JP7373340B2 - judgment device - Google Patents

Description

The present invention relates to a determination device, a substrate processing device, a substrate processing system, and a method of manufacturing an article.

In recent years, with the miniaturization and expansion of demand for electronic devices, it is necessary to achieve both miniaturization and productivity of semiconductor elements represented by memories and MPUs.
Therefore, in a substrate processing apparatus that processes substrates used for manufacturing semiconductor devices, high precision is also required for alignment of the substrates.

In substrate alignment, a method is often used in which the position of the substrate is determined by capturing an image of a mark formed on the substrate and performing pattern matching processing on the obtained image data.
Patent Document 1 discloses an exposure apparatus that detects marks with high accuracy by simultaneously extracting the edges of marks and the directions of such edges, and performing pattern matching processing focusing on the edges for each edge direction. .

Japanese Patent Application Publication No. 2000-260699

Conventionally, when alignment of a substrate fails in a substrate processing apparatus, a user determines whether or not the apparatus needs to be maintained by referring to image data and related data related to the image data.
Therefore, in some cases, the processing of the device may be interrupted or it may take time to make a decision, resulting in a reduction in throughput.
SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a determination device that can determine whether a substrate processing apparatus needs to be maintained.

The determination device according to the present invention classifies image data of marks on a substrate captured by a substrate processing device regarding causes of alignment failure using a learning model obtained by machine learning, and based on the classification results. The method is characterized in that it is determined whether or not the substrate processing apparatus needs to be maintained.

According to the present invention, it is possible to provide a determination device that can determine whether a substrate processing apparatus needs to be maintained.

FIG. 1 is a block diagram showing the configuration of a substrate processing system according to a first embodiment. FIG. 1 is a schematic diagram showing the configuration of an exposure apparatus included in the substrate processing system according to the first embodiment. FIG. 3 is a diagram showing a configuration for determining whether maintenance is necessary in the substrate processing system according to the first embodiment. FIG. 3 is a diagram exemplarily showing a screen displayed on a display device in the substrate processing system according to the first embodiment. FIG. 3 is a diagram exemplarily showing a learning data creation screen in the substrate processing system according to the first embodiment. 5 is a flowchart showing a process for creating learning data in the substrate processing system according to the first embodiment. FIG. 3 is a diagram exemplarily showing buttons for displaying a learning data creation screen in the substrate processing system according to the first embodiment. FIG. 7 is a diagram showing a configuration for determining whether maintenance is necessary in a substrate processing system according to a second embodiment.

Hereinafter, the determination device according to the present embodiment will be described in detail with reference to the drawings. Note that the embodiments shown below merely show specific examples of implementation, and the present embodiments are not limited to the following embodiments.
Furthermore, not all combinations of features described in the embodiments described below are essential for solving the problems of this embodiment.
Further, the drawings shown below may be drawn on a different scale from the actual scale in order to make the present embodiment easily understandable.

[First embodiment]
When manufacturing devices such as semiconductor elements, liquid crystal display elements, and thin-film magnetic heads using photolithography technology, exposure involves projecting a pattern on an original such as a reticle onto a substrate such as a wafer using a projection optical system and transferring the pattern. The device is in use.

In exposure apparatuses, as electronic equipment becomes smaller and demand increases, it is necessary to achieve both miniaturization of semiconductor elements such as memories and MPUs and productivity.
Therefore, exposure apparatuses are required to improve their basic performance such as resolution, overlay accuracy, and throughput.

The resolution of an exposure device is inversely proportional to the numerical aperture (NA) of the projection optical system and proportional to the wavelength of the light used for exposure (exposure light), so it is possible to increase the numerical aperture of the projection optical system and shorten the wavelength of the exposure light. It's progressing.
Furthermore, as semiconductor elements become smaller, it is necessary to improve overlay accuracy, and therefore alignment, which adjusts the relative positions of the original and the substrate, also needs to be highly accurate.

Further, as a technique for further improving overlay accuracy, a technique is known that controls variations in semiconductor manufacturing process, changes over time, etc. by feedforward or the like. Specifically, AEC (Advanced Equipment Control), APC (Advanced Process Control), and the like are known as such techniques. Furthermore, a technique is known in which the results measured by an inspection device are learned by machine learning and fed forward to a lithography device or a coating/developing device (coater/developer).

Factors that cause alignment measurement failures in exposure equipment include the mark being located within the measurement field of view but not clearly due to poor substrate processing or scope aberrations, or the mark formation position being significantly shifted. Various reasons can be considered, such as not being within the measurement field of view.
If the position of the mark deviates significantly within the measurement field of view, it may be due to factors originating from the exposure device, such as misalignment when the exposure device receives the substrate, or due to the mark being dependent on the processing process of the substrate in equipment other than the exposure device. Factors other than those originating from the exposure apparatus, such as positional fluctuations, may be considered.
Therefore, when there are multiple possible causes of alignment measurement failure, the countermeasures, including maintenance, will differ depending on each failure factor.
Therefore, when maintaining the exposure apparatus, it is necessary to accurately classify the causes of failure and to make accurate judgments based on the classification.

FIG. 1 is a block diagram showing the configuration of a substrate processing system 50 including a determination device according to the first embodiment.

The substrate processing system 50 includes at least one semiconductor manufacturing line 1.
Each semiconductor manufacturing line 1 includes a plurality of substrate processing apparatuses 10 (semiconductor manufacturing apparatuses) that process substrates, and a host computer 11 (host control apparatus) that controls the operation of the plurality of substrate processing apparatuses 10. .
Examples of the substrate processing apparatus 10 include a lithography apparatus (exposure apparatus, imprint apparatus, charged particle beam drawing apparatus, etc.), a film forming apparatus (CVD apparatus, etc.), a processing apparatus (laser processing apparatus, etc.), and an inspection apparatus (overlay inspection apparatus, etc.). equipment, etc.).
The substrate processing apparatus 10 also includes a coating and developing device (coater/developer) that applies a resist material (adhesive material) to the substrate as a pre-processing of the lithography process, and also performs a developing process as a post-processing of the lithography process. It can be done.

Note that in the exposure apparatus, a latent image corresponding to a pattern of the original is formed on the photoresist on the substrate by exposing the photoresist supplied onto the substrate through an original (reticle, mask).
In an imprint apparatus, a pattern is formed on a substrate by curing the imprint material while an original plate (mold, template) is in contact with the imprint material supplied onto the substrate.
In a charged particle beam drawing apparatus, a latent image is formed on the photoresist on the substrate by drawing a pattern on the photoresist supplied onto the substrate using a charged particle beam.

