display unit 104 to display an error screen.

[0153] Next, the determination process (step S5) will be described with reference to Figures 1 to 13. Figure 13 is a flow diagram showing the determination process (step S5). As shown in Figure 13, the determination process (step S5) includes steps S51 to S54.

[0154] When the judgment process (step S5) starts, the control unit 102 generates input data based on the pre-processing measurement data (pre-processing thickness) and the target data (target thickness), and obtains the Euclidean distance (first Euclidean distance) between the reference data RE and the input data (step S51).

[0155] When the control unit 102 acquires the first Euclidean distance, the control unit 102 determines whether the first Euclidean distance is greater than the first threshold value TH1 (step S52). When the control unit 102 determines that the first Euclidean distance is greater than the first threshold value TH1 (Yes in step S52), the control unit 102 decides not to perform the substrate processing. As a result, the determination process (step S5) ends, and the process executed by the control unit 102 proceeds to step S9 described with reference to FIG.

[0156] When it is determined that the first Euclidean distance is not greater than the first threshold value TH1 (No in step S52), the control unit 102 determines whether the first Euclidean distance is greater than a second threshold value TH2 (step S53).

[0157] When the control unit 102 determines that the first Euclidean distance is greater than the second threshold value TH2 (Yes in step S53), it stores flag information indicating that the substrate processing has been performed in the storage unit 103. As a result, the determination process (step S5) ends, and the process executed by the control unit 102 proceeds to step S6 described with reference to FIG.

[0158] If the control unit 102 determines that the first Euclidean distance is not greater than the second threshold value TH2 (No in step S53), the determination process (step S5) ends, and the process executed by the control unit 102 proceeds to step S6 described with reference to FIG. 12.

[0159] Next, the condition setting process (step S6) will be described with reference to Figures 1 to 14. Figure 14 is a flow diagram showing the condition setting process (step S6). As shown in Figure 14, the condition setting process (step S6) includes steps S61 to S63.

[0160] When the condition setting process (step S6) starts, the control unit 102 inputs input data to the trained model LM (step S61). As a result, scan speed information (processing conditions) is output from the trained model LM, and the control unit 102 acquires the scan speed information (step S62). The control unit 102 sets the scan speed for each speed setting position in the recipe data RP based on the scan speed information acquired from the trained model LM (step S63). As a result, the condition setting process (step S6) ends, and the process executed by the control unit 102 proceeds to step S7 described with reference to FIG. 12.

[0161] Next, the substrate processing (step S7) will be described with reference to Figures 1 to 15. Figure 15 is a flow diagram showing the substrate processing (step S7). As shown in Figure 15, the substrate processing (step S7) includes steps S71 to S73.

[0162] When the substrate processing (step S7) is started, the control unit 102 controls the chemical liquid supply unit 5 and the nozzle movement mechanism 6 to eject the etching liquid from the first nozzle 51 toward the substrate W while moving the first nozzle 51 (step S71). As a result, the substrate W is etched (etching process).

[0163] When etching of the substrate W is completed, the control unit 102 controls the rinsing liquid supply unit 7 to discharge the rinsing liquid from the second nozzle 71 toward the substrate W (step S72). As a result, the etching liquid is removed from the substrate W. Specifically, the etching liquid is swept outward from the substrate W by the rinsing liquid and discharged around the substrate W. Therefore, the liquid film of the rinsing liquid on the substrate W is replaced with a liquid film of the rinsing liquid.

[0164] After replacing the etching liquid with the rinsing liquid, the control unit 102 controls the spin motor unit 4 to dry the substrate W (step S73). As a result, the substrate processing (step S7) ends, and the process executed by the control unit 102 proceeds to step S8 described with reference to FIG.

[0165] Specifically, the control unit 102 increases the rotation speed of the substrate W to be higher than the rotation speed during the etching process and the rinsing process. As a result, a large centrifugal force is applied to the rinsing liquid on the substrate W, and the rinsing liquid adhering to the substrate W is thrown off around the substrate W. In this manner, the rinsing liquid is removed from the substrate W, and the substrate W is dried. Note that the control unit 102 stops the rotation of the substrate W by the spin motor unit 4, for example, after a predetermined time has elapsed since the substrate W started to rotate at high speed.

[0166] Above, the first embodiment of the present invention has been described with reference to Figs. 1 to 15. According to this embodiment, the reliability of the processing conditions output from the trained model LM can be guaranteed. Specifically, when there is a possibility that the processing conditions output from the trained model LM are not optimal, the control unit 102 does not input input data to the trained model LM. Therefore, it is possible to prevent processing conditions whose reliability cannot be guaranteed from being output from the trained model LM. Furthermore, according to this embodiment, it is possible to prevent the substrate W from being processed when there is a possibility that the processing conditions (scan speed information) output from the trained model LM are not optimal.

