JP7594914B2 - Substrate processing condition setting support method, substrate processing system, storage medium, and learning model - Google Patents

Description

The present disclosure relates to a method for supporting condition setting for substrate processing, a substrate processing system, a storage medium, and a learning model.

Patent document 1 discloses an apparatus for forming a photosensitive film on the surface of a substrate, exposing the photosensitive film to light, and then developing the photosensitive film.

JP 2017-73522 A

The present disclosure provides a condition setting support method that is effective in simplifying the task of setting processing conditions for substrate processing.

A method for supporting condition setting for substrate processing according to one aspect of the present disclosure includes inputting a dataset including processing conditions for substrate processing performed by a substrate processing apparatus, including supply of processing liquid to a substrate, and performance data related to the quality of the substrate processing into a machine learning device, and deriving recommended processing conditions for the substrate processing based on a model generated by the machine learning device through machine learning based on a plurality of sets of the datasets, the model outputting predictive data related to the quality of the substrate processing in response to the input of processing conditions.

The present disclosure provides a condition setting support method that is effective in simplifying the task of setting processing conditions for substrate processing.

1 is a schematic diagram illustrating a configuration of a substrate processing system according to an exemplary embodiment. FIG. 2 is a schematic diagram illustrating a schematic configuration of a coating unit. FIG. 2 is a schematic diagram illustrating a schematic configuration of a development unit; FIG. 2 is a schematic diagram illustrating a schematic configuration of a post-processing inspection device. FIG. 2 is a schematic diagram illustrating a schematic configuration of an in-process inspection device; FIG. 2 is a block diagram illustrating an example of the functional configuration of a control device and a machine learning device. FIG. 2 is a block diagram illustrating an example of the hardware configuration of a control device and a machine learning device. 4 is a flowchart illustrating a condition setting assistance procedure executed by the control device. 10 is a flowchart illustrating a condition setting assistance procedure further executed by the control device. 11 is a flowchart illustrating a condition setting assistance procedure executed by the machine learning device. 13 is a flowchart illustrating a condition setting support procedure further executed by the machine learning device.

Various exemplary embodiments are described below. In the description, identical elements or elements having the same functions are given the same reference numerals, and duplicate descriptions are omitted.

[Substrate Processing System]
The substrate processing system 1 is a system that forms a photosensitive film on a surface of a substrate and performs a development process on the photosensitive film after an exposure process. The substrate to be processed is, for example, a semiconductor wafer W. The photosensitive film is, for example, a resist film.

As illustrated in FIG. 1, the substrate processing system 1 includes a coating/developing apparatus 2 and a control device 100. The coating/developing apparatus 2 includes a carrier block 4, a processing block 5, and an interface block 6.

The carrier block 4 introduces wafers W (substrates) into the coating and developing apparatus 2 and removes wafers W from the coating and developing apparatus 2. For example, the carrier block 4 can support multiple carriers C for wafers W and has a built-in transfer arm A1. The carrier C accommodates, for example, multiple circular wafers W. The transfer arm A1 removes unprocessed wafers W from the carrier C and returns processed wafers W to the carrier C.

The processing block 5 has multiple processing modules 11, 12, 13, and 14. The processing modules 11, 12, and 13 (processing sections) perform a film formation process in which a film formation liquid (processing liquid for film formation) is applied to the surface Wa of the wafer W to form a coating. For example, the processing modules 11, 12, and 13 each incorporate a coating unit U1, a heat treatment unit U2, and a transport arm A3 that transports the wafer W to these units.

The processing module 11 forms an underlayer film on the surface of the wafer W using a coating unit U1 and a heat treatment unit U2. The coating unit U1 of the processing module 11 applies a treatment liquid for forming the underlayer film onto the wafer W. The heat treatment unit U2 of the processing module 11 performs various heat treatments associated with the formation of the underlayer film.

The processing module 12 forms a resist film on the underlayer film using a coating unit U1 and a heat treatment unit U2. The coating unit U1 of the processing module 12 applies a treatment liquid for forming a resist film onto the underlayer film. The heat treatment unit U2 of the processing module 12 performs various heat treatments associated with the formation of the resist film.

The processing module 13 forms an upper layer film on the resist film using a coating unit U1 and a thermal processing unit U2. The coating unit U1 of the processing module 13 applies a liquid for forming the upper layer film onto the resist film. The thermal processing unit U2 of the processing module 13 performs various thermal processes associated with the formation of the upper layer film.

As illustrated in FIG. 2, the coating unit U1 has a rotating holder 50 and a film forming liquid supply unit 60. The rotating holder 50 holds and rotates the wafer W. For example, the rotating holder 50 has a holder 51 and a rotation drive unit 52. The holder 51 supports the horizontally arranged wafer W and holds it, for example, by vacuum suction. The rotation drive unit 52 uses, for example, an electric motor as a power source, to rotate the holder 51 around a vertical axis. This causes the wafer W held by the holder 51 to also rotate.

The film formation liquid supply unit 60 supplies the film formation liquid to the surface Wa of the wafer W held in the holding unit 51. For example, the film formation liquid supply unit 60 has a nozzle 61 and a liquid source 62. The nozzle 61 is disposed above the wafer W held in the holding unit 51, and ejects the processing liquid downward. The liquid source 62 pressure-feeds the processing liquid to the nozzle 61.

Returning to FIG. 1, the processing module 14 (processing section) performs a development process in which a developing solution is supplied to the resist film (photosensitive film) that has been subjected to an exposure process on the surface Wa of the wafer W. For example, the processing module 14 incorporates a development unit U3, a heat treatment unit U4, and a transport arm A3 that transports the wafer W to these units. The processing module 14 performs a development process on the resist film after exposure using the development unit U3 and the heat treatment unit U4. The development unit U3 applies a developing solution (developing solution) to the surface of the exposed wafer W, and then rinses it off with a rinsing solution (rinsing solution) to perform a development process on the resist film. The heat treatment unit U4 performs various heat treatments associated with the development process. Specific examples of heat treatments include a heating process before the development process (PEB: Post Exposure Bake), a heating process after the development process (PB: Post Bake), etc.

As illustrated in FIG. 3, the developing unit U3 has a rotating holder 20, a developer supply unit 30, and a rinse liquid supply unit 40. The rotating holder 20 holds and rotates the wafer W. For example, the rotating holder 20 has a holder 21 and a rotation drive unit 22. The holder 21 supports the horizontally arranged wafer W and holds it, for example, by vacuum suction. The rotation drive unit 22 uses, for example, an electric motor as a power source, to rotate the holder 21 around a vertical axis. This causes the wafer W held by the holder 21 to also rotate.

The developer supply unit 30 supplies developer to the surface Wa of the wafer W held in the holder 21. For example, the developer supply unit 30 has a nozzle 31, a nozzle transfer unit 32, and a liquid source 33. The nozzle 31 is disposed above the wafer W held in the holder 21, and ejects developer downward. The nozzle transfer unit 32 moves the nozzle 31 horizontally using an electric motor or the like as a power source. The liquid source 33 pressure-feeds the developer to the nozzle 31.

The rinse liquid supply unit 40 supplies rinse liquid to the surface Wa of the wafer W held in the holder 21. For example, the rinse liquid supply unit 40 has a nozzle 41, a nozzle transfer unit 42, and a liquid source 43. The nozzle 41 is disposed above the wafer W held in the holder 21, and ejects rinse liquid downward. The nozzle transfer unit 42 moves the nozzle 41 horizontally using an electric motor or the like as a power source. The liquid source 43 pressure-feeds the rinse liquid to the nozzle 41.

Returning to FIG. 1, the interface block 6 transfers the wafer W between the exposure device (not shown) that performs an exposure process on the resist film formed on the wafer W. For example, the interface block 6 has a built-in transfer arm A8 that is connected to the exposure device. The transfer arm A8 transfers the wafer W before exposure processing to the exposure device and receives the wafer W after exposure processing from the exposure device.

An accommodation unit U10 is provided between the processing block 5 and the carrier block 4. The accommodation unit U10 is divided into a plurality of cells arranged in the vertical direction, and each cell can accommodate a wafer W. The accommodation unit U10 is used for transferring the wafer W between the carrier block 4 and the processing block 5, etc. A lift arm A7 is provided near the accommodation unit U10. The lift arm A7 raises and lowers the wafer W between the cells of the accommodation unit U10. An accommodation unit U11 is provided between the processing block 5 and the interface block 6. The accommodation unit U11 is also divided into a plurality of cells arranged in the vertical direction, and each cell can accommodate a wafer W. The accommodation unit U11 is used for transferring the wafer W between the processing block 5 and the interface block 6, etc.