As shown in FIG. 1, each of the plurality of substrate processing apparatuses 10 provided in each semiconductor manufacturing line 1 is connected to a management apparatus 12 that manages maintenance.
Thereby, the management device 12 can manage each of the plurality of substrate processing apparatuses 10 provided in each semiconductor manufacturing line 1.
The determination device according to this embodiment is provided in one of the substrate processing device 10, the host computer 11, and the management device 12.

In addition, in the substrate processing system 50, the connection between the plurality of substrate processing apparatuses 10 and the host computer 11 and the connection between the plurality of substrate processing apparatuses 10 and the management apparatus 12 can be either wired connection or wireless connection. I do not care.

Next, a specific example in which each substrate processing apparatus 10 is configured as an exposure apparatus in the substrate processing system 50 will be described.

FIG. 2A is a block diagram showing the configuration of the exposure apparatus 10 provided in the substrate processing system 50. Further, FIG. 2(b) is a schematic diagram showing the configuration of a substrate alignment optical system 190 included in the exposure apparatus 10.

The exposure apparatus 10 is a lithography apparatus that is used to manufacture devices such as semiconductor elements, liquid crystal display elements, and thin film magnetic heads as articles, and forms a pattern on a substrate.
Further, the exposure apparatus 10 exposes the substrate using a step-and-scan method or a step-and-repeat method.

As shown in FIG. 2A, the exposure apparatus 10 includes a main controller 100, a light source controller 110, a light source 120, an image processor 130, a stage controller 140, and an interferometer 150.
The exposure apparatus 10 also includes an original alignment optical system 160, an original stage 171, a projection optical system 180, a substrate alignment optical system 190, and a substrate stage 200.

The original stage 171 moves while holding the original 170 illuminated by an illumination optical system (not shown). A pattern to be transferred to the substrate 210 is drawn on the original plate 170.
Projection optical system 180 projects the pattern of original 170 onto substrate 210 . The substrate stage 200 can hold and move the substrate 210.

The original alignment optical system 160 is used to align the original 170. For example, the original alignment optical system 160 includes an image sensor 161 configured of a storage type photoelectric conversion element, and an optical system 162 that guides light from a mark provided on the original 170 to the image sensor 161.
Substrate alignment optical system 190 is used to align substrate 210. In this embodiment, the substrate alignment optical system 190 is an off-axis optical system that detects marks 211 provided on the substrate 210.

The main control section 100 includes a CPU, a memory, and the like, and controls each section of the exposure apparatus 10 to perform an exposure process for exposing the substrate 210 and related processes.
In the substrate processing system 50, the main controller 100 controls the position of the substrate stage 200 based on the position of the mark formed on the original 170 and the position of the mark 211 formed on the substrate 210. In other words, the main control unit 100 performs positioning between the original 170 and the substrate 210, for example, global alignment.

The light source 120 includes a halogen lamp or the like, and illuminates the mark 211 formed on the substrate 210.
The light source control unit 110 controls the illumination intensity of the light from the light source 120, that is, the light for illuminating the mark 211.

The image processing unit 130 processes image signals (detection signals) from the image sensor 161 in the original alignment optical system 160 and the image sensors 191A and 191B in the substrate alignment optical system 190 to obtain mark positions, that is, mark images. .
In the substrate processing system 50, the image processing section 130 and the substrate alignment optical system 190 function as a measuring device that measures the position of the mark 211 formed on the substrate 210.

The interferometer 150 measures the position of the substrate stage 200 by irradiating light onto a mirror 212 provided on the substrate stage 200 and detecting the light reflected by the mirror 212.
The stage control unit 140 moves (drives and controls) the substrate stage 200 to an arbitrary position based on the position of the substrate stage 200 measured by the interferometer 150.

In the exposure apparatus 10 , light (exposure light) from an illumination optical system (not shown) passes through an original 170 held on an original stage 171 and enters a projection optical system 180 .
Since the original 170 and the substrate 210 are arranged in an optically conjugate positional relationship with each other, the pattern of the original 170 is projected onto the substrate 210 held on the substrate stage 200 via the projection optical system 180. The image is formed and transferred.

The substrate alignment optical system 190 functions as a detection unit that detects the mark 211 formed on the substrate 210 and generates a detection signal (an image signal in this embodiment).
As shown in FIG. 2B, the substrate alignment optical system 190 includes imaging elements 191A and 191B, imaging optical systems 192A and 192B, and a half mirror 193. Further, the substrate alignment optical system 190 includes an illumination optical system 194, a polarizing beam splitter 195, a relay lens 196, a λ/4 plate 197, and an objective lens 198.

In the exposure apparatus 10, light from a light source 120 is guided to a substrate alignment optical system 190 via an optical fiber (not shown) or the like.
The light guided to the substrate alignment optical system 190 then enters the polarizing beam splitter 195 via the illumination optical system 194, as shown in FIG. 2(b).
The light reflected by the polarizing beam splitter 195 passes through the relay lens 196, the λ/4 plate 197, and the objective lens 198, and illuminates the mark 211 formed on the substrate 210.

The light reflected by the mark 211 passes through the objective lens 198, the λ/4 plate 197, the relay lens 196, and the polarizing beam splitter 195, and enters the half mirror 193.
Then, the light incident on the half mirror 193 is divided into two beams at an appropriate intensity ratio by the half mirror 193, and then guided to imaging optical systems 192A and 192B with different imaging magnifications.

Imaging optical systems 192A and 192B form images of marks 211 on the imaging surfaces of imaging elements 191A and 191B, respectively.
The imaging elements 191A and 191B each include an imaging surface that images an area including the mark 211, and generate an image signal corresponding to the area imaged on the imaging surface.

The image signals generated by the image sensors 191A and 191B are then read out by the image processing unit 130.
In this embodiment, the image processing unit 130 acquires the positions of the marks 211 on the imaging surfaces of the image sensors 191A and 191B by performing pattern matching processing as image processing on the read image signals.

Pattern matching processing is generally classified into the following two types.
One method is to binarize an image (shaded image), match it with a template prepared in advance, and set the position with the most correlation as the position of the mark 211.
The other method is to obtain the position of the mark 211 by performing a correlation calculation with a template containing gradation information using the gradation image as it is.