[0167] In this embodiment, the pre-treatment thickness is measured, but the post-treatment thickness may be measured in addition to the pre-treatment thickness.

[0168] Furthermore, in the present embodiment, the first threshold TH1 and the second threshold TH2 are set, but of the first threshold TH1 and the second threshold TH2, only the first threshold TH1 may be set.

[0169] [Embodiment 2] Next, a second embodiment of the present invention will be described with reference to Figures 1 to 7, 9 to 13, and 15 to 17. However, differences from the first embodiment will be described, and a description of the same matters as in the first embodiment will be omitted. The second embodiment differs from the first embodiment in the condition setting process (step S6).

[0170] First, the control device 101 provided in the substrate processing apparatus 100 of this embodiment will be described with reference to Fig. 16. Fig. 16 is a block diagram of the control device 101 provided in the substrate processing apparatus 100 of this embodiment.

[0171] 16, the storage unit 103 further stores default scan speed information SC. The default scan speed information SC is an example of a default condition. The default scan speed information SC indicates scan speed information that is defined in advance.

[0172] The control unit 102 compares the input data with the reference data RE to obtain a comparison result, and selects one of a plurality of decision items based on the comparison result and the thresholds TH (the first threshold TH1 and the second threshold TH2).

[0173] In this embodiment, the multiple decision items further include a fourth decision item. The fourth decision item is an item for deciding to execute substrate processing based on default scan speed information SC and to store information indicating that the substrate processing has been executed in storage unit 103. The operator can input the second decision item or the fourth decision item in fourth setting field 115 on setting screen SE described with reference to FIG.

[0174] The first decision item is an item for deciding not to perform substrate processing. The second decision item is an item for deciding to perform substrate processing based on the processing conditions output from the trained model LM and to store flag information indicating that substrate processing has been performed in the storage unit 103. The third decision item is an item for deciding to perform substrate processing based on the processing conditions output from the trained model LM.

[0175] Next, the condition setting process executed by the control unit 102 will be described with reference to Fig. 17. Fig. 17 is a flow diagram showing the condition setting process (step S6). In detail, Fig. 17 shows the condition setting process when a fourth decision item is input in the fourth setting field 115 of the setting screen SE. The condition setting process (step S6) shown in Fig. 17 further includes steps S611 and S612 in addition to steps S61 to S63 shown in Fig. 14.

[0176] When the condition setting process (step S6) is started, the control unit 102 determines whether or not flag information indicating that the substrate processing has been performed is stored in the storage unit 103 (step S611). The flag information indicating that the substrate processing has been performed is stored in the storage unit 103 in step S54 shown in FIG.

[0177] When the control unit 102 determines that the flag information is stored in the storage unit 103 (Yes in step S611), the control unit 102 sets the scan speed for each speed setting position in the recipe data RP based on the default scan speed information SC (step S612). As a result, the condition setting process (step S6) ends, and the process executed by the control unit 102 proceeds to step S7 described with reference to FIG.

[0178] If the control unit 102 determines that the flag information is not stored in the storage unit 103 (No in step S611), the control unit 102 executes the processes of steps S61 to S63 described with reference to Fig. 14. As a result, the condition setting process (step S6) ends, and the process executed by the control unit 102 proceeds to step S7 described with reference to Fig. 12.

[0179] The second embodiment of the present invention has been described above with reference to Figures 1 to 7, 9 to 13, and 15 to 17. According to this embodiment, similar to the first embodiment, it is possible to guarantee the reliability of the processing conditions output from the trained model LM.

[0180] Also, as already described, when the input data is input to the learned model LM when the first Euclidean distance is equal to or less than the first threshold TH1 and greater than the second threshold TH2, there is a slight possibility that unexpected scan speed information is output from the learned model LM. In contrast, according to the present embodiment, when the first Euclidean distance is equal to or less than the first threshold TH1 and greater than the second threshold TH2, the scan speed of each speed setting position can be set in the recipe data RP based on the default scan speed information SC. Therefore, when the first Euclidean distance is equal to or less than the first threshold TH1 and greater than the second threshold TH2, it is possible to prevent an unexpected scan speed from being set in the recipe data RP.

[0181] [Embodiment 3] Next, a third embodiment of the present invention will be described with reference to Figures 1 to 11 and 13 to 19. However, differences from the first and second embodiments will be described, and a description of the same aspects as the first and second embodiments will be omitted. The third embodiment differs from the first and second embodiments in that the control unit 102 generates additional learning data.