The control device 100 controls the coating/developing device 2 to perform coating/developing processes, for example, in the following manner. First, the control device 100 controls the transfer arm A1 to transport the wafer W in the carrier C to the accommodation unit U10, and then controls the lift arm A7 to place the wafer W in the cell for the processing module 11.

Next, the control device 100 controls the transport arm A3 to transport the wafer W from the accommodation unit U10 to the coating unit U1 and heat treatment unit U2 in the processing module 11. The control device 100 also controls the coating unit U1 and heat treatment unit U2 to form an underlayer film on the surface of the wafer W. The control device 100 then controls the transport arm A3 to return the wafer W with the underlayer film formed thereon to the accommodation unit U10, and controls the lift arm A7 to place the wafer W in a cell for the processing module 12.

Next, the control device 100 controls the transport arm A3 to transport the wafer W from the accommodation unit U10 to the coating unit U1 and heat treatment unit U2 in the processing module 12. The control device 100 also controls the coating unit U1 and heat treatment unit U2 to form a resist film on the underlying film of the wafer W. Thereafter, the control device 100 controls the transport arm A3 to return the wafer W to the accommodation unit U10, and controls the lift arm A7 to place the wafer W in a cell for the processing module 13.

Next, the control device 100 controls the transfer arm A3 to transfer the wafer W in the accommodation unit U10 to each unit in the processing module 13. The control device 100 also controls the coating unit U1 and the heat treatment unit U2 to form an upper layer film on the resist film of the wafer W. Thereafter, the control device 100 controls the transfer arm A3 to transfer the wafer W to the accommodation unit U11.

Next, the control device 100 controls the transfer arm A8 to send the wafer W in the accommodation unit U11 to the exposure device 3. The control device 100 then controls the transfer arm A8 to receive the wafer W that has been subjected to the exposure process from the exposure device 3 and place it in a cell for the processing module 14 in the accommodation unit U11.

Next, the control device 100 controls the transport arm A3 to transport the wafer W from the accommodation unit U11 to each unit in the processing module 14, and controls the development unit U3 and the heat treatment unit U4 to perform a development process on the resist film on the wafer W. The control device 100 then controls the transport arm A3 to return the wafer W to the accommodation unit U10, and controls the lift arm A7 and the transfer arm A1 to return the wafer W into the carrier C. This completes the coating and development process.

The specific configuration of the substrate processing system is not limited to the above example. The substrate processing system may be any system that includes a processing section that performs substrate processing, including supplying a processing liquid to the substrate, and a control device 100 that can control the processing section.

[Condition setting support system]
The substrate processing system 1 further includes a condition setting system 7. The condition setting system 7 has a quality inspection device 70. At least a part of the condition setting system 7 is configured by the above-mentioned control device 100. That is, the condition setting system 7 has the quality inspection device 70 and the control device 100. The quality inspection device 70 detects information related to the quality of the substrate processing performed by the coating/developing apparatus 2.

The control device 100 is configured to execute the following: causing the coating/developing device 2 (substrate processing device) to perform substrate processing, including supplying a processing liquid to the wafer W, according to preset processing conditions; acquiring performance data on the quality of the substrate processing according to the processing conditions from the quality inspection device 70; inputting a data set including the processing conditions of the substrate processing and the performance data of the substrate processing into the machine learning device 200; and deriving recommended processing conditions for the substrate processing based on a learning model generated by the machine learning device 200 by machine learning based on multiple sets of data sets so as to output predicted data on the quality of the substrate processing in response to the input of the processing conditions. The predicted data is, for example, data that predicts the above-mentioned performance data. The performance data may be any data as long as it is related to the quality of the substrate processing. Data on the quality of the substrate after the substrate processing is related to the quality of the substrate processing. In addition, the supply state of the processing liquid during the substrate processing is also related to the quality of the substrate processing.

The condition setting system 7 may further include a machine learning device 200. The machine learning device 200 is configured to acquire the above-mentioned dataset and generate the above-mentioned learning model by machine learning based on a plurality of sets of datasets. The machine learning device 200 may be housed in the same housing as the control device 100, or may be installed at a location remote from the control device 100. When installed at a location remote from the control device 100, the machine learning device 200 is connected to the control device 100, for example, via a local area network. The machine learning device 200 may be connected to the control device 100 via a wide area network such as the so-called Internet. The configuration of each part will be described in detail below.

(Quality Data Detector)
The quality inspection device 70 has a post-processing inspection unit 80 shown in FIG. 4, for example. The post-processing inspection unit 80 detects information related to the quality of the substrate after the substrate processing. As an example, the post-processing inspection unit 80 detects information related to the line width of the resist pattern formed on the surface of the wafer W after the development processing. For example, the post-processing inspection unit 80 detects image information that allows a difference in the line width of the resist pattern to be recognized as a difference in at least one of hue, brightness, and saturation.

Specifically, the post-processing inspection unit 80 includes a holding unit 83, a linear drive unit 84, an imaging unit 81, and a light projecting/reflecting unit 82. The holding unit 83 holds the wafer W horizontally. The linear drive unit 84 is powered by, for example, an electric motor and moves the holding unit 83 along a horizontal, straight path. The imaging unit 81 acquires image data of the surface of the wafer W. The imaging unit 81 is provided at one end of the post-processing inspection unit 80 in the movement direction of the holding unit 83, and is directed to the other end of the movement direction.

The light projecting/reflecting unit 82 projects light onto the imaging range and directs the reflected light from the imaging range towards the imaging unit 81. For example, the light projecting/reflecting unit 82 has a half mirror 86 and a light source 87. The half mirror 86 is provided in a position higher than the holding unit 83, in the middle of the moving range of the holding unit 83, and reflects light from below towards the imaging unit 81. The light source 87 is provided above the half mirror 86 and irradiates illumination light downward through the half mirror 86.

The post-processing inspection unit 80 operates as follows to acquire image data of the surface of the wafer W. First, the linear drive unit 84 moves the holder 83. This causes the wafer W to pass under the half mirror 86. During this passage, reflected light from each part of the surface of the wafer W is sent sequentially to the imaging unit 81. The imaging unit 81 forms an image of the reflected light from each part of the surface of the wafer W, and acquires image data of the surface of the wafer W. This allows image information of the resist pattern to be detected.

The post-processing inspection unit 80 may detect information regarding the thickness of the coating formed on the surface of the wafer W after the film formation process. For example, the post-processing inspection unit 80 detects image information that allows differences in the coating thickness to be recognized as differences in at least one of hue, brightness, and saturation. The image information can also be detected by the configuration illustrated in FIG. 4.

The quality inspection device 70 may further include an in-process inspection unit 90 shown in FIG. 5. The in-process inspection unit 90 detects information relating to the supply state of processing liquid during substrate processing. As an example, the in-process inspection unit 90 detects information relating to the supply state of developing liquid during development processing. For example, the in-process inspection unit 90 includes a liquid splash detection unit 91, a liquid puddle detection unit 92, and a liquid drip detection unit 93.

The splash detection unit 91 detects information regarding the state of splashing during the supply of the developing solution. For example, the splash detection unit 91 includes an irradiation unit 94 and an imaging unit 95. The irradiation unit 94 is fixed to, for example, the nozzle 31, and irradiates laser light in the horizontal direction above the wafer W. The installation height of the irradiation unit 94 is set to a height that can be reached by droplets that have splashed up from the surface Wa. The imaging unit 95 acquires image data of the irradiation range of the laser light from the irradiation unit 94. When splashing occurs, the splashed droplets cause scattering of the laser light, etc., and a change occurs in the image data acquired by the imaging unit 95. For this reason, the image data acquired by the imaging unit 95 includes information regarding the state of droplet generation.

The liquid puddle detection unit 92 detects information regarding the formation state of a liquid film of the developing liquid on the surface Wa. For example, the liquid puddle detection unit 92 includes an imaging unit 96. The imaging unit 96 acquires image data of the surface Wa of the wafer W held by the holding unit 21. The image data acquired by the imaging unit 96 includes information regarding the formation state of the liquid film.

The dripping detection unit 93 detects information regarding the state of dripping of the developing solution from the nozzle 31. Dripping refers to the phenomenon in which the developing solution falls from the nozzle 31 outside of a preset developer supply period. For example, the dripping detection unit 93 includes an imaging unit 97. The imaging unit 97 acquires image data of the nozzle 31 and the area below it. The image data acquired by the imaging unit 97 includes information regarding the state of dripping.