Note that the image processing performed by the image processing unit 130 is not limited to pattern matching processing, and may be other processing such as edge detection processing as long as it is possible to obtain position information of the mark 211.

Further, as alignment methods, there are a movement measurement method and an image processing method.
In the moving measurement method, the mark 211 provided on the substrate 210 is irradiated with light (laser) while the substrate stage 200 is moved. Then, the position of the mark 211 is determined by measuring the change in the intensity of the light reflected from the mark 211 and the position of the substrate stage 200 in parallel.
In the image processing method, marks 211 provided on the substrate 210 are irradiated with white light while the substrate stage 200 is kept stationary. Then, the position of the mark 211 is determined by detecting the light reflected from the mark 211 with a storage type photoelectric conversion element and performing image processing.

In addition, alignment optical systems used in such an alignment method include a TTL (through-the-lens) optical system, a TTR (through-the-reticle) optical system, and an off-axis optical system.
The TTL optical system detects marks provided on the substrate via the projection optical system. The TTR optical system simultaneously detects marks provided on the reticle and marks provided on the substrate via the projection optical system. An off-axis optical system is a dedicated optical system that has an optical axis located a predetermined distance away from the optical axis of the projection optical system without going through the projection optical system, and it emits white light from a dedicated light source onto the substrate. Detect the mark provided.
As described above, the substrate alignment optical system 190 provided in the substrate processing system 50 according to this embodiment is an off-axis optical system.

In the exposure apparatus 10, two types of alignment, pre-alignment and fine alignment, are performed using the acquired position information of the mark 211.
Pre-alignment here refers to roughly aligning the substrate 210 so that the amount of positional deviation of the substrate 210 sent to the substrate stage 200 from a substrate transport system (not shown) can be detected and fine alignment can be started. positioning).
Furthermore, fine alignment here refers to measuring the position of the substrate 210 held by the substrate stage 200 with high precision, and precisely aligning the substrate 210 so that the alignment error of the substrate 210 is within an allowable range. (positioning).

In the pre-alignment, as described above, it is necessary to detect the amount of positional deviation of the substrate 210 sent to the substrate stage 200 from the substrate transport system (not shown).
Therefore, the substrate alignment optical system 190 that detects the mark 211 has a wide detection range (field of view) relative to the size of the mark 211.

To find the position (XY coordinates) of the mark 211 from such a wide detection range, pattern matching (template matching) processing as described above is often used.
In the pattern matching process, it is difficult to detect the mark 211 in a low-contrast image, a noise image, or an image including a mark where an abnormality has occurred when processing the substrate 210.

Measurement of mark 211 may fail due to various reasons. That is, in fine alignment, the mark 211 may not be detected. Further, even if the mark 211 can be detected, the image processing may fail because the position cannot be obtained for some reason.
For example, there may be cases where the mark 211 is not clear due to the influence of the processing process of the substrate 210, or a case where the mark 211 is not clearly visible due to the influence of the aberration of the substrate alignment optical system 190.
It is also conceivable that the position of the mark 211 is shifted from the field of view of the imaging surfaces of the imaging elements 191A and 191B.

If a clear image of the mark 211 can be obtained within the field of view of the imaging surface of the image sensor 191A or 191B, the position of the mark 211 can be accurately measured by image processing.
However, if the contrast of the image is low or if the image is distorted due to the influence of aberrations, the position of the mark 211 may not be measured correctly.

Furthermore, possible causes for the mark 211 to shift from the field of view of the imaging planes of the image sensors 191A and 191B are those caused by the apparatus, such as erroneous measurement during pre-alignment or positional deviation during transport processing before measurement.
In addition, the cause of the mark 211 being shifted from the field of view of the imaging surface of the image sensor 191A or 191B may be due to the processing steps of the substrate 210, such as a change in the transfer position of the mark 211.

If the measurement of the mark 211 fails, the substrate 210 cannot be properly aligned.
If the alignment of the substrate 210 cannot be performed normally, maintenance processing is performed to enable the alignment to be performed normally.

The maintenance processing includes, for example, changing the mark to be used among the plurality of marks 211, expanding the search range for the mark image, changing the imaging conditions, and the like.

If alignment processing fails on the substrate 210, sufficient alignment accuracy cannot be achieved even if exposure processing is performed on the substrate 210 thereafter.
At that time, an error is normally issued, processing of the substrate 210 is stopped, and work is performed to investigate the cause of the failure and resolve it.

On the other hand, if the alignment process is successful on the substrate 210, the exposure process is subsequently performed on the substrate 210, but even if the alignment process is successful, there is a possibility that sufficient alignment accuracy will not be achieved in the exposure process.
One of the causes of such a possibility is that the calculation result for positioning with respect to the substrate 210 becomes inaccurate due to erroneous measurement of the position of the mark 211.

Mismeasurement of the position of the mark 211 can occur, for example, when an incorrect image signal is generated in a mark image obtained by imaging an area including the mark 211 due to adhesion of dust or other conditions at the time of imaging. Occur.
If the position of the mark 211 is incorrectly measured, an incorrect value will be used when calculating the alignment of the substrate 210.
Therefore, as a result of calculation, even if the positioning error of the substrate 210 falls within the allowable range and the alignment process is successful, the alignment accuracy will deteriorate during the exposure process.

FIGS. 3A and 3B are a block diagram and a processing flow diagram, respectively, showing a configuration for determining whether the exposure apparatus 10 requires maintenance.

First, the image processing means 300 provided in the exposure apparatus 10 performs image processing on the substrate 210 to obtain a mark image (image data) (step S401).

If the image processing means 300 determines that the alignment error of the substrate 210 is within the allowable range based on the mark image and that the alignment is successful, the exposure processing means 350 provided in the exposure apparatus 10 executes the exposure process. Ru.
Further, at the same time as the exposure process is executed, by adding related data to the mark image, the image processing means 300 acquires the alignment data 301, and then the alignment data 301 is passed to the image classification means 400 (step S402).
Further, even if the image processing means 300 determines that the alignment error of the substrate 210 is not within the allowable range based on the mark image and the alignment has failed, the alignment data 301 is similarly acquired and then passed to the image classification means 400. (Step S402).

Note that even if the image processing means 300 determines that the alignment is successful, it may actually fail due to erroneous measurement.
In such a case, the alignment data 301 that was determined to be successful due to erroneous measurement may be re-determined to have failed based on the result of overlay measurement of the substrate 210 by an external measuring device (not shown).