[0182] 18 is a flow chart showing a process executed by the control device 101 (controller 102) of the substrate processing apparatus 100 according to this embodiment. The process shown in FIG. 18 further includes step S10 and step S20 in addition to step S1 to step S9 shown in FIG.

[0183] In this embodiment, when the drying process (step S73 in FIG. 15) is completed, the rotation of the substrate W by the spin motor unit 4 is not stopped. For example, after the drying process, the control unit 102 reduces the rotation speed of the substrate W to the rotation speed during the etching process and the rinsing process.

[0184] After performing the substrate processing (step S7), the control unit 102 determines whether or not flag information indicating that the substrate processing has been performed is stored in the storage unit 103 (step S10).

[0185] When the control unit 102 determines that the flag information is stored in the storage unit 103 (Yes in step S10), it generates additional learning data (step S20). After generating the additional learning data, the control unit 102 stops the rotation of the substrate W by the spin motor unit 4. Then, it executes step S8 described with reference to Fig. 12. As a result, the process shown in Fig. 18 ends.

[0186] When it is determined that the flag information is not stored in the storage unit 103 (No in step S10), the control unit 102 stops the rotation of the substrate W by the spin motor unit 4. Then, step S8 described with reference to Fig. 12 is executed. As a result, the process shown in Fig. 18 ends.

[0187] Next, the process of generating additional learning data (step S20) will be described with reference to Fig. 19. Fig. 19 is a flow diagram showing the process of generating additional learning data (step S20). As shown in Fig. 19, the process of generating additional learning data (step S20) includes steps S21 to S23.

[0188] When the control unit 102 starts the process of generating additional learning data (step S20), it controls the thickness measurement unit 8 to measure the thickness (post-processing thickness) of the target object TG included in the substrate W (step S21).

[0189] Specifically, as described with reference to Fig. 3(a), the control unit 102 controls the probe movement mechanism 9 to move the optical probe 81 to the measurement position P (probe movement process). Then, the control unit 102 causes the thickness measurement unit 8 to measure the thickness of the target object TG (thickness measurement process). The control unit 102 acquires post-processing measurement data based on a measurement signal output from the thickness gauge 85 of the thickness measurement unit 8. The post-processing measurement data indicates the distribution of the thickness of the target object TG after the substrate processing is performed (post-processing thickness).

[0190] After acquiring the post-processing measurement data, the control unit 102 acquires a processing amount based on the pre-processing measurement data and the post-processing measurement data (step S22). Specifically, the processing amount indicates a difference between the pre-processing thickness and the post-processing thickness.

[0191] When the control unit 102 acquires the processing amount, it generates additional learning data (step S23). The additional learning data includes scan speed information and processing amount information, similar to the learning data LD described with reference to Fig. 9. The control unit 102 generates the additional learning data based on the scan speed and the processing amount at each speed setting position set in the recipe data RP.

[0192] After generating the additional learning data, the control unit 102 stops the rotation of the substrate W by the spin motor unit 4. As a result, the process of generating additional learning data (step S20) ends, and the process executed by the control unit 102 proceeds to step S8 shown in FIG.

[0193] The third embodiment of the present invention has been described above with reference to FIGS. 1 to 11 and 13 to 19. According to this embodiment, when the first Euclidean distance is equal to or less than the first threshold TH1 and greater than the second threshold TH2, additional learning data can be generated. As already described, the value B of the second threshold TH2 is set to a value in an area where the number of learning data LD is small. Therefore, when the first Euclidean distance is equal to or less than the first threshold TH1 and greater than the second threshold TH2, additional learning data is generated, so that the number of learning data LD in an area where the number of learning data LD is small can be increased during additional learning. As a result, the prediction accuracy of the trained model LM is improved.

[0194] In this embodiment, the substrates W whose first Euclidean distance is equal to or smaller than the first threshold value TH1 and larger than the second threshold value TH2 are subject to additional learning, but all substrates W that have undergone substrate processing may be subject to additional learning. In this case, all substrates W whose first Euclidean distance is equal to or smaller than the first threshold value TH1 are subject to additional learning.

[0195] The additional learning data may further include information indicating that the processing amount (etching amount) does not match the target processing amount (target etching amount) or that the processing amount is not within the allowable range. In this case, the control unit 102 determines whether the processing amount (etching amount) matches the input data (target etching amount). Alternatively, the control unit 102 determines whether the processing amount is within the allowable range.