The in-process inspection unit 90 may detect information related to the supply state of the film-forming liquid during the film-forming process. In this case, too, information related to the supply state of the film-forming liquid in the coating unit U1 can be detected by a configuration similar to the above-mentioned splash detection unit 91, puddle detection unit 92, and drip detection unit 93.

(Control device and machine learning device)
As shown in FIG. 6, the control device 100 has, as its functional configuration (hereinafter referred to as "functional modules"), a processing condition storage unit 111, a processing control unit 112, a data acquisition unit 113, a data input unit 114, and a recommended condition derivation unit 115.

The processing condition storage unit 111 stores preset processing conditions. For example, the processing condition storage unit 111 stores the development processing conditions by the processing module 14. The development processing conditions include the heat processing conditions by the heat processing unit U4 and the liquid processing conditions by the development unit U3. The liquid processing conditions by the development unit U3 include sequences such as supply of the developer, supply of the rinse liquid, and drying (shaking off and drying by rotation). In addition, the liquid processing conditions by the development unit U3 include the rotation speed of the wafer W in each sequence, the supply amount of the developer, the supply time of the developer, the supply amount of the rinse liquid, the discharge time of the rinse liquid, and the shaking off and drying time. When the developer is supplied while moving the nozzle 31 by the nozzle transfer unit 32, the liquid processing conditions by the development unit U3 may further include the movement start position, movement speed, movement end position, etc. of the nozzle 31 during the supply of the developer.

The processing condition storage unit 111 may store the film formation processing conditions by the processing modules 11, 12, and 13. The film formation processing conditions include the liquid processing conditions by the coating unit U1 and the heat processing conditions by the heat processing unit U2. The liquid processing conditions by the coating unit U1 include a sequence such as the supply of the film formation liquid. In addition, the liquid processing conditions by the coating unit U1 include the rotation speed of the wafer W, the supply amount of the film formation liquid, the supply time of the film formation liquid, etc. in each sequence.

The process control unit 112 causes the processing unit to perform substrate processing according to the process conditions stored in the process condition storage unit 111. For example, the process control unit 112 causes the processing module 14 to perform development processing according to the development processing conditions stored in the process condition storage unit 111. As an example, the process control unit 112 controls the heat processing unit U4 to perform heat processing (e.g., the above-mentioned PEB) on the wafer W after the exposure processing according to the preset heat processing conditions. The process control unit 112 then controls the development unit U3 to perform development processing on the wafer W according to the preset liquid processing conditions. The process control unit 112 then controls the heat processing unit U4 to perform heat processing (e.g., the above-mentioned PB) on the wafer W according to the preset heat processing conditions.

The process control unit 112 may cause the process modules 11, 12, and 13 to perform a film formation process according to the film formation process conditions stored in the process condition storage unit 111. As an example, the process control unit 112 controls the coating unit U1 to apply a film formation liquid to the front surface Wa of the wafer W according to preset liquid processing conditions. The process control unit 112 then controls the heat treatment unit U2 to perform a heat treatment on the wafer W according to preset heat treatment conditions.

The data acquisition unit 113 acquires performance data related to the quality of substrate processing according to processing conditions. The data acquisition unit 113 may acquire performance data including performance values of multiple items. The performance values of the multiple items may include performance values of post-processing items indicating the quality of the wafer W after substrate processing and in-processing items indicating the supply state of processing liquid during substrate processing. The performance values of the multiple items may be acquired performance data including multiple performance values of the same type. Multiple performance values of the same type refer to multiple performance values that should ideally be the same value. A specific example of multiple performance values of the same type is multiple performance values acquired at multiple locations.

For example, the data acquisition unit 113 acquires, as an example of a post-processing item, an actual value indicating the actual value of the line width of a resist pattern formed on the surface Wa of the wafer W by the development process (hereinafter referred to as the "actual line width value"). Specifically, the data acquisition unit 113 acquires the actual line width value based on information detected by the post-processing inspection unit 80. The data acquisition unit 113 may acquire the actual line width values at multiple locations on the surface Wa based on the information detected by the post-processing inspection unit 80.

The data acquisition unit 113 acquires, as an example of an in-process item, a performance value indicating the supply state of the developer during the development process. Specifically, the data acquisition unit 113 acquires performance values of the presence or absence of developer splashes, poor liquid film formation, and dripping, based on the information detected by the in-process inspection unit 90.

The data acquisition unit 113 may acquire, as an example of a post-processing item, an actual value indicating the actual value of the film thickness of a coating formed on the front surface Wa of the wafer W by a film formation process (hereinafter referred to as the "actual film thickness value"). Specifically, the data acquisition unit 113 may acquire the actual film thickness value based on information detected by the post-processing inspection unit 80. The data acquisition unit 113 may acquire the actual film thickness values at multiple locations on the front surface Wa based on the information detected by the post-processing inspection unit 80.

The data acquisition unit 113 may acquire, as an example of an in-process item, a performance value indicating the supply state of the film-forming liquid during the film-forming process. Specifically, the data acquisition unit 113 may acquire performance values regarding the presence or absence of splashing of the film-forming liquid, poor formation of the liquid film, and dripping, based on information detected by the in-process inspection unit 90.

The data input unit 114 inputs a dataset including processing conditions and performance data corresponding to the processing conditions to a model generation unit 214 (described later) of the machine learning device 200. The data input unit 114 may select a dataset to be input to the model generation unit 214 based on the performance values of the above-mentioned items being processed. For example, the data input unit 114 may exclude a dataset in which the supply state of the processing liquid is poor from the input targets to the model generation unit 214. Specific examples of a poor supply state of the processing liquid include at least one of the above-mentioned liquid splashing, poor formation of a liquid film, and dripping.

The recommended condition derivation unit 115 derives recommended processing conditions for substrate processing based on a learning model generated by the model generation unit 214 through machine learning based on multiple sets of data sets. As described below, the learning model is generated to output predictive data regarding the quality of substrate processing in response to input of processing conditions. The recommended processing conditions are processing conditions that are determined to be recommended for adoption based on the learning model and predetermined evaluation conditions of the predictive data.

For example, the recommended condition derivation unit 115 includes, as more detailed functional modules, an evaluation condition input unit 121 and a search result acquisition unit 122. The evaluation condition input unit 121 inputs the evaluation conditions of the predicted data to a condition search unit 216 (described later) of the machine learning device 200. The evaluation conditions are conditions for determining whether or not the predicted data is at an acceptable level.

The evaluation condition input unit 121 may input evaluation conditions for evaluating predicted values of multiple items to the condition search unit 216. The evaluation condition input unit 121 may input evaluation conditions including conditions related to the variability of predicted values in at least some of the multiple items to the condition search unit 216. For example, the evaluation conditions include a method for deriving an evaluation score of the predicted data and an acceptable level of the evaluation score.

As an example, the evaluation condition input unit 121 inputs evaluation conditions for evaluating the predicted values of the line width (hereinafter referred to as "predicted line width values") at multiple locations on the surface Wa to the condition search unit 216. The evaluation conditions include a calculation formula (e.g., a calculation formula for standard deviation) for the variation of the predicted line width values at at least some of the multiple locations (e.g., all locations) as an example of a method for deriving the evaluation score. The evaluation conditions include an allowable upper limit value of the variation calculated by the calculation formula as an allowable level of the evaluation score.

The evaluation condition input unit 121 may input evaluation conditions for evaluating the predicted film thickness values at multiple locations on the surface Wa to the condition search unit 216. The evaluation conditions include a calculation formula (e.g., a calculation formula for a standard deviation) for the variation of the predicted film thickness values at at least some (e.g., all) of the multiple locations as an example of a method for deriving the evaluation score. The evaluation conditions include an allowable upper limit value of the variation calculated by the calculation formula as an allowable level of the evaluation score.

The search result acquisition unit 122 acquires the recommended processing conditions derived by the condition search unit 216 and stores them in the processing condition storage unit 111. As described below, the recommended processing conditions are derived based on multiple sets of data sets, a learning model, and evaluation conditions input by the evaluation condition input unit 121.