Here, the related data is data that includes information related to the acquired mark image. For example, the related data may include information specifying the configuration of the exposure apparatus 10, such as the model, number, hardware configuration, software configuration, and installation line.
Further, the related data may include information specifying the configuration such as lot, substrate 210, original plate 170, recipe, environmental conditions, processing date and time.
Further, the related data may include information specifying configurations such as various offset settings of the exposure apparatus 10 during mark measurement, illumination conditions such as the light amount of the light source 120 that illuminates the mark 211 and the focus amount of the optical system.
Further, the related data may include information specifying the configuration, such as operating conditions during alignment of the exposure apparatus 10, such as the type of the mark 211, and position information of the stage.
Further, the related data may include information specifying the configuration, such as the previous alignment measurement result and the operating state during alignment, such as the pressure with which the substrate stage 200 attracts the substrate 210.

Furthermore, the alignment data 301 delivered to the image classification means 400 is not limited to the above, and may be the result of classifying mark images by an image classification means (not shown) using machine learning or the like.

As described above, the alignment data 301 delivered to the image classification means 400 includes any alignment data 301 for which the image processing means 300 has determined that the alignment was successful or failed; however, the alignment data 301 is not limited to this. I can't do it.
In order to improve throughput, only the alignment data 301 for which alignment has been determined to have failed by the image processing means 300 may be passed to the image classification means 400.

Further, the number of marks 211 measured in step S401 may be one or more, and the number of alignment data 301 transferred to the image classification means 400 may be one or more.
Further, the alignment data 301 may be transferred to the image classification means 400 sequentially every time image processing for one mark 211 is completed. Further, the present invention is not limited to this, and the image processing may be performed all at once after the image processing is completed for all the marks 211 on the substrate 210.

Next, the image classification means 400 classifies the received alignment data 301 into one of a plurality of types related to alignment failure factors (step S403).
Note that the image classification means 400 can be realized by a software program executed in at least one of the main control unit 100, the management device 12, and the host computer 11 of the exposure apparatus 10.

The determination device according to this embodiment uses machine learning as described below as a specific method for classifying the alignment data 301 in the substrate processing system 50.

As a method for making decisions using machine learning, there is supervised learning that creates learning data and performs machine learning.
In supervised learning, it is necessary to create learning data (teacher data) that includes input data and output data that is correct data corresponding to the input data.

In the determination device according to this embodiment, the image classification means 400 uses a learning model obtained by machine learning using a plurality of alignment data 301 in which classification type numbers are input as learning data 305.
Here, machine learning can be performed using, for example, a neural network. A neural network is a model that has a multilayer network structure including an input layer, a middle layer, and an output layer.
Then, a learning model can be obtained by optimizing random variables inside the network using an algorithm such as error backpropagation using learning data indicating the relationship between input data and output data.

Here, an example has been described in which a learning model is obtained using a neural network, but the present invention is not limited to this, and a learning model may be obtained using other models or algorithms such as a support vector machine or a decision tree.
Then, the image classification means 400 inputs the alignment data 301 into the acquired learning model, and outputs classification information 302 including the type number corresponding to the alignment data 301 as output data.

Next, specific creation of learning data in the judgment device according to this embodiment will be described.
First, using the results of the alignment process previously performed on the board 210, the learning data 305 is created by using the alignment data 301 as input data and the classification information 302 corresponding to the classification into each type number as output data. create.

As the classification information 302, for example, type numbers 0 to 5 as shown in Table 1 below can be set.

Note that each type of number as shown in Table 1 may be manually set based on the alignment data 301 when the alignment process failed in the past, the result of the automatic recovery operation of the exposure apparatus 10, etc. Alternatively, each type of number as shown in Table 1 may be automatically provided using machine learning or the like.

As shown in Table 1, in the determination device according to the present embodiment, the classification information 302 includes a cause of failure in alignment processing and a maintenance method for improving the cause for each type number.
In addition, in Table 1, one factor is identified for one type number and one maintenance method is presented, but this is not limited to this; multiple factors are identified for one type number, and multiple A conservation method may also be presented.

Specifically, type number 0 is normal alignment data 301 that does not include a cause of alignment failure, and corresponds to the classification when maintenance is not required.
Further, type number 1 corresponds to the classification when the cause of alignment failure cannot be identified and the maintenance method is unknown.
Further, type number 2 corresponds to a classification in which the cause of alignment failure is a positional shift when the substrate 210 is delivered, and the maintenance method is adjustment of the delivery position of the board 210.
Further, type number 3 corresponds to a classification in which the cause of alignment failure is a setting error in the light intensity of the light source 120 during alignment measurement, and the maintenance method is adjustment of the light intensity of the light source 120 during alignment measurement.
Further, type number 4 corresponds to a classification in which the cause of alignment failure is vibration of the substrate stage 200 during alignment measurement, and the maintenance method is adjustment for the vibration of the substrate stage 200 during alignment measurement.
Further, type number 5 corresponds to a classification in which the cause of alignment failure is deterioration of the light source 120 and the maintenance method is replacement of the light source 120.

For these classifications, the related data added in the alignment data 301 is effectively used.
Note that the above types are just examples, and other classification types may be set.

In order to create classification information 302 as output data, the alignment data 301 as input data is classified into type numbers 0 to 5 by the image classification means 400, and classification information 302 can be obtained.

Note that if the alignment data 301 applies to the above types in multiple ways, it may be classified into the item with the greatest degree of applicability.
Further, the present invention is not limited to this, and if the alignment data 301 applies to the above types in multiple ways, it may be classified by weighting according to the degree to which it applies to each type.

In the manner described above, the learning data 305 can be created by using the alignment data 301 as input data and the classification information 302 corresponding to the classification into each type number as output data.
Then, by learning a plurality of alignment data 301 to which type numbers are attached, inference logic can be created.

Note that in the above description, the classification of the alignment data 301 for creating the learning data 305 is performed by the image classification means 400, but the classification is not limited to this.
For example, in order to create learning data 305 necessary to obtain a learning model, the user can check a plurality of pieces of alignment data 301 and manually input the type number.
Furthermore, in order to increase the accuracy rate of the classification information 302 output from the learning model, it is necessary to create learning data 305 for a large amount of alignment data 301.

As shown in FIG. 3A, the display device 206 displays information necessary to operate the exposure apparatus 10, information regarding the operation of the exposure apparatus 10, and the like.
FIG. 4 is a diagram exemplarily showing a screen 900 displayed on the display device 206.