[0196] [Embodiment 4] Next, a fourth embodiment of the present invention will be described with reference to Figures 1 to 20. However, differences from the first to third embodiments will be described, and a description of the same matters as the first to third embodiments will be omitted. The fourth embodiment differs from the first to third embodiments in that the thickness of the target object TG is measured by an external device of the substrate processing apparatus 100.

[0197] Fig. 20 is a diagram showing a substrate processing system 1000 of the present embodiment. As shown in Fig. 20, the substrate processing system 1000 of the present embodiment includes a substrate processing apparatus 100 and a film thickness measuring apparatus 200. The film thickness measuring apparatus 200 is an example of a thickness measuring apparatus.

[0198] Before the substrate processing is performed, the film thickness measuring device 200 measures the thickness of the target TG included in the substrate W. After the film thickness measuring device 200 measures the thickness of the target TG, the substrate processing apparatus 100 performs the substrate processing.

[0199] In detail, the control unit 102 of the substrate processing apparatus 100 acquires the measurement result of the film thickness measuring apparatus 200. For example, the substrate processing apparatus 100 and the film thickness measuring apparatus 200 may be communicably connected to transmit data indicating the measurement result of the film thickness measuring apparatus 200 to the substrate processing apparatus 100. The communication medium may be a communication cable or may be wireless. Alternatively, the control unit 102 of the substrate processing apparatus 100 may acquire the measurement result of the film thickness measuring apparatus 200 using a medium that holds data. For example, the medium that holds data may be an optical disk such as a compact disk or a DVD, or a storage device such as a USB memory.

[0200] The control unit 102 of the substrate processing apparatus 100 acquires pre-processing measurement data (thickness distribution of the target object TG before substrate processing is performed) from the measurement results of the film thickness measurement apparatus 200. Then, as described in the first to third embodiments, input data is generated based on the pre-processing measurement data and target data indicating a target value (target thickness) of the thickness of the target object TG, and the input data is compared with the reference data RE to determine whether or not to perform substrate processing.

[0201] In this embodiment, the substrate processing apparatus 100 may or may not include the thickness measuring unit 8.

[0202] The fourth embodiment of the present invention has been described above with reference to Fig. 1 to Fig. 20. According to this embodiment, similarly to the first to third embodiments, it is possible to guarantee the reliability of the processing conditions output from the trained model LM.

[0203] [Embodiment 5] Next, a fifth embodiment of the present invention will be described with reference to Figures 1 to 19, 21, and 22. However, differences from the first to fourth embodiments will be described, and a description of the same aspects as the first to fourth embodiments will be omitted. The fourth embodiment differs from the first to fourth embodiments in that an external device of the substrate processing apparatus 100 determines whether or not to perform substrate processing.

[0204] Fig. 21 is a diagram showing a substrate processing system 1000 of the present embodiment. As shown in Fig. 21, the substrate processing system 1000 of the present embodiment includes a substrate processing apparatus 100, a film thickness measuring apparatus 200, and an information processing apparatus 300. The film thickness measuring apparatus 200 is an example of a thickness measuring apparatus. The information processing apparatus 300 is an example of a determining apparatus.

[0205] The film thickness measuring device 200 measures the thickness of the target TG included in the substrate W before the substrate processing is performed. The information processing device 300 acquires the measurement results of the film thickness measuring device 200. For example, the information processing device 300 and the film thickness measuring device 200 may be communicably connected to transmit data indicating the measurement results of the film thickness measuring device 200 from the film thickness measuring device 200 to the information processing device 300. The communication medium may be a communication cable or may be wireless. Alternatively, the information processing device 300 may acquire the measurement results of the film thickness measuring device 200 using a medium that holds the data. For example, an optical disk such as a compact disk or a DVD, or a storage device such as a USB memory may be used as the medium that holds the data.

[0206] The information processing device 300 determines whether or not to perform substrate processing based on the measurement result of the film thickness measurement device 200. Specifically, the information processing device 300 executes the determination process described with reference to FIGS. 8 to 11 to determine whether or not to perform substrate processing.

[0207] The substrate processing apparatus 100 acquires the judgment result and the input data to be input to the trained model LM from the information processing apparatus 300. The judgment result indicates whether or not to execute the substrate processing. For example, the information processing apparatus 300 and the substrate processing apparatus 100 may be communicably connected, and data indicating the judgment result and the input data may be transmitted from the information processing apparatus 300 to the substrate processing apparatus 100. The communication medium may be a communication cable or may be wireless. Alternatively, the substrate processing apparatus 100 may acquire the judgment result and the input data using a medium that holds data. For example, an optical disk such as a compact disk or a DVD, or a storage device such as a USB memory may be used as the medium that holds data.