Here, the processing control unit 112 may cause the processing unit to further perform substrate processing according to the recommended processing conditions. The data acquisition unit 113 may further acquire additional performance data regarding the quality of the substrate processing according to the recommended processing conditions. The data input unit 114 may further input an additional data set including the recommended processing conditions and the additional performance data to the model generation unit 214. The recommended condition derivation unit 115 may update the recommended processing conditions based on the learning model updated by the model generation unit 214 based on the additional data set. Updating the learning model means generating a new learning model based on multiple sets of data sets to which the additional data set has been added. Updating the recommended processing conditions means deriving new recommended processing conditions based on the learning model updated by the model generation unit 214.

In this case, the control device 100 may further include a condition evaluation unit 116 and a repetition management unit 117. The condition evaluation unit 116 evaluates whether or not the recommended processing conditions are adoptable. The repetition management unit 117 repeats at least the following steps until the evaluation result by the condition evaluation unit 116 indicates that the recommended processing conditions are adoptable.
i) The process control unit 112 causes the processing unit to further perform substrate processing in accordance with the recommended processing conditions.
ii) The data acquisition unit 113 further acquires additional performance data.
iii) The data input component 114 further inputs additional data sets to the model generation component 214 .
iv) The recommended condition derivation unit 115 updates the recommended processing conditions based on the learning model updated by the model generation unit 214 based on the additional data set.

There are no particular restrictions on the method of evaluating the recommended processing conditions by the condition evaluation unit 116. For example, the condition evaluation unit 116 evaluates whether or not to adopt the recommended processing conditions based on the evaluation results of the additional performance data based on predetermined evaluation conditions. The evaluation conditions may be the same as the evaluation conditions of the prediction data described above. For example, the evaluation conditions include a method of deriving an evaluation score for the additional performance data and an acceptable level of the evaluation score.

As an example, the condition evaluation unit 116 evaluates the line width actual values at multiple locations on the surface Wa based on predetermined evaluation conditions. The evaluation conditions include a calculation formula (e.g., a calculation formula for standard deviation) for the variation of the line width actual values at at least some of the multiple locations (e.g., all locations) as an example of a method for deriving the evaluation score. The evaluation conditions include an allowable upper limit value of the variation calculated by the calculation formula as an allowable level of the evaluation score.

The evaluation condition input unit 121 may evaluate the actual film thickness values at multiple locations on the surface Wa based on predetermined evaluation conditions. The evaluation conditions include a calculation formula (e.g., a calculation formula for a standard deviation) for the variation of the actual film thickness values at at least some (e.g., all) of the multiple locations as an example of a method for deriving the evaluation score. The evaluation conditions include an allowable upper limit value of the variation calculated by the calculation formula as an allowable level of the evaluation score.

The condition evaluation unit 116 may evaluate whether or not to adopt the latest recommended processing conditions based on whether the difference between the latest recommended processing conditions and past recommended processing conditions (e.g., the previous recommended processing conditions) is at an acceptable level. It is expected that the recommended processing conditions will gradually converge to one condition through the iterative processing by the iterative management unit 117. By reducing the difference between the latest recommended processing conditions and past recommended processing conditions to an acceptable level, it is possible to adopt recommended processing conditions that are close to the convergence result.

The condition evaluation unit 116 may evaluate whether or not to adopt the latest recommended processing conditions based on whether or not a difference between the latest added performance data and the previously added performance data is at an acceptable level. The condition evaluation unit 116 may evaluate whether or not to adopt the latest recommended processing conditions based on whether or not a difference between the evaluation score of the latest added performance data and the evaluation score of the previously added performance data is at an acceptable level.

The control device 100 may further include an actual data correction unit 118. Before the data input unit 114 inputs the data set to the model generation unit 214, the actual data correction unit 118 removes components attributable to factors other than the substrate processing by the processing unit of the coating/developing apparatus 2 from the actual data of the data set. For example, the actual data correction unit 118 removes variation components attributable to the exposure process from the line width actual values at the above-mentioned multiple locations. Specifically, the actual data correction unit 118 removes a variation pattern specific to the exposure process that has been investigated in advance from the line width actual values at the multiple locations.

The machine learning device 200 has, as functional modules, a search calculation unit 211, a data acquisition unit 212, a data storage unit 213, a model generation unit 214, a model storage unit 215, and a condition search unit 216. The search calculation unit 211 is a machine learning engine in the machine learning device 200. For example, the search calculation unit 211 searches for a solution using a genetic algorithm based on preset learning conditions. The learning conditions include first generation individuals, a method for deriving the evaluation scores of the individuals, and an acceptable level of the evaluation score.

The search calculation unit 211 acquires multiple individuals of the first generation and calculates an evaluation score for each individual. The search calculation unit 211 then selects individuals whose evaluation scores are far from the acceptable level, and evolves the multiple individuals into multiple individuals of the next generation through calculations such as crossover, inversion, and mutation. Thereafter, the search calculation unit 211 repeats the derivation of individual evaluation scores, the selection of individuals, and the evolution of individuals to derive individuals whose evaluation scores are at an acceptable level.

The data acquisition unit 212 acquires the above dataset and the additional dataset from the data input unit 114. The data storage unit 213 accumulates the dataset acquired by the data acquisition unit 212 as a database for learning.

The model generation unit 214 generates the learning model by machine learning based on multiple sets of data sets accumulated by the data storage unit 213. The model generation unit 214 may generate the learning model by machine learning including a calculation process of searching for the learning model by a genetic program. For example, the model generation unit 214 generates a learning model including multiple model formulas that each output predicted values of multiple items in response to input of processing conditions. In generating each model formula, the model generation unit 214 sets the learning conditions for deriving the model formula and requests the search calculation unit 211 to derive a model formula according to the learning conditions.

For example, the model generation unit 214 generates a plurality of provisional model formulas that generate predicted values in response to input of processing conditions, and these are regarded as the above-mentioned plurality of individuals of the first generation. The provisional model formulas are mathematical expressions expressed in a tree structure with various operators and random numerical values as elements. The model generation unit 214 regards a deviation score indicating the deviation between the predicted value based on the provisional model formula and the actual value as an evaluation score under the above-mentioned learning conditions, and determines a derivation method thereof. For example, the model generation unit 214 determines a derivation method including at least the following procedure.
a1) Inputting processing conditions for a plurality of sets of data sets into a provisional model formula to derive a plurality of predicted values.
a2) deriving a deviation score that indicates the deviation between a plurality of predicted values and actual values of a plurality of data sets;

The deviation score may be any value as long as it indicates the deviation between multiple predicted values and the actual values of multiple sets of data sets. Specific examples of deviation scores include the sum of squares of the differences between the predicted values and the actual values, or the square root of the sum of squares. The model generation unit 214 sets a preset upper limit value for the deviation score as the acceptable level of the evaluation score under the above learning conditions.

The search calculation unit 211 derives a model formula whose deviation score is equal to or less than an upper limit by repeatedly deriving the deviation score of the provisional model formula, selecting the provisional model formula, and evolving the provisional model formula. The model generation unit 214 acquires the model formula derived by the search calculation unit 211 and stores it in the model holding unit 215. By the above procedure, the model generation unit 214 stores each model formula in the model holding unit 215, and a learning model including multiple model formulas is generated in the model holding unit 215.

The condition search unit 216 derives recommended processing conditions based on multiple sets of data sets stored in the data storage unit 213, the learning model stored in the model storage unit 215, and the evaluation conditions input by the evaluation condition input unit 121. The condition search unit 216 may derive recommended processing conditions by a search process including a calculation process for searching for recommended processing conditions by a genetic algorithm. For example, the condition search unit 216 sets the above-mentioned learning conditions for deriving recommended processing conditions, and requests the search calculation unit 211 to derive recommended processing conditions according to the learning conditions.

For example, the condition search unit 216 treats the processing conditions of multiple sets of data sets stored in the data storage unit 213 as multiple individuals of the first generation. Each processing condition is a tree structure representing multiple conditions.

The condition searching unit 216 determines a method for deriving an evaluation score under the above learning conditions so as to include at least the following steps.
b1) Inputting processing conditions for multiple sets of data sets into a learning model stored in the model holding unit 215 to derive prediction data.
b2) Deriving an evaluation score for the predicted data in accordance with the derivation method in the evaluation conditions input by the evaluation condition input unit 121.

The condition search unit 216 sets the tolerance level for the evaluation conditions input by the evaluation condition input unit 121 as the tolerance level for the evaluation score under the above learning conditions.