Further, in the input device 205, the user inputs information necessary for operating the exposure apparatus 10, information necessary for displaying a screen on the display device 206, and the like.
Further, by displaying information necessary for inputting the classification type number on the display device 206, the user can input information on the type number for classification via the input device 205.

Further, a CPU (not shown) executes processing by a display unit 800 that causes information to be displayed on the display device 206 and an input unit 810 that causes information to be input to the input device 205.
Further, a CPU (not shown) executes processing by a determining unit 820 that determines whether the display on the display device 206 and the input on the input device 205 are valid.

Furthermore, the storage device 204 stores uncreated data 801 in which a type number has not been input and created data 802 in which a type number has been input.
Uncreated data 801 is alignment data 301 acquired in alignment processing, and is data for creating learning data.
Further, the created data 802 is data to which a type number is added to the uncreated data 801, and becomes the learning data 305 input to the image classification means 400.

Here, the display device 206, the input device 205, and the storage device 204 may be provided in the exposure apparatus 10, but are not limited to this, and may be provided in external information processing devices such as the host computer 11 and the management device 12. do not have.
Further, the display means 800 and the input means 810 can be realized by a software program executed in at least one of the main control section 100, the management device 12, and the host computer 11 of the exposure apparatus 10.
Further, the determining means 820 can be realized by a software program executed in at least one of the main control unit 100, the management device 12, and the host computer 11 of the exposure apparatus 10.

The display unit 800 causes the display device 206 to display information necessary for creating the learning data 305.
FIG. 5 is a diagram exemplarily showing a creation screen for the learning data 305.

As shown in FIG. 5, information related to data included in the uncreated data 801 is displayed on the screen 910.
For example, a mark image 911 of the mark 211 captured by the substrate alignment optical system 190 is displayed on the screen 910.
Further, on the screen 910, data 912 related to when the mark 211 is imaged, such as the model of the exposure apparatus 10, the installation line of the exposure apparatus 10, the illumination conditions of the exposure apparatus 10, and the position information of the substrate stage 200, is displayed. .

Further, on the screen 910, classification information 913 indicating classification options and a selection state indicating whether or not a classification has been selected is displayed.
As for the selection state of the classification information 913, selection or non-selection can be input using the input device 205.
When the confirm button 914 is pressed with a certain category selected, information on the selected category is input for the displayed uncreated data 801.
Furthermore, when the cancel button 915 is pressed, the creation of the learning data 305 is canceled.

Note that the display unit 800 may display a plurality of uncreated data 801, a plurality of related data 912, and a plurality of classification information 913 on the screen 910, and allow the user to select a classification for the plurality of uncreated data 801. .

The input means 810 acquires the classification selection information input from the input device 205. Then, the input means 810 associates the classification selection information with the uncreated data 801 stored in the storage device 204 and stores it in the storage device 204 as created data 802.
The determining means 820 determines the start and end of creation of the learning data 305 based on predetermined conditions. That is, the determining means 820 determines whether to start the process of causing the display means 800 to display information for selecting a classification on the display device 206 and inputting the classification selection information to the input means 810.
Further, the determining means 820 determines whether to end the displaying of information for selecting a classification on the display device 206 by the displaying means 800 and the inputting of classification selection information by the inputting means 810.

When the number of created data 802 reaches a predetermined number, the image classification means 400 may perform additional learning using the created data 802 as learning data 305 and delete the created data 802 from the storage device 204. .
Further, the storage device 204 may store the numbers of uncreated data 801 and created data 802, and the numbers of these data may be updated by the input means 810 and the image classification means 400.

The storage device 204 also stores the number of learned data (data added to the learning data 305) and unlearned data (data not added to the learning data 305) among the created data 802. Good too. Then, the number of data items may be updated by the input means 810 or the image classification means 400.
Further, the display unit 800 may display the number of data items on the display device 206.

Next, the process of creating learning data 305 will be explained.
FIG. 6 is a flowchart showing the process of creating learning data 305.

In S110, the determining means 820 determines whether to start creating the learning data 305 based on a predetermined condition for starting creating the learning data 305.
If the determining means 820 determines not to start creating the learning data 305, the process returns to S110 after a predetermined period has elapsed, and it is determined whether to start creating the learning data 305 again.

On the other hand, if the determining means 820 determines to start creating the learning data 305, the process advances to S111 and starts creating the learning data 305.
Then, in S111, information on the classification type number is added to the alignment data 301 included in the uncreated data 801 according to the above-described procedure.

Then, in S112, the alignment data 301 to which the type number has been added is deleted from the uncreated data 801 and added to the created data 802.

Then, in S113, the determining means 820 determines whether to end the creation of the learning data 305 based on a predetermined condition for ending the creation of the learning data 305.
If the determining means 820 determines that the creation of the learning data 305 is not finished, the process returns to S111, and the next uncreated data 801 is displayed on the display device 206.

On the other hand, when the determining means 820 determines that the creation of the learning data 305 is to be completed, the display of the screen 910 is ended, and the process of creating the learning data 305 is ended.

Further, the display unit 800 may allow the user to determine whether to display a screen for classifying the uncreated data 801 on the display device 206.
FIG. 7 is an exemplary diagram showing buttons for displaying a creation screen for learning data 305.

The button 901 is a button that allows the user to determine whether to display a screen for classifying the uncreated data 801 on the display device 206.
The display unit 800 displays a button 901 on the screen 900, and when the button is pressed by the user, causes the display device 206 to display a screen for classifying the uncreated data 801.

Further, the display unit 800 may display a message 902 indicating the number of uncreated data 801 together with a button 901.
By displaying the message 902, the user can determine whether to start creating the learning data 305 based on the number of uncreated data 801.

Further, by providing the image classification means 400 in a device provided outside the exposure apparatus 10, such as the management apparatus 12, it is possible to receive alignment data 301 from a plurality of exposure apparatuses 10 and create learning data 305.

Further, the image classification means 400 may store all or part of the received alignment data 301 for a preset period or up to the upper limit of the number of items.

Furthermore, an arbitrary type may be selected from a plurality of types displayed in a list, and the alignment data 301 classified and stored according to the selected type may be called up and displayed on the screen.
Further, the number of alignment data 301 included in the selected type may be totaled and the result may be displayed.
Alternatively, the alignment data 301 included in the selected type may be classified by associated data, aggregated, and the results may be displayed.

Then, when the alignment data 301 is classified by the image classification means 400 in step S403, the classification result is displayed by the exposure device 10 or the management device 12 (step S404).