[0208] When the information processing apparatus 300 determines that a substrate processing operation is to be performed, the substrate processing apparatus 100 inputs input data to the trained model LM as described in the first to fourth embodiments. Then, the substrate processing operation is performed based on the processing conditions output from the trained model LM.

[0209] The substrate processing apparatus 100 may obtain the input data only when the information processing apparatus 300 has determined that the substrate processing is to be performed.

[0210] Next, the configuration of the information processing device 300 will be described with reference to Fig. 22. Fig. 22 is a block diagram showing the configuration of the information processing device 300. As shown in Fig. 22, the information processing device 300 includes a processing unit 301, a storage unit 302, a display unit 303, and an input unit 304. The information processing device 300 is, for example, a server or a personal computer (PC). However, the information processing device 300 is not particularly limited as long as it is a device capable of performing arithmetic processing.

[0211] The processing unit 301 includes a processor. The processing unit 301 includes, for example, a CPU or an MPU. Alternatively, the processing unit 301 may include a general-purpose computing device or a dedicated computing device.

[0212] The storage unit 302 stores data and computer programs. The data includes reference data RE. In this embodiment, the data further includes a threshold value TH. The processing unit 301 executes various arithmetic operations based on the computer programs and data stored in the storage unit 302.

[0213] The storage unit 302 includes a main storage device. The main storage device is, for example, a semiconductor memory. The storage unit 302 may further include an auxiliary storage device. The auxiliary storage device includes, for example, at least one of a semiconductor memory and a hard disk drive. The storage unit 302 may include a removable medium.

[0214] The display unit 303 displays various types of information. In this embodiment, the display unit 303 displays the setting screen SE described with reference to Fig. 11. The display unit 303 has, for example, a liquid crystal display or an organic EL display.

[0215] The input unit 304 receives input from an operator and outputs various information to the processing unit 301. The input unit 304 includes input devices such as a keyboard, a pointing device, and a touch panel. The touch panel may be disposed on the display surface of the display unit 303, for example, and may configure a graphical user interface together with the display unit 303.

[0216] When the setting screen SE is displayed on the display unit 303, the operator can operate the input unit 304 to input the values ​​of the first threshold TH1 and the second threshold TH2, respectively, in the first setting field 112 and the second setting field 113 described with reference to Fig. 11. In addition, the operator can operate the input unit 304 to input the operations to be performed by the substrate processing apparatus 100, respectively, in the third setting field 114 and the fourth setting field 115 described with reference to Fig. 11.

[0217] The processing unit 301 is an example of a determination unit. The processing unit 301, like the control unit 102 of the substrate processing apparatus 100 described in the first to fourth embodiments, acquires pre-processing measurement data (thickness distribution of the target object TG before the substrate processing is performed) from the measurement result of the film thickness measurement device 200. Then, based on the pre-processing measurement data and target data indicating a target value (target thickness) of the thickness of the target object TG, generates input data, compares the input data with the reference data RE, and determines whether or not to perform the substrate processing. Specifically, the processing unit 301 compares the input data with the reference data RE to acquire a comparison result, and determines whether or not to perform the substrate processing based on the comparison result and a threshold value TH.

[0218] The fifth embodiment of the present invention has been described above with reference to Fig. 1 to Fig. 19, Fig. 21, and Fig. 22. According to this embodiment, similarly to the first to fourth embodiments, it is possible to guarantee the reliability of the processing conditions output from the trained model LM.

[0219] The embodiments of the present invention have been described above with reference to the drawings (FIGS. 1 to 22). However, the present invention is not limited to the above embodiments, and can be implemented in various aspects without departing from the gist of the present invention. In addition, the components disclosed in the above embodiments can be modified as appropriate. For example, a certain component among all the components shown in one embodiment may be added to a component of another embodiment, or some of all the components shown in one embodiment may be deleted from the embodiment.

[0220] The drawings are mainly schematic illustrations of each component to facilitate understanding of the invention, and the thickness, length, number, spacing, etc. of each component shown in the drawings may differ from the actual ones due to the convenience of creating the drawings. Furthermore, the configurations of each component shown in the above embodiment are merely examples and are not particularly limited, and it goes without saying that various modifications are possible within a range that does not substantially deviate from the effects of the present invention.

[0221] 1 to 22, a clamping chuck that clamps the substrate W has been described as a configuration for holding the substrate W, but the configuration for holding the substrate W is not particularly limited as long as it can hold the substrate W horizontally. For example, the spin chuck 3 may be a vacuum chuck or a Bernoulli chuck.

[0222] 1 to 22, the value of the first threshold TH1 is set by an operator, but the value of the first threshold TH1 may be a preset value. Similarly, in the embodiment described with reference to Figs. 1 to 22, the value of the second threshold TH2 is set by an operator, but the value of the second threshold TH2 may be a preset value.