The search calculation unit 211 derives the evaluation scores of the processing conditions, selects the processing conditions, and evolves the processing conditions repeatedly to derive recommended processing conditions with evaluation scores at an acceptable level. The condition search unit 216 acquires the recommended processing conditions derived by the search calculation unit 211 and outputs them to the search result acquisition unit 122.

7 is a block diagram illustrating the hardware configuration of the control device 100 and the machine learning device 200. The control device 100 includes a circuit 190. The circuit 190 includes at least one processor 191, a memory 192, a storage 193, a display device 194, an input device 195, an input/output port 196, and a communication port 197. The storage 193 is a non-volatile storage medium (e.g., a flash memory) that can be read by a computer. For example, the storage 193 stores a program for causing the control device 100 to execute the following: causing the coating/developing device 2 to perform substrate processing according to preset processing conditions; acquiring performance data regarding the quality of the substrate processing according to the processing conditions from the quality inspection device 70; inputting a data set including the processing conditions of the substrate processing and the performance data of the substrate processing into the machine learning device 200; and deriving recommended processing conditions for the substrate processing based on the learning model generated by the machine learning device 200 based on multiple sets of data sets. For example, the storage 193 includes a storage area for storing programs for configuring the above-mentioned functional modules, and a storage area allocated to the processing condition holding unit 111 .

The display device 194 is used for displaying recommended processing conditions, etc. The display device 194 and the input device 195 function as a user interface for the control device 100. The display device 194 includes, for example, an LCD monitor, and is used to display information to the user. The input device 195 is, for example, a keyboard, and acquires information input by the user. The display device 194 and the input device 195 may be integrated as a so-called touch panel. The input device 195 is used for inputting processing conditions and evaluation conditions, etc.

The memory 192 temporarily stores the programs loaded from the storage 193 and the results of calculations by the processor 191. The processor 191 executes the programs in cooperation with the memory 192 to control the coating and developing apparatus 2. The input/output port 196 inputs and outputs electrical signals between the display device 194 and the input device 195 in response to commands from the processor 191. The communication port 197 performs network communication with the machine learning device 200 in response to commands from the processor 191.

The machine learning device 200 includes a circuit 290. The circuit 290 includes a processor 291, a memory 292, a storage 293, and a communication port 294. The storage 293 is a non-volatile storage medium (e.g., a flash memory) that is readable by a computer. For example, the storage 293 stores a program for causing the machine learning device 200 to acquire the above dataset and generate the above learning model by machine learning based on a plurality of sets of datasets. For example, the storage 293 includes a storage area for storing a program for configuring the above functional module, and a storage area allocated to the data holding unit 213 and the model holding unit 215.

The memory 292 temporarily stores the program loaded from the storage 293 and the results of calculations by the processor 291. The processor 291 executes the program in cooperation with the memory 292 to generate the learning model. The communication port 294 performs network communication with the control device 100 in response to a command from the processor 291.

[Condition setting support procedure]
Next, as an example of a condition setting support method, a condition setting support procedure executed by each of the control device 100 and the machine learning device 200 will be described. The condition setting support procedure executed by the control device 100 includes a procedure for deriving recommended processing conditions and a procedure for brushing up the recommended processing conditions. The condition setting support procedure executed by the machine learning device 200 includes a procedure for generating a learning model and a procedure for searching for recommended processing conditions. Each procedure will be specifically exemplified below.

(Procedure for deriving recommended processing conditions)
The procedure for deriving the recommended processing conditions by the control device 100 includes causing the coating/developing device 2 to perform substrate processing including supplying a processing liquid to the wafer W according to preset processing conditions, acquiring performance data on the quality of the substrate processing according to the processing conditions, inputting a data set including the processing conditions of the substrate processing and the performance data of the substrate processing to the machine learning device 200, and deriving the recommended processing conditions based on the learning model generated by the machine learning device 200 based on the multiple sets of data sets. Deriving the recommended processing conditions may include inputting evaluation conditions of the prediction data to the machine learning device 200, and acquiring the recommended processing conditions derived by the machine learning device 200 based on the multiple sets of data sets, the learning model, and the evaluation conditions.

8, the control device 100 first executes steps S01, S02, and S03. In step S01, the process control unit 112 causes the coating/developing device 2 to start substrate processing according to the process conditions stored in the process condition storage unit 111. In step S02, the data acquisition unit 113 acquires the performance values of the above-mentioned processing items. The data acquisition unit 113 may acquire performance values of multiple processing items. For example, the data acquisition unit 113 acquires performance values of the presence or absence of splashing of the developing solution, poor formation of the liquid film, and dripping based on information detected by the processing inspection unit 90. The data acquisition unit 113 may acquire performance values of the presence or absence of splashing of the film-forming solution, poor formation of the liquid film, and dripping based on information detected by the processing inspection unit 90. In step S03, the process control unit 112 checks whether the substrate processing according to the processing conditions has been completed.

If it is determined in step S03 that the substrate processing is not complete, the control device 100 returns the process to step S02. Thereafter, acquisition of the performance values of the items being processed continues until the substrate processing is complete. If it is determined in step S03 that the substrate processing is complete, the control device 100 executes step S04. In step S04, the data input unit 114 checks whether any defects have occurred in the supply state of the processing liquid based on the performance values of the items being processed.

If it is determined in step S04 that no defect has occurred in the supply state of the processing liquid, the control device 100 executes steps S05, S06, and S07. In step S05, the data acquisition unit 113 acquires the actual value of the post-processing item. The data acquisition unit 113 may acquire the actual values of multiple post-processing items. For example, the data acquisition unit 113 acquires the line width actual value at multiple locations on the surface Wa based on information detected by the post-processing inspection unit 80. The data acquisition unit 113 may acquire the film thickness actual value at multiple locations on the surface Wa based on information detected by the post-processing inspection unit 80. In step S06, the actual data correction unit 118 removes components caused by factors other than the substrate processing from the actual values of the multiple post-processing items. In step S07, the data input unit 114 inputs a data set including the processing conditions and the actual data (actual values of multiple post-processing items) corresponding to the processing conditions to the machine learning device 200.

Next, the control device 100 executes step S08. If it is determined in step S04 that a problem has occurred in the supply state of the processing liquid, the control device 100 executes step S08 without executing steps S05, S06, and S07. In step S08, the data input unit 114 checks whether the input of the number of data sets required for machine learning in the machine learning device 200 has been completed.

If it is determined in step S08 that the input of the number of data sets required for machine learning has not been completed, the control device 100 executes step S09. In step S09, the process control unit 112 changes the process conditions. For example, the process control unit 112 changes the process conditions based on user input to the input device 195, etc. The control device 100 then returns the process to step S01. Thereafter, the change of process conditions, the execution of substrate processing, and the input of data sets are repeated until the input of the number of data sets required for machine learning has been completed.

In step S08, if it is determined that the input of the number of data sets required for machine learning has been completed, the control device 100 executes steps S11, S12, S13, and S14. In step S11, the evaluation condition input unit 121 waits for a learning completion notification from the machine learning device 200. In step S12, the evaluation condition input unit 121 sets the evaluation conditions for the predicted data. For example, the evaluation condition input unit 121 sets the evaluation conditions for the predicted data based on a user's input to the input device 195, etc. In step S13, the evaluation condition input unit 121 inputs the evaluation conditions set in step S12 to the machine learning device 200. In step S14, the search result acquisition unit 122 acquires the recommended processing conditions derived by the machine learning device 200 based on multiple sets of data sets, the learning model, and the evaluation conditions input by the evaluation condition input unit 121, and stores them in the processing module 11. This completes the procedure for deriving the recommended processing conditions.

(Procedure for improving recommended processing conditions)
The procedure for brushing up the recommended processing conditions by the control device 100 includes having the coating/developing apparatus 2 further perform substrate processing according to the recommended processing conditions, further acquiring additional performance data related to the quality of the substrate processing according to the recommended processing conditions, further inputting an additional data set including the recommended processing conditions and the additional performance data to the machine learning device 200, and updating the recommended processing conditions based on a learning model updated by the machine learning device 200 based on the additional data set. This brushing up procedure may further include evaluating the recommended processing conditions, and may repeat the steps of having the coating/developing apparatus 2 further perform substrate processing according to the recommended processing conditions, further acquiring additional performance data, further inputting the additional data set to the machine learning device 200, and updating the recommended processing conditions based on a learning model updated by the machine learning device 200 based on the additional data set until the evaluation result of the recommended processing conditions reaches a predetermined level.