Then, based on the displayed classification results, the user or the determination device determines whether maintenance processing 303 is possible (determines the necessity of maintenance) (step S405).
If it is determined that maintenance processing is possible (Yes in step S405), maintenance processing 303 is executed manually or automatically (step S406). On the other hand, if it is determined that the maintenance process 303 is not possible (No in step S405), the determination of whether to perform the maintenance process is ended.
Note that the maintenance process 303 here is shown in Table 1, for example, and may be automatically performed by the device, or may display a warning so that the user can manually perform the maintenance process 303.

Next, it is monitored whether the alignment data 301 is classified again into the type number for which the maintenance process 303 was performed, that is, whether a similar alignment failure occurs again despite the maintenance process 303 being performed. (Step S407).

If the alignment data 301 is classified again into the type number for which the maintenance process 303 was performed, that is, if the problem has not been resolved (No in step S407), the alignment data 301 is classified into another type number. Additional learning is performed (step S408).
Note that this additional learning (changing the classification criteria) may be performed manually by the user, or may be automatically performed by the device.
On the other hand, if the alignment data 301 is not classified again into the type number for which the maintenance process 303 was performed at the predetermined time (Yes in step S407), the determination of whether to perform the maintenance process is ended.

As described above, the determination device according to the present embodiment classifies alignment data 301 acquired by the exposure apparatus 10 regarding causes of alignment failure using a learning model acquired by machine learning. Then, it is determined whether the exposure apparatus 10 needs to be maintained based on the classified causes of alignment failure.
Thereby, it is possible to obtain a determination device that can determine whether the exposure apparatus 10 needs to be maintained.

[Second embodiment]
FIGS. 8A and 8B are a block diagram and a process flow diagram, respectively, showing a configuration for determining whether maintenance is required for the exposure apparatus 10 in a determination device according to the second embodiment.
Note that the determination device according to the present embodiment has the same configuration as the determination device according to the first embodiment except that a failure determination means 430 is newly provided, so the same members are given the same reference numbers. , and the explanation will be omitted.

First, the image processing means 300 provided in the exposure apparatus 10 performs image processing on the substrate 210, and a mark image (image data) is obtained (step S601).

If the image processing means 300 determines that the alignment error of the substrate 210 is within the allowable range based on the mark image and that the alignment is successful, the exposure processing means 350 provided in the exposure apparatus 10 executes the exposure process. Ru.
Further, at the same time as the exposure process is executed, the image processing means 300 acquires the alignment data 301 by adding related data to the mark image, and then the alignment data 301 is passed to the image classification means 400 (step S602).
Further, even if the image processing means 300 determines that the alignment error of the substrate 210 is not within the allowable range based on the mark image and the alignment has failed, the alignment data 301 is similarly acquired and then passed to the image classification means 400. (Step S602).

Note that even if the image processing means 300 determines that the alignment is successful, it may actually fail due to erroneous measurement.
In such a case, the alignment data 301 that was determined to be successful due to erroneous measurement may be re-determined to have failed based on the result of overlay measurement of the substrate 210 by an external measuring device (not shown).

Here, the related data is data that includes information related to the acquired mark image. For example, the related data may include information specifying the configuration of the exposure apparatus 10, such as the model, number, hardware configuration, software configuration, and installation line.
Further, the related data may also include information specifying the configuration such as lot, substrate 210, original plate 170, recipe, environmental conditions, processing date and time.
Further, the related data may include information specifying configurations such as various offset settings of the exposure apparatus 10 during mark measurement, illumination conditions such as the light amount of the light source 120 that illuminates the mark 211 and the focus amount of the optical system.
Further, the related data may include information specifying the configuration, such as operating conditions during alignment of the exposure apparatus 10, such as the type of the mark 211, and position information of the stage.
Further, the related data may include information specifying the configuration, such as the previous alignment measurement result and the operating state during alignment, such as the pressure with which the substrate stage 200 attracts the substrate 210.

Furthermore, the alignment data 301 delivered to the image classification means 400 is not limited to the above, and may be the result of classifying mark images by an image classification means (not shown) using machine learning or the like.

As described above, the alignment data 301 passed to the image classification means 400 includes, but is not limited to, any alignment data 301 for which the image processing means 300 determines that the alignment was successful or failed. .
In order to improve throughput, only the alignment data 301 for which alignment has been determined to have failed by the image processing means 300 may be passed to the image classification means 400.

Further, the number of marks 211 measured in step S601 may be one or more, and the number of alignment data 301 transferred to the image classification means 400 may be one or more.
Further, the alignment data 301 may be transferred to the image classification means 400 sequentially every time image processing for one mark 211 is completed. Further, the present invention is not limited to this, and the image processing may be performed all at once after the image processing is completed for all the marks 211 on the substrate 210.

Next, the image classification means 400 obtains classification information 302 by classifying the received alignment data 301 into one of a plurality of types related to alignment failure factors (step S603).
Note that the image classification means 400 can be realized by a software program executed in at least one of the main control unit 100, the management device 12, and the host computer 11 of the exposure apparatus 10.

As the classification information 302, for example, type numbers 0 to 5 as shown in Table 2 below can be set.

Note that the numbers for each type as shown in Table 2 may be manually set based on the alignment data 301 when the alignment process failed in the past, the result of the automatic recovery operation of the exposure apparatus 10, etc. Alternatively, each type of number as shown in Table 2 may be automatically provided using machine learning or the like.

As shown in Table 2, in the determination device according to the present embodiment, the classification information 302 includes a cause of failure in alignment processing and a maintenance method for improving the cause for each type number.

Specifically, type number 0 is normal alignment data 301 that does not include a cause of alignment failure, and corresponds to the classification when maintenance is not required.

Further, type number 1 corresponds to the classification when the cause of alignment failure cannot be identified and the maintenance method is unknown.
In addition, type number 2 indicates that the cause of alignment failure is a positional shift when transferring the substrate 210 or a positional change of the mark 211 depending on the processing process of the substrate 210, and in the former case, the maintenance method is It corresponds to classification when it is a position adjustment.
Further, type number 3 corresponds to a classification in which the cause of alignment failure is a setting error in the light intensity of the light source 120 during alignment measurement, and the maintenance method is adjustment of the light intensity of the light source 120 during alignment measurement.
Further, type number 4 corresponds to a classification in which the cause of alignment failure is vibration of the substrate stage 200 during alignment measurement or a decrease in contrast depending on the processing process of the substrate 210. If the cause of alignment failure is the former, it corresponds to the classification where the maintenance method is adjustment for vibration of substrate stage 200 during alignment measurement.
Further, type number 5 corresponds to a classification in which the cause of alignment failure is deterioration of the light source 120 and the maintenance method is replacement of the light source 120.