[0223] 1 to 22, the operation of the substrate processing apparatus 100 when the first Euclidean distance is greater than the first threshold TH1 is set by an operator, but the operation of the substrate processing apparatus 100 when the first Euclidean distance is greater than the first threshold TH1 may be predefined. Similarly, in the embodiment described with reference to Figs. 1 to 22, the operation of the substrate processing apparatus 100 when the first Euclidean distance is greater than the second threshold TH2 and equal to or less than the first threshold TH1 is set by an operator, but the operation of the substrate processing apparatus 100 when the first Euclidean distance is greater than the second threshold TH2 and equal to or less than the first threshold TH1 may be predefined.

[0224] In the embodiment described with reference to FIGS. 1 to 22, two thresholds (the first threshold TH1 and the second threshold TH2) are set as the threshold TH, but three or more thresholds may be set.

[0225] 1 to 22, the reference data RE and the input data are compared to calculate the Euclidean distance (first Euclidean distance), but the reference data RE and the input data may be compared to calculate the similarity between the reference data RE and the input data. Specifically, the similarity between the reference data RE and the input data may be calculated by DTW (Dynamic time warping), or the similarity between the reference data RE and the input data may be calculated by DDTW (Derivative DTW). In this case, instead of the Euclidean distance (second Euclidean distance) between the processing amount of the learning data LD and the reference data RE, the similarity between the processing amount of the learning data LD and the reference data RE is calculated by DTW or DDTW.

[0226] In the embodiment described with reference to Figs. 1 to 22, the display unit 104 displays an error screen to notify the operator of an error, but the notification of an error is not limited to displaying an error screen. For example, the substrate processing apparatus 100 may include a speaker. In this case, the control unit 102 may output an error sound from the speaker. Alternatively, the substrate processing apparatus 100 may include a lamp. In this case, the control unit 102 may notify the operator of an error by blinking the lamp, or may notify the operator of an error by turning on the lamp.

[0227] 1 to 22, the scan speed information indicates the scan speed at each speed setting position, but the scan speed information may indicate only one setting value of the scan speed. In this case, the first nozzle 51 moves at a constant speed from the start position to the end position of the movement section of the first nozzle 51.

[0228] 1 to 22, the movement speed of the first nozzle 51 is set as a processing condition, but the relative movement speed between the first nozzle 51 and the substrate W may be set. When the substrate W rotates, the relative movement speed between the first nozzle 51 and the substrate W indicates the relative movement speed between the surface of the rotating substrate W and the first nozzle 51. For example, the relative movement speed indicates the sum of the velocity component (vector) of the first nozzle 51 and the velocity component (vector) of a portion of the rotating substrate W that faces the first nozzle 51. Here, the velocity component of the substrate W indicates the velocity in the circumferential direction CD.

[0229] Furthermore, in the embodiment described with reference to FIGS. 1 to 22, the first nozzle 51 turns, but the first nozzle 51 may move linearly.

[0230] In the embodiment described with reference to FIGS. 1 to 22, the first nozzle 51 is a scan nozzle, but the first nozzle 51 may be a fixed nozzle. In this case, the processing unit 1 includes a substrate moving mechanism for moving the substrate W, instead of the nozzle moving mechanism 6. The moving speed of the substrate W is set in the recipe data RP, instead of the scan speed. In this case, the trained model LM outputs substrate speed information indicating the moving speed of the substrate W. The control unit 102 controls the substrate moving mechanism based on the substrate speed information. The substrate speed information may indicate the moving speed of the substrate W set for each position (each substrate position) that divides the moving section in which the substrate W moves into a plurality of sections, similar to the scan speed information. The substrate W may rotate or move linearly.

[0231] 1 to 22, only the first nozzle 51 moves during processing of the substrate W, but the first nozzle 51 and the substrate W may also move. In this case, the processing unit 1 further includes a substrate moving mechanism for moving the substrate W, in addition to the nozzle moving mechanism 6. The recipe data RP sets the moving speed of the substrate W together with the scan speed. In this case, the learned model LM outputs substrate speed information that specifies the moving speed of the substrate W, in addition to the scan speed information. The control unit 102 controls the substrate moving mechanism based on the substrate speed information.

[0232] 1 to 22, the optical probe 81 is fixed at the measurement position P (a fixed position) when measuring the thickness of the object TG, but the optical probe 81 may move when measuring the thickness of the object TG. In this case, as shown in Fig. 3(a), the optical probe 81 moves so that the thickness measurement position for the object TG forms an arc-shaped locus TJ1 in a plan view.