As illustrated in FIG. 9, the control device 100 first executes steps S21, S22, S23, S24, and S25. In step S21, the process control unit 112 causes the coating/developing device 2 to perform substrate processing according to the recommended processing conditions stored in the processing condition storage unit 111. In step S22, the data acquisition unit 113 acquires additional performance values for the above-mentioned post-processing items. The data acquisition unit 113 may acquire additional performance values for multiple post-processing items. In step S23, the performance data correction unit 118 removes components caused by factors other than the substrate processing from the additional performance values for the multiple post-processing items. In step S24, the condition evaluation unit 116 evaluates the recommended processing conditions. In step S25, the repetition management unit 117 checks whether the recommended processing conditions can be adopted based on the evaluation results in step S24.

If it is determined in step S25 that the recommended processing conditions cannot be adopted, the control device 100 executes steps S26, S27, and S28. In step S26, the data input unit 114 inputs an additional data set including processing conditions and additional performance data (additional performance values of multiple post-processing items) corresponding to the processing conditions to the machine learning device 200. In step S27, the search result acquisition unit 122 waits for a notification of completion of the update of the learning model from the machine learning device 200. In step S28, the search result acquisition unit 122 acquires the recommended processing conditions updated by the machine learning device 200 based on the additional data set and stores them in the processing module 11. After that, the control device 100 returns the process to step S21. Thereafter, the acquisition of additional performance data and the update of the recommended processing conditions are repeated until the recommended processing conditions are adoptable.

If it is determined in step S25 that the recommended processing conditions can be adopted, the control device 100 completes the process. This completes the procedure for brushing up the recommended processing conditions.

(Procedure for generating learning model)
The procedure for generating a learning model by the machine learning device 200 includes acquiring the above-mentioned dataset and generating a learning model by machine learning based on a plurality of sets of datasets. Generating a learning model by machine learning may include a calculation process for searching for a learning model by a genetic program. A learning model including a plurality of model formulas each outputting a predicted value of a plurality of items in response to an input of a processing condition may be generated.

As illustrated in FIG. 10, the machine learning device 200 first executes steps S31, S32, and S33. In step S31, the data acquisition unit 212 waits for input of a dataset from the data input unit 114. In step S32, the data acquisition unit 212 accumulates the input dataset in the data holding unit 213. In step S33, the data holding unit 213 checks whether the number of datasets accumulated in the data holding unit 213 has reached the number required for machine learning.

If it is determined in step S33 that the number of accumulated datasets has not reached the number required for machine learning, the control device 100 returns the process to step S31. Thereafter, acquisition of datasets is repeated until the number of datasets required for machine learning is accumulated.

In step S33, if it is determined that the number of accumulated data sets has reached the number required for machine learning, the control device 100 executes steps S34, S35, and S36. In step S34, the model generation unit 214 sets the learning conditions for deriving a model formula corresponding to any of the predicted values, and requests the search operation unit 211 to derive a model formula according to the learning conditions. For example, the model generation unit 214 generates a plurality of provisional model formulas that generate predicted values according to the input of the processing conditions, and these are regarded as the plurality of individuals of the first generation. In addition, the model generation unit 214 determines a derivation method using the deviation score as an evaluation score, and sets the upper limit value of the deviation score as the allowable level of the evaluation score. In step S35, the search operation unit 211 calculates the deviation score of each provisional model formula according to the learning conditions. In step S36, the search operation unit 211 checks whether or not there is a provisional model formula whose deviation score is equal to or less than the upper limit value according to the learning conditions.

If it is determined in step S36 that there is no provisional model formula whose deviation score is equal to or less than the upper limit, the machine learning device 200 executes step S37. In step S37, the search calculation unit 211 selects provisional model formulas whose deviation scores greatly exceed the upper limit, and evolves the provisional model formulas into provisional model formulas of the next generation by calculations such as crossover, inversion, and mutation. After that, the machine learning device 200 returns the process to step S35. Thereafter, the derivation of the deviation scores of the provisional model formulas, the selection of the provisional model formulas, and the evolution of the provisional model formulas are repeated until a provisional model formula whose deviation score is equal to or less than the upper limit is derived.

If it is determined in step S36 that there is a provisional model formula with a deviation score equal to or less than the upper limit, the machine learning device 200 executes steps S38 and S39. In step S38, the search calculation unit 211 selects the provisional model formula with the best (smallest) deviation score, and stores this in the model holding unit 215 as one model formula of the learning model. In step S39, the model generation unit 214 checks whether the derivation of all model formulas for constituting the learning model (i.e., all model formulas required to derive predicted values for multiple items) has been completed.

If it is determined in step S39 that the derivation of all model formulas has not been completed, the machine learning device 200 executes step S41. In step S41, the model generation unit 214 changes the model formula to be derived. In other words, the model generation unit 214 changes the item to be predicted by the model formula. The machine learning device 200 then returns the process to step S34. Thereafter, the setting of learning conditions and the derivation of model formulas based on the conditions are repeated until the derivation of all model formulas is completed.

If it is determined in step S39 that the derivation of all model formulas has been completed, the machine learning device 200 completes the generation of the learning model. This completes the procedure for generating the learning model.

(Procedure for searching recommended processing conditions)
The procedure for searching for recommended processing conditions by the machine learning device 200 includes deriving recommended processing conditions for substrate processing based on a plurality of sets of data sets, a learning model, and evaluation conditions for predicted data. Deriving the recommended processing conditions may include a calculation process for searching for recommended processing conditions using a genetic algorithm. The recommended processing conditions may be derived based on a plurality of sets of data sets, a plurality of model formulas, and evaluation conditions for evaluating predicted values of a plurality of items. For example, the recommended processing conditions may be derived based on evaluation conditions including conditions related to the variability of predicted values of a plurality of items.

As illustrated in FIG. 11, the machine learning device 200 first executes steps S51 and S52. In step S51, the condition search unit 216 waits for input of evaluation conditions from the evaluation condition input unit 121. In step S52, the condition search unit 216 sets the above-mentioned learning conditions for deriving recommended processing conditions, and requests the search calculation unit 211 to derive recommended processing conditions according to the learning conditions. For example, the condition search unit 216 sets the processing conditions of multiple sets of data sets stored in the data storage unit 213 as multiple individuals of the first generation. In addition, the condition search unit 216 determines a method for deriving an evaluation score and an acceptable level of the evaluation score based on the evaluation conditions input by the evaluation condition input unit 121.

Next, the machine learning device 200 executes steps S53, S54, and S55. In step S53, the search calculation unit 211 inputs each processing condition into the learning model stored in the model holding unit 215 to derive predicted data. In step S54, the search calculation unit 211 derives an evaluation score for the predicted data. In step S55, the search calculation unit 211 checks whether there is a processing condition for which the evaluation score is at an acceptable level.

If it is determined in step S55 that there is no processing condition whose evaluation score is at an acceptable level, the machine learning device 200 executes step S56. In step S56, the search operation unit 211 evolves the multiple processing conditions into multiple processing conditions of the next generation by operations such as crossover, inversion, and mutation, while selecting processing conditions whose evaluation scores are far from the acceptable level. The machine learning device 200 then returns the process to step S53. Thereafter, the derivation of the evaluation scores of the processing conditions, the selection of the processing conditions, and the evolution of the processing conditions are repeated until a processing condition whose evaluation score is at an acceptable level is derived.

If it is determined in step S55 that there is a processing condition with an acceptable evaluation score, the machine learning device 200 executes steps S57 and S58. In step S57, the search calculation unit 211 sets the processing condition with the best evaluation score as the recommended processing condition. In step S58, the condition search unit 216 acquires the recommended processing condition derived by the search calculation unit 211 and outputs it to the search result acquisition unit 122. This completes the search procedure for recommended processing conditions.

Deriving the recommended processing conditions is not limited to the computational process of searching for the recommended processing conditions using the genetic algorithm described above. For example, in step S55, the recommended processing conditions can be derived by a computational process of repeatedly changing the processing conditions and deriving the evaluation score until the evaluation score reaches an acceptable level.

[Specific examples]
As an example, a procedure for supporting the setting of processing conditions for the development process in the development unit U3 will be specifically illustrated. The processing conditions for the development process in the development unit U3 include, for example, the rotation speed of the wafer W, the supply amount of the developing solution, the supply time of the developing solution, the supply amount of the rinsing solution, the discharge time of the rinsing solution, the shake-off drying time, the movement start position of the nozzle 31, the movement speed of the nozzle 31, and the movement end position of the nozzle 31. Among these, items requiring recommended processing conditions are, for example, the rotation speed of the wafer W and the movement speed of the nozzle 31 during the supply of the developing solution. In this case, in the above steps S01 to S09, inputting a data set into the machine learning device 200 is repeated while changing the rotation speed of the wafer W and the movement speed of the nozzle 31.