For these classifications, the related data added in the alignment data 301 is effectively used.
Note that the above types are just examples, and other classification types may be set.

Note that if the alignment data 301 applies to the above types in multiple ways, it may be classified into the item with the greatest degree of applicability.
Further, the present invention is not limited to this, and if the alignment data 301 applies to the above types in multiple ways, it may be classified by weighting according to the degree to which it applies to each type.

In the determination device according to the first embodiment, as shown in Table 1, one factor is specified for one type number, and one maintenance method is presented.
However, in the determination device according to the present embodiment, as shown in Table 2, multiple factors may be identified for one type number, and maintenance methods may be presented for each factor. be.

For example, in type number 2 shown in Table 2, the alignment failure factor is "positional shift during transfer of the substrate 210" which is a factor originating from the exposure apparatus 10, or "positional shift during transfer of the substrate 210" which is a factor originating from the substrate processing process. "positional fluctuation of the mark 211 depending on the
This means that it is not possible to narrow down the cause of alignment failure to one using only the existing alignment data 301.

For example, consider a case where the position of the mark 211 in the mark image included in the alignment data 301 is significantly shifted.

At this time, an alignment failure occurs because the position of the mark 211 is shifted so much that it cannot be measured by the pattern matching process of the exposure apparatus 10.
Assume that the alignment data 301 is classified into type number 2 based on the classification by the image classification means 400, based on the fact that the position of the mark 211 is largely shifted.

At this time, depending on the factor specified in type number 2, it is possible to perform recovery by maintaining the exposure apparatus 10.
That is, if this alignment failure is caused by a shift in the receiving position of the substrate 210 in the exposure apparatus 10, recovery can be performed by adjusting the receiving position of the substrate 210 in the exposure apparatus 10.
However, if the position of the mark 211 varies depending on the processing process of the substrate 210, that is, if the mark 211 is formed at a shifted position on the substrate 210, the substrate processing process in an apparatus other than the exposure apparatus 10 may be taken into account. Adjustments will be required. Therefore, it is not possible to restore the exposure apparatus 10 through maintenance processing.

In the determination device according to this embodiment, when the alignment data 301 is classified by the image classification means 400 into a type number in which a plurality of factors are suspected as described above, the failure determination means 430 is used.
That is, the image classification means 400 passes the classification information 302 thus classified to the failure determination means 430, and determines whether the cause is due to the exposure apparatus (step S604).
Here, the failure determination means 430 can be realized, for example, by a software program executed in the management device 12.

Table 3 exemplarily shows the number of times that a plurality of mark images acquired during a predetermined number of past alignment processing failures were classified into each type of number for each exposure device.

As described above, the cause of alignment failure may be that alignment operating conditions such as the type of light source 120 used during alignment measurement and the shape of mark 211 differ depending on the exposure apparatus, that is, it may depend on the apparatus.
For this reason, the failure determination unit 430 makes the determination by comparing the classification information 302 obtained by the same number of alignment processes in the past between exposure apparatuses, as shown in Table 3.

For example, if we focus on type number 2 in Table 3, it can be seen that the number of times the exposure apparatus EQ2 is classified is significantly greater than that of other exposure apparatuses.
Therefore, if the alignment data 301 classified into type number 2 is acquired by the exposure apparatus EQ2, the failure determination means 430 identifies the above comparison result, that is, the cause is originating from the exposure apparatus. The determination result 431 is sent back to the image classification means 400.

Then, the image classification means 400 selects an appropriate alignment failure factor from the identified plurality of alignment failure factors based on the determination result 431, that is, performs further classification, and outputs classification information 302 (step S605). .
That is, when classifying the alignment data 301 acquired by the exposure apparatus EQ2 into type number 2, the image classification means 400 classifies that the cause of the alignment failure is a positional shift when the substrate 210 is transferred. Then, the maintenance method can be classified as adjusting the delivery position of the board 210.

Note that the failure determination means 430 makes a determination by calculating the median or average value of the number of times of classification in each exposure device, and determining that the difference between the number of times of classification in a predetermined exposure device exceeds a threshold value. You may go.
Further, the determination by the failure determination means 430 is based on the test statistic |x−μ This may be done by determining that |/σ exceeds a threshold value.
Further, the determination method in the failure determination means 430 is not limited to the above, and it is also possible to adopt a method of statistically selecting outliers.

Further, in the determination apparatus according to the present embodiment, the determination by the failure determination means 430 is performed by comparing the classifications of the same number of mark images acquired by each exposure device, but the present invention is not limited to this. For example, the determination by the failure determining means 430 may be made by comparing the classification of mark images acquired within the same period in each exposure apparatus.
Furthermore, the determination by the failure determining means 430 may be limited to mark images acquired in alignment processing performed under specific operating conditions such as a predetermined alignment mode or recipe. That is, the determination by the failure determining means 430 may be made by comparing the classification of mark images acquired under the same operating conditions in each substrate processing apparatus.

Then, when the image classification means 400 outputs the classification information 302 in step S605, the classification result is displayed by the device (step S606).

Then, based on the displayed classification results, the user or the device determines whether the maintenance process 303 is possible (step S607).
If it is determined that the maintenance process 303 is possible (Yes in step S607), the maintenance process 303 is executed manually or automatically (step S608). On the other hand, if it is determined that the maintenance process 303 is not possible (No in step S607), the determination of whether to perform the maintenance process is ended.
The maintenance process 303 here is shown in Table 2, for example, and may be automatically executed by the device, or may display a warning so that the user can manually execute the maintenance process 303.

Next, it is monitored whether the alignment data 301 is classified again into the type number for which the maintenance process 303 was performed, that is, whether a similar alignment failure occurs again despite the maintenance process 303 being performed. (Step S609).

Then, if the alignment data 301 is classified again into the type number for which the maintenance process 303 was performed, that is, if it is determined that the problem has not been resolved (No in step S609), one of the following two steps is performed. .
That is, additional learning is performed so that the image classification means 400 classifies the alignment data 301 into a different type number (changing the classification standard), or the threshold value for the judgment in step S604 is changed (the judgment standard is changed). ) (Step S610).
Note that the additional learning in step S610 may be performed manually by the user or automatically by the device. Further, the threshold value change in step S610 may be performed manually by the user, or may be automatically performed by the device using machine learning or the like.