[0233] Specifically, when measuring the thickness of the object TG, the optical probe 81 emits light toward the object TG while moving between the center CT and the edge EG of the substrate W in a plan view. As a result, the thickness of the object TG is measured at each measurement position included in the trajectory TJ1. Each measurement position corresponds to each radial position of the substrate W. Therefore, the thickness measurement process measures the distribution of the thickness of the object TG in the radial direction of the substrate W.

[0234] 1 to 22, the trained model LM outputs the scan speed information as the objective variable, but instead of or in addition to the scan speed information, other processing conditions that affect the throughput may be output. For example, the trained model LM may output at least one of the rotation speed of the substrate W, the chemical temperature, the chemical concentration, and the discharge flow rate of the chemical as the other processing condition.

[0235] In the embodiment described with reference to Figs. 1 to 22, the chemical liquid is an etching liquid, but the chemical liquid is not limited to an etching liquid. The chemical liquid may be any liquid that processes the substrate W. For example, the chemical liquid may be a removal liquid that removes the target object TG. By using the removal liquid, it is possible to perform a process for removing a specific film or a process for removing a specific film containing foreign matter. When a resist removal process is performed on the substrate W, the removal liquid is, for example, a sulfuric acid / hydrogen peroxide mixture (SPM). The resist removal process is a process for removing resist from the surface of a semiconductor substrate.

[0236] In the embodiment described with reference to Figs. 1 to 22, the substrate processing includes an etching process, but the substrate processing is not limited to an etching process. For example, the substrate processing may include a film forming process. When the substrate processing apparatus 100 performs a film forming process, the processing liquid is, for example, SPM or ozone water. In this case, an oxide film is formed on the substrate W. Also, the processing amount indicates the amount of film formed.

[0237] Furthermore, in the embodiment described with reference to Figures 1 to 22, the substrate processing apparatus 100 is a single-wafer type apparatus that performs substrate processing for each substrate W, but the substrate processing apparatus 100 may also be a batch type apparatus that performs substrate processing on multiple substrates W simultaneously. [Industrial Applicability]

[0238] The present invention is useful in the field of substrate processing. [Explanation of symbols]

[0239] 6: Nozzle movement mechanism 8: Thickness measurement section 51: First nozzle 100: Substrate processing apparatus 102: Control unit 103: Storage section 104:Display section 111: Graph display area 112: First setting field 113: Second setting field 114: Third setting field 115: 4th setting column 200: Film thickness measuring device 300: Information processing device 301: Processing section 302: Storage section 1000: Substrate processing system GR:Graph LD: Learning data LM: Pre-trained model RE: Reference data SC : Default scan speed information SE: Setting screen TG:Target TH: Threshold TH1: First threshold TH2: Second threshold W: Substrate

Claims

1. A substrate processing apparatus that performs substrate processing, which is processing on a substrate, a thickness measuring unit for measuring a thickness of an object included in the substrate; a control unit that outputs processing conditions from a trained model by inputting input data indicating a target value of a processing amount for the substrate processing to the trained model that outputs processing conditions during execution of the substrate processing; and A storage unit that stores reference data acquired based on a plurality of pieces of learning data used to construct the trained model; Equipped with the learning data indicates a processing amount of the substrate processing during learning, The control unit is before the substrate processing is performed, the thickness measuring unit measures a thickness of the object to obtain pre-processing measurement data indicating a thickness of the object before the substrate processing is performed; generating the input data based on target data indicating a target value of a thickness of the object and the pre-processing measurement data; The substrate processing apparatus compares the input data with the reference data to determine whether or not to execute the substrate processing.

2. the storage unit stores at least one threshold value for a comparison result between the input data and the reference data; The substrate processing apparatus according to claim 1 , wherein the control unit compares the input data with the reference data to obtain the comparison result, and determines whether or not to perform the substrate processing based on the comparison result and the at least one threshold value.

3. The control unit selects one of a plurality of decision items based on the comparison result and the at least one threshold value; The plurality of decision items include: a first decision item for deciding not to perform the substrate processing; A second decision item of deciding to execute the substrate processing based on the processing conditions output from the trained model and to store information indicating that the substrate processing has been executed in the storage unit; a third decision item for deciding to perform the substrate processing based on the processing conditions output from the trained model; The substrate processing apparatus of claim 2 .

4. The at least one threshold value includes a first threshold value and a second threshold value that is smaller than the first threshold value; The control unit is If the comparison result is greater than the first threshold, select the first decision item; If the comparison result is greater than the second threshold and equal to or less than the first threshold, select the second decision item; The substrate processing apparatus according to claim 3 , wherein the third decision item is selected when the comparison result is equal to or smaller than the second threshold value.