For example, in steps S01 to S09, the movement speed of the nozzle 31 is set to 15 mm/s, 20 mm/s, and 25 mm/s when the rotation speed of the wafer W is 200 rpm, then the movement speed of the nozzle 31 is set to 15 mm/s, 20 mm/s, and 25 mm/s when the rotation speed of the wafer W is 250 rpm, and then the movement speed of the nozzle 31 is set to 15 mm/s, 20 mm/s, and 25 mm/s when the rotation speed of the wafer W is 300 rpm. In step S04, which is executed under any of these processing conditions, if it is determined that a defect has occurred in the supply state of the processing liquid, the data set corresponding to that processing condition is excluded from the input target to the machine learning device 200. In this case, in order to obtain the number of data sets required for machine learning, the processing conditions are further changed in step S09. For example, if it is determined that splashing of developer has occurred under processing conditions of a rotation speed of 300 rpm and a movement speed of 25 mm/s, the rotation speed is changed to 290 rpm, and performance data is again obtained under conditions of a rotation speed of 290 rpm and a movement speed of 25 mm/s.

In step S05, for example, the average value of the line width in each of the divided regions of the wafer W divided into n locations is acquired as n line width actual values. An example of the data set in this case is shown below.
Processing conditions: rotation speed of wafer W = 200 rpm, nozzle movement speed = 15 mm/s
Actual data: W1 = 23 nm, W2 = 28 nm, W3 = 31 nm, ... Wn = 24 nm (Wi: average line width in divided region i)

The learning model generated in the machine learning device 200 based on this data set outputs a predicted value of the line width average value in n divided regions, for example, in response to inputs of the rotation speed of the wafer W and the movement speed of the nozzle. In step S12, for example, a calculation formula for the standard deviation of the n line width prediction values is set as a calculation formula for the evaluation score, and a tolerance value of the standard deviation is set as the tolerance level. Based on the evaluation conditions set in this way, the machine learning device 200 derives recommended values of the rotation speed of the wafer W and the movement speed of the nozzle 31 (for example, the rotation speed of the wafer W = 234 rpm, the movement speed of the nozzle 31 = 22 rpm) as the recommended processing conditions.

[Effects of this embodiment]
As described above, the method for supporting condition setting for substrate processing according to this embodiment includes inputting a dataset including processing conditions for substrate processing performed by the coating/developing apparatus 2, including the supply of processing liquid to the wafer W, and performance data related to the quality of the substrate processing, into the machine learning device 200, and deriving recommended processing conditions for the substrate processing based on a learning model generated by the machine learning device 200 through machine learning based on multiple sets of datasets, the learning model outputting predictive data related to the quality of the substrate processing in response to the input of processing conditions.

This condition setting support method derives recommended processing conditions based on a learning model generated by machine learning, making it possible to efficiently search for appropriate processing conditions. This is therefore effective in simplifying the task of setting processing conditions for substrate processing.

The method for supporting substrate processing condition setting may include causing the coating/developing apparatus 2 to further perform substrate processing according to the recommended processing conditions, further acquiring additional performance data related to the quality of the substrate processing according to the recommended processing conditions, further inputting an additional data set including the recommended processing conditions and the additional performance data into the machine learning device 200, and updating the recommended processing conditions based on a learning model updated by the machine learning device 200 based on the additional data set. In this case, the recommended processing conditions are updated based on feedback of the recommended processing conditions and the additional performance data. Therefore, more appropriate processing conditions can be efficiently searched for.

The method for supporting the setting of conditions for substrate processing may further include evaluating the recommended processing conditions, and may repeat the steps of causing the coating/developing apparatus 2 to further perform substrate processing in accordance with the recommended processing conditions, further acquiring additional performance data, further inputting the additional data set into the machine learning device 200, and updating the recommended processing conditions based on the learning model updated by the machine learning device 200 based on the additional data set, until the evaluation result of the recommended processing conditions reaches a predetermined level. In this case, more appropriate processing conditions can be efficiently searched for by the repeated processing.

Deriving the recommended processing conditions may include inputting evaluation conditions for the predicted data into the machine learning device 200, and obtaining the recommended processing conditions derived by the machine learning device 200 based on the multiple sets of data sets, the learning model, and the evaluation conditions. In this case, the machine learning device 200 also searches for the recommended processing conditions, so that appropriate processing conditions can be searched for more efficiently.

In the method for supporting condition setting for substrate processing, a data set may be input to a machine learning device 200 that acquires performance data including performance values of multiple items, generates a learning model including multiple model formulas that each output a predicted value of the multiple items in response to input of processing conditions, and inputs evaluation conditions for evaluating the predicted values of the multiple items to the machine learning device 200. In this case, by expanding the evaluation conditions to multiple items, the quality of the processing can be more appropriately evaluated and more appropriate processing conditions can be searched for.

In the substrate processing condition setting support method, evaluation conditions including conditions regarding the variability of predicted values for at least some of the multiple items may be input to the machine learning device 200. In this case, multiple items can be efficiently evaluated, making it possible to efficiently search for more appropriate processing conditions.

In the substrate processing condition setting support method, performance data including performance values of post-processing items indicating the quality of the wafer W after substrate processing and in-processing items indicating the supply state of the processing liquid during the substrate processing may be acquired, and a data set to be input to the machine learning device 200 may be selected based on the performance values of the in-processing items. In this case, by directly capturing abnormalities during processing as in-processing data, the search range for recommended processing conditions based on the post-processing quality can be narrowed. Therefore, appropriate processing conditions can be searched for more efficiently.

The method for supporting the setting of substrate processing conditions may further include excluding components attributable to factors other than the substrate processing from the performance data of the data set before inputting the data set into the machine learning device 200. In this case, more appropriate processing conditions can be searched for.

The substrate processing may include a development process in which a developer is supplied to the photosensitive film on the surface Wa of the wafer W that has been subjected to an exposure process, and performance data including the performance value of the line width of the pattern formed on the surface Wa of the wafer W by the development process may be acquired. When the substrate processing includes a development process, it tends to require a great deal of effort to derive suitable processing conditions. For this reason, the above-mentioned condition setting support method makes it possible to efficiently search for suitable processing conditions, and is highly effective.

The substrate processing may include a film formation process in which a film formation liquid is applied to the surface Wa of the wafer W to form a coating, and performance data may be acquired that includes the performance value of the film thickness of the coating formed on the surface Wa of the wafer W by the film formation process. Even when the substrate processing includes a film formation process, the quality of the substrate processing is very sensitive to the processing conditions, so it tends to require a great deal of effort to derive suitable processing conditions. For this reason, the above-mentioned condition setting support method makes it possible to efficiently search for suitable processing conditions, and is highly effective.

Although the embodiments have been described above, the present disclosure is not necessarily limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure. For example, the substrate to be processed is not limited to a semiconductor wafer, but may be, for example, a glass substrate, a mask substrate, or a flat panel display (FPD).

2...coating/developing apparatus (substrate processing apparatus), 11, 12, 13, 14...processing module (processing section), 112...processing control section, 113...data acquisition section, 114...data input section, 115...recommended condition derivation section, 121...evaluation condition input section, 122...search result acquisition section, 214...model generation section, W...wafer, Wa...surface.