On the other hand, if the alignment data 301 is not classified again into the type number for which the maintenance process 303 was performed at the predetermined time (Yes in step S609), the determination of whether to perform the maintenance process is ended.

As described above, in the determination device according to the present embodiment, the alignment data 301 acquired by the exposure apparatus 10 is classified regarding alignment failure factors using a learning model acquired by machine learning, and each exposure apparatus The classifications in 10 are compared with each other.
Thereby, it is determined whether the classified alignment failure factor originates from the exposure apparatus 10, and based on this, it is determined whether the exposure apparatus 10 needs to be maintained.

Thereby, it is possible to obtain a determination device that can determine with higher accuracy whether or not the exposure apparatus 10 needs to be maintained.

[Method for manufacturing articles]
The article manufacturing method using the determination device according to the present embodiment is suitable for manufacturing articles such as devices (semiconductor elements, magnetic storage media, liquid crystal display elements, etc.), for example.
The method for manufacturing an article according to the present embodiment also includes a step of exposing a substrate coated with a photosensitive agent (forming a pattern on the substrate) using the exposure apparatus 10, and developing the exposed substrate (not shown). and a step of developing (processing the substrate) using the apparatus.

Further, the manufacturing method according to the present embodiment may include other well-known steps (oxidation, film formation, vapor deposition, doping, planarization, etching, resist stripping, dicing, bonding, packaging, etc.).
The method for manufacturing an article according to the present embodiment is advantageous in at least one of the performance, quality, productivity, and production cost of the article compared to the conventional method.

Although preferred embodiments have been described above, it goes without saying that the present invention is not limited to these embodiments, and various modifications and changes can be made within the scope of the gist thereof.
Further, although an exposure apparatus has been described as an example of the substrate processing apparatus 10, the present invention is not limited to this.
For example, an example of the substrate processing apparatus 10 may be an imprint apparatus that forms a pattern of imprint material on a substrate using a mold.

Further, as an example of the substrate processing apparatus 10, it may be a drawing apparatus that forms a pattern on the substrate by drawing on the substrate with a charged particle beam (electron beam, ion beam, etc.) via a charged particle optical system.
The substrate processing apparatus 10 also includes a coating apparatus that applies a photosensitive medium onto the surface of a substrate, a developing apparatus that develops a substrate on which a pattern is formed, and an imprint apparatus that is used in manufacturing articles such as devices. It may also include manufacturing equipment that performs processes other than those performed by the equipment.

Furthermore, the scope of the present embodiment also includes a method and a program for implementing the embodiments described above, and a computer-readable recording medium on which the program is recorded.

10 Substrate processing device 12 Management device (judgment device)
210 Board 211 Mark 301 Alignment data (image data)

Claims (19)

Classify the image data of the marks on the substrate captured by the substrate processing equipment into alignment failure factors using a learning model obtained by machine learning, and maintain the substrate processing equipment based on the classification results. A determination device that determines whether it is necessary to do so. 2. The determining device according to claim 1, wherein the determining device performs the classification on alignment data obtained by adding related data including information related to the image data to the image data. 3. The determination device according to claim 1, wherein the determination device performs the classification by comparing the classifications among a plurality of the substrate processing apparatuses. 4. The determining device according to claim 3, wherein the determining device determines, based on the comparison, whether the cause of the alignment failure originates from the substrate processing apparatus. 2. The determination device determines whether the cause of the alignment failure originates from the substrate processing device by comparing the classifications for the same number of image data acquired by each substrate processing device. 4. The determination device according to 4. The determining device determines whether the cause of the alignment failure originates from the substrate processing apparatus by comparing the classifications of the image data acquired within the same period in each substrate processing apparatus. The determination device according to item 4. The determination device is characterized in that it determines whether the cause of the alignment failure originates from the substrate processing device by comparing the classification of the image data acquired under the same operating conditions in each substrate processing device. The determination device according to any one of claims 4 to 6. 8. The determining device executes maintenance processing for the substrate processing device when determining that the cause of the alignment failure originates from the substrate processing device. Judgment device. The determination device determines whether the cause of alignment failure has been resolved after performing the maintenance process, and if not resolved, a criterion for determining whether the cause of alignment failure originates from the substrate processing device. The determination device according to claim 8, characterized in that the determination device changes: 10. Any one of claims 1 to 9, wherein, after classifying the image data, the determination device executes maintenance processing for the substrate processing device based on a maintenance method corresponding to the classified alignment failure factor. The determination device according to item (1). 11. The determining device determines whether the alignment failure factor has been resolved after executing the maintenance process, and if the cause has not been resolved, changes the classification criterion. judgment device. 12. The determination device according to claim 1, wherein the determination device displays at least one of the classification result and a maintenance method corresponding to the classification result. 13. The image data includes only image data for which the alignment process has failed, and the determining device performs the classification only for the image data for which the alignment process has failed. The determination device according to item (1). 14. The substrate processing apparatus is an exposure apparatus that exposes the substrate using exposure light so as to transfer a pattern formed on an original onto the substrate. The judgment device described in . 15. A substrate processing apparatus for processing a substrate, comprising the determination device according to claim 1. A step of processing a substrate using the substrate processing apparatus according to claim 15,
A method for manufacturing an article, comprising manufacturing an article from the treated substrate. a plurality of substrate processing devices that process the substrate;
a host computer that controls operations of the plurality of substrate processing apparatuses;
a management device that manages maintenance of the plurality of substrate processing devices;
Equipped with
A substrate processing system, wherein the management device includes the determination device according to any one of claims 1 to 14. A step of classifying image data of marks on the substrate captured by the substrate processing apparatus regarding causes of alignment failure using a learning model obtained by machine learning;
a step of determining whether it is necessary to maintain the substrate processing apparatus based on the result of the step of performing the classification;
A determination method characterized by having the following. A computer-readable recording medium on which a program that causes a computer to determine the necessity of maintenance is recorded,
A step of classifying image data of marks on the substrate captured by the substrate processing apparatus regarding causes of alignment failure using a learning model obtained by machine learning;
a step of determining whether it is necessary to maintain the substrate processing apparatus based on the result of the step of performing the classification;
A computer-readable recording medium on which a program for causing a computer to execute is recorded.

Images