5. The storage unit further stores a predefined default condition as the processing condition, The substrate processing apparatus of claim 3 , wherein the plurality of decision items further include a fourth decision item that decides to perform the substrate processing based on the default conditions and to store information indicating that the substrate processing has been performed in the memory unit.

6. The at least one threshold value includes a first threshold value and a second threshold value that is smaller than the first threshold value; The control unit is If the comparison result is greater than the first threshold, select the first decision item; When the comparison result is greater than the second threshold and equal to or less than the first threshold, the second decision item or the fourth decision item is selected; The substrate processing apparatus according to claim 5 , wherein the third decision item is selected when the comparison result is equal to or smaller than the second threshold value.

7. Further comprising a display unit for displaying a setting screen, The substrate processing apparatus according to claim 2 , wherein the setting screen includes a setting field for setting the at least one threshold value.

8. Further comprising a display unit for displaying a setting screen, The substrate processing apparatus according to claim 3 , wherein the setting screen includes a first setting field for setting one of the plurality of decision items for the at least one threshold value.

9. The substrate processing apparatus according to claim 8 , wherein the setting screen further includes a second setting field for setting the at least one threshold value.

10. The substrate processing apparatus according to claim 7 , wherein the setting screen further includes a graph display field for displaying a graph in which the plurality of learning data are quantified.

11. The substrate processing apparatus according to claim 10 , wherein the setting screen displays the at least one threshold value in the graph display field.

12. a nozzle for ejecting a processing liquid toward the substrate; The substrate processing apparatus according to claim 1 , wherein the substrate processing includes a process of discharging the processing liquid from the nozzle toward the substrate.

13. The substrate processing apparatus according to claim 12 , further comprising a nozzle moving mechanism that moves the nozzle when the substrate processing is performed.

14. The substrate processing apparatus according to claim 13 , wherein the processing conditions include a moving speed of the nozzle.

15. The substrate processing apparatus according to claim 12 , wherein the processing liquid includes an etching liquid for etching the object.

16. a thickness measuring device for measuring the thickness of an object included in a substrate; a substrate processing apparatus that performs a substrate processing on the substrate after the thickness of the object is measured by the thickness measuring device; A substrate processing system comprising: The substrate processing apparatus includes: a control unit that outputs processing conditions from a trained model by inputting input data indicating a target value of a processing amount for the substrate processing to the trained model that outputs processing conditions during execution of the substrate processing; and A storage unit that stores reference data acquired based on a plurality of pieces of learning data used to construct the trained model; Equipped with the learning data indicates a processing amount of the substrate processing during learning, The control unit is obtaining pre-processing measurement data indicating a thickness of the object before the substrate processing is performed from a measurement result of the thickness measuring device; generating the input data based on target data indicating a target value of a thickness of the object and the pre-processing measurement data; A substrate processing system that compares the input data with the reference data to determine whether or not to perform the substrate processing.

17. a thickness measuring device for measuring the thickness of an object included in a substrate; a decision device for deciding whether or not to perform a substrate processing, which is a processing for the substrate; a substrate processing apparatus that outputs processing conditions when the substrate processing is performed from a trained model that outputs the processing conditions; and A substrate processing system comprising: The determination device comprises: A storage unit that stores reference data acquired based on a plurality of pieces of learning data used to construct the trained model; a decision unit that decides whether or not to perform the substrate processing; Equipped with the learning data indicates a processing amount of the substrate processing during learning, The determination unit is obtaining pre-processing measurement data indicating a thickness of the object before the substrate processing is performed from a measurement result of the thickness measuring device; generating input data indicating a target value of a processing amount of the substrate processing based on target data indicating a target value of a thickness of the object and the pre-processing measurement data; comparing the input data with the reference data to determine whether to perform the substrate processing; A substrate processing system, comprising: a control unit that, when the decision device decides to perform the substrate processing, inputs the input data to the trained model and outputs the processing conditions from the trained model.

18. measuring a thickness of an object included in a substrate before performing a substrate processing, and acquiring pre-processing measurement data indicating a measurement result of the thickness of the object; generating input data indicating a target value of a processing amount of the substrate processing based on target data indicating a target value of a thickness of the object and the pre-processing measurement data; A step of comparing the input data with reference data acquired based on a plurality of learning data used to construct a trained model, and determining whether or not to execute the substrate processing. Including, the learning data indicates a processing amount of the substrate processing during learning, A data processing method in which the trained model outputs processing conditions at the time of executing the substrate processing when the input data is input.

Images