Claims (28)

acquiring in-process performance data indicating a supply state of the developer during a development process in which the developer is supplied to the photosensitive film on the surface of the substrate that has been subjected to an exposure process;
acquiring post-processing performance data including a performance value of a line width of a pattern formed on the surface of the substrate by the developing process;
determining whether the supply state is poor based on the in-processing performance data, and when it is determined that the supply state is not poor , inputting a data set including processing conditions of the development processing and the post-processing performance data into a machine learning device;
deriving recommended processing conditions for the development processing based on a learning model generated by the machine learning device through machine learning based on a plurality of sets of the data sets, the learning model outputting predicted data of the line width in response to input of the processing conditions ;
The poor supply state of the developer includes a case where the developer splashes on the substrate,
acquiring the in-process performance data based on an image of the substrate and the developing solution captured by an imaging unit;
acquiring the post-processing performance data based on a detection result by a sensor different from the imaging unit;
A method for supporting setting conditions for substrate processing. 2. The method for supporting setting of substrate processing conditions according to claim 1, wherein the processing performance data is acquired based on the captured image including a reachable height of droplets of the developer splashed up from the substrate. 3. The method for supporting setting conditions for substrate processing according to claim 1, wherein the poor supply state of the developer further includes at least one of poor formation of a liquid film of the developer and dripping of the developer . acquiring the in-process performance data based on an image of the substrate and the developing solution captured by an imaging unit;
The method for supporting setting of substrate processing conditions according to claim 1 , further comprising the step of acquiring the post-processing performance data based on a detection result obtained by a sensor other than the imaging unit. further acquiring additional performance data including a performance value of the line width by the development processing executed according to the recommended processing conditions;
further inputting an additional data set into the machine learning device, the additional data set including the recommended processing conditions and the additional performance data;
The substrate processing condition setting support method according to claim 1 , further comprising: updating the recommended processing conditions based on the learning model updated by the machine learning device based on the additional data set. 6. The substrate processing condition setting support method of claim 5, further comprising: repeating the steps of further acquiring the additional performance data, further inputting the additional data set into the machine learning device, updating the recommended processing conditions based on the learning model updated by the machine learning device based on the additional data set, and evaluating the recommended processing conditions based on the additional performance data until an evaluation result of the recommended processing conditions reaches a predetermined level. 7. The method for supporting setting substrate processing conditions according to claim 1 , further comprising excluding components attributable to factors other than the development processing from the post-processing performance data of the data set before inputting the data set into the machine learning device. acquiring in-process performance data indicating a supply state of the film-forming liquid during a film-forming process in which the film-forming liquid is applied to the surface of the substrate to form a film;
acquiring post-processing performance data including a performance value of a film thickness of the coating formed on the surface of the substrate by the film formation process;
determining whether the supply state is poor based on the in-process performance data, and when it is determined that the supply state is not poor , inputting a data set including processing conditions of the film formation process and the post-processing performance data into a machine learning device;
deriving recommended processing conditions for the film formation process based on a learning model generated by the machine learning device through machine learning based on a plurality of sets of the data sets, the learning model outputting predicted data of the film thickness in response to input of the processing conditions ;
The poor supply state of the film forming liquid includes the occurrence of splashing of the film forming liquid on the substrate,
acquiring the in-process performance data based on an image of the substrate and the film forming solution captured by an imaging unit;
The method for supporting setting conditions for substrate processing further comprises acquiring the post-processing performance data based on a detection result obtained by a sensor separate from the imaging unit . 9. The method for supporting setting of substrate processing conditions according to claim 8, wherein the processing performance data is acquired based on a captured image including a reachable height of droplets of the film forming liquid splashed up from the substrate. The substrate processing condition setting support method according to claim 8 or 9 , wherein the poor supply state of the film forming liquid further includes at least one of poor formation of a liquid film of the film forming liquid and dripping of the film forming liquid. acquiring the in-process performance data based on an image of the substrate and the film forming solution captured by an imaging unit;
11. The method for supporting setting of substrate processing conditions according to claim 8, further comprising acquiring the post-processing performance data based on a detection result by a sensor separate from the imaging unit. further acquiring additional performance data including a performance value of the film thickness by the film formation process performed according to the recommended processing conditions;
further inputting an additional data set into the machine learning device, the additional data set including the recommended processing conditions and the additional performance data;
The substrate processing condition setting support method according to claim 8 , further comprising: updating the recommended processing conditions based on the learning model updated by the machine learning device based on the additional data set. 13. The substrate processing condition setting support method of claim 12, further comprising: repeating the steps of further acquiring the additional performance data, further inputting the additional data set into the machine learning device, updating the recommended processing conditions based on the learning model updated by the machine learning device based on the additional data set, and evaluating the recommended processing conditions based on the additional performance data until an evaluation result of the recommended processing conditions reaches a predetermined level. The substrate processing condition setting support method according to any one of claims 8 to 13 , further comprising excluding components attributable to factors other than the film formation process from the post-processing performance data of the data set before inputting the data set into the machine learning device. deriving the recommended processing conditions
inputting an evaluation condition for the prediction data into the machine learning device;
The substrate processing condition setting support method according to any one of claims 1 to 14 , further comprising obtaining the recommended processing conditions derived by the machine learning device based on the plurality of sets of data sets, the learning model, and the evaluation conditions. acquiring the processed performance data including performance values of a plurality of items;
inputting the data set into the machine learning device that generates the learning model including a plurality of model formulas that respectively output predicted values of the plurality of items in response to the input of the processing conditions;
The method for supporting setting of substrate processing conditions according to claim 15 , further comprising inputting the evaluation conditions for evaluating the predicted values of the plurality of items to the machine learning device. The method for supporting setting conditions for substrate processing according to claim 16 , further comprising inputting the evaluation conditions, including a condition regarding a variation in predicted values in at least a part of the plurality of items, into the machine learning device. a processing section for performing a developing process by supplying a developer to the photosensitive film on the surface of the substrate that has been subjected to an exposure process;
a processing control unit that causes the processing unit to execute the developing process in accordance with preset processing conditions;
a data acquisition unit that acquires during the developing process, data indicating a supply state of a developer, and acquires post-processing data including a result value of a line width of a pattern formed on a surface of the substrate by the developing process;
a data input unit that judges whether the supply state is poor based on the in-process performance data, and inputs a data set including the processing conditions and the post-processing performance data to a model generation unit when it is judged that the supply state is not poor ;
a recommended condition derivation unit that derives recommended processing conditions for the development processing based on a learning model generated by the model generation unit through machine learning based on a plurality of sets of the data sets so as to output predicted data of the line width in response to input of the processing conditions ,
The poor supply state of the developer includes a case where the developer splashes on the substrate,
The data acquisition unit is
acquiring the in-process performance data based on an image of the substrate and the developing solution captured by an imaging unit;
The substrate processing system acquires the post-processing performance data based on a detection result obtained by a sensor separate from the imaging unit . The substrate processing system according to claim 18 , wherein the data acquisition unit acquires the in-process performance data based on the captured image including a reachable height of droplets of the developer splashed up from the substrate. 20. The substrate processing system according to claim 18 , wherein the poor supply state of the developer further includes at least one of a defective formation of a liquid film of the developer and dripping of the developer. a processing section that performs a film formation process by applying a film formation liquid to a surface of a substrate to form a film;
a process control unit that causes the processing unit to execute the film forming process in accordance with preset process conditions;
a data acquisition unit that acquires during the film formation process, data indicating a supply state of the film formation solution, and acquires post-processing data including a film thickness of the film formed on the surface of the substrate by the film formation process;
a data input unit that judges whether the supply state is poor based on the in-process performance data, and inputs a data set including the processing conditions and the post-processing performance data to a model generation unit when it is judged that the supply state is not poor ;
a recommended condition derivation unit that derives recommended processing conditions for the film formation process based on a learning model generated by the model generation unit through machine learning based on a plurality of sets of the data sets so as to output predicted data of the film thickness in response to input of the processing conditions ;
The poor supply state of the film forming liquid includes the occurrence of splashing of the film forming liquid on the substrate,
The data acquisition unit is
acquiring the in-process performance data based on an image of the substrate and the film forming solution captured by an imaging unit;
The substrate processing system acquires the post-processing performance data based on a detection result obtained by a sensor separate from the imaging unit . The substrate processing system according to claim 21 , wherein the data acquisition unit acquires the processing performance data based on the captured image including a reachable height of droplets of the film forming liquid splashed up from the substrate. 23. The substrate processing system according to claim 21 , wherein the poor supply state of the film deposition liquid further includes at least one of a poor formation of a liquid film of the film deposition liquid and dripping of the film deposition liquid. The substrate processing system according to claim 18 , further comprising the model generating unit. The recommended condition derivation unit
an evaluation condition input unit that inputs evaluation conditions for the prediction data into a condition search unit;
The substrate processing system according to any one of claims 18 to 24 , further comprising a search result acquisition unit that acquires the recommended processing conditions derived by a condition search unit based on the plurality of sets of data sets, the learning model, and the evaluation conditions. The substrate processing system according to claim 25 , further comprising the condition searching unit. A computer-readable storage medium storing a program for causing an apparatus to execute the method for supporting setting of substrate processing conditions according to any one of claims 1 to 17 . A program for causing an apparatus to execute the method for supporting setting of substrate processing conditions according to any one of claims 1 to 17 .

Images