Feature estimation method, estimation program, and feature estimation device

Abstract

To provide a feature amount estimation method, an estimation program, and a feature amount estimation apparatus for improving the estimation accuracy of a shape feature amount on a surface of a substrate.SOLUTION: The method for estimating a shape feature amount on a surface of a substrate after applying processing for semiconductor manufacturing, includes: generating a plurality of regression models representing a relation between a target parameter and the shape feature amount according to each of a plurality of pre-processing conditions having a plurality of parameters, based on actually measured data of the shape feature amount on the surface of the substrate that has been pre-processed; calculating, for each of the plurality of regression models, an index value representing the reliability of estimated results of the shape feature amount; evaluating the suitability of a setting method of the plurality of pre-processing conditions; selecting a model to be selected as a selection model from among the plurality of regression models based on calculated results of the index value and evaluated results of the suitability of the setting method; and calculating an estimated value of the shape feature amount corresponding to any value of the target parameter by using the selection model.SELECTED DRAWING: Figure 11

Description

Technical Field

[0001] The present disclosure relates to a feature amount estimation method, an estimation program, and a feature amount estimation device.

Background Art

[0002] Patent Document 1 discloses an information processing apparatus including a prediction unit configured to calculate a predicted film thickness, and an output unit configured to output instruction information regarding the processing of a substrate based on the predicted film thickness.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The present disclosure provides a feature amount estimation method, an estimation program, and a feature amount estimation device useful for improving the estimation accuracy of shape feature amounts on the surface of a substrate.

Means for Solving the Problems

[0005] A feature estimation method relating to one aspect of this disclosure is a method for estimating shape features that represent the shape characteristics of the surface of a substrate after it has been subjected to a predetermined process for semiconductor manufacturing. This feature estimation method includes: generating multiple regression models that represent the relationship between one or more target parameters included in a plurality of parameters and the shape features, based on measured data of the shape features on the surface of a substrate that has been subjected to a predetermined process according to each of a plurality of pre-processing conditions having a plurality of parameters relating to the predetermined process; calculating an index value representing the reliability of the shape feature estimation result for each of the plurality of regression models; evaluating the suitability of the method for setting the plurality of pre-processing conditions; selecting one regression model as the selected model from among the plurality of regression models based on the calculation result of the index value and the evaluation result of the suitability of the setting method; and using the selected model to calculate an estimated value of the shape features corresponding to an arbitrary value of one or more target parameters. [Effects of the Invention]

[0006] This disclosure provides a feature estimation method, estimation program, and feature estimation apparatus that are useful for improving the estimation accuracy of shape features on the surface of a substrate. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a schematic diagram showing an example of a substrate processing system. [Figure 2] Figure 2 is a schematic diagram showing an example of a coating and developing apparatus. [Figure 3] Figure 3 is a schematic diagram showing an example of a liquid processing unit. [Figure 4] Figure 4 is a flowchart showing an example of liquid processing. [Figure 5] Figure 5 is a graph showing an example of the change in rotational speed over time during liquid processing. [Figure 6] Figure 6 is a block diagram showing an example of the functional configuration of a control device. [Figure 7] Figure 7 is a block diagram showing an example of the hardware configuration of a control device. [Figure 8] Figure 8 is a flowchart showing an example of a method for generating a regression model. [Figure 9] Figure 9 is a table showing an example of several pretreatment conditions and an example of measured data. [Figure 10] Figure 10(a) is a schematic diagram showing an example of measurement points set on the surface of a workpiece. Figure 10(b) is a diagram illustrating an example of a regression model. [Figure 11] Figure 11 is a flowchart showing an example of a method for estimating film thickness. [Figure 12] Figure 12 is a graph showing various examples of setting preconditions. [Figure 13] Figure 13 is a graph illustrating an example of a method for evaluating data bias related to the setting of preconditions. [Figure 14] Figure 14 is a graph illustrating an example of a method for evaluating data bias related to the setting of preconditions. [Figure 15] Figure 15 schematically illustrates the process of selecting a regression model. [Figure 16] Figure 16 is a graph showing the verification results of the method for estimating shape features. [Modes for carrying out the invention]

[0008] An embodiment will be described below with reference to the drawings. In this description, the same elements or elements having the same function will be denoted by the same reference numeral, and redundant descriptions will be omitted.

[0009] [Circuit board processing system] The substrate processing system 1 (substrate processing apparatus) shown in FIG. 1 is a system that performs formation of a photosensitive film, exposure of the photosensitive film, and development of the photosensitive film on a workpiece W. The workpiece W to be processed is, for example, a substrate or a substrate in a state where a film and a circuit etc. are formed by performing predetermined processing. The substrate included in the workpiece W is, as an example, a wafer containing silicon. The workpiece W (substrate) may be formed in a circular shape. The workpiece W to be processed may be a glass substrate, a mask substrate, an FPD (Flat Panel Display), etc., or may be an intermediate obtained by performing predetermined processing on these substrates etc. The photosensitive film is, for example, a resist film.

[0010] The substrate processing system 1 includes a coating / developing apparatus 2, an exposure apparatus 3, and a control apparatus 100. The exposure apparatus 3 is an apparatus that exposes the resist film (photosensitive film) formed on the workpiece W (substrate). Specifically, the exposure apparatus 3 irradiates an energy beam to the exposure target portion of the resist film by a method such as immersion exposure. Hereinafter, as an example of the substrate processing apparatus, the configuration of the coating / developing apparatus 2 will be described. As shown in FIGS. 1 and 2, the coating / developing apparatus 2 includes a carrier block 4, a processing block 5, and an interface block 6.

[0011] The carrier block 4 introduces the workpiece W into the coating / developing apparatus 2 and exports the workpiece W from the coating / developing apparatus 2. For example, the carrier block 4 can support a plurality of carriers C for the workpiece W and incorporates a transfer device A1 including a transfer arm. The carrier C accommodates, for example, a plurality of circular workpieces W. The transfer device A1 takes out the workpiece W from the carrier C and delivers it to the processing block 5, receives the workpiece W from the processing block 5, and returns it into the carrier C.

[0012] Processing block 5 performs one or more substrate processes (predetermined processes) for semiconductor manufacturing on the workpiece W. The one or more substrate processes executed by processing block 5 include a process (predetermined process) of forming a film of a processing liquid such as a resist film on the surface Wa, and a process (predetermined process) of developing the resist film after exposure. Processing block 5 has a plurality of processing modules 11, 12, 13, 14.

[0013] Processing module 11 incorporates a liquid processing unit U1, a heat treatment unit U2, and a transfer device A3 that transfers the workpiece W to these units. Processing module 11 forms a lower layer film on the surface of the workpiece W by the liquid processing unit U1 and the heat treatment unit U2. The liquid processing unit U1 applies a processing liquid for forming the lower layer film onto the workpiece W. The heat treatment unit U2 performs various heat treatments associated with the formation of the lower layer film.

[0014] Processing module 12 incorporates a liquid processing unit U1, a heat treatment unit U2, and a transfer device A3 that transfers the workpiece W to these units. Processing module 12 forms a resist film on the lower layer film by the liquid processing unit U1 and the heat treatment unit U2. The liquid processing unit U1 applies a processing liquid (resist) for forming the resist film onto the lower layer film. The heat treatment unit U2 performs various heat treatments associated with the formation of the film.

[0015] Processing module 13 incorporates a liquid processing unit U1, a heat treatment unit U2, and a transfer device A3 that transfers the workpiece W to these units. Processing module 13 forms an upper layer film on the resist film by the liquid processing unit U1 and the heat treatment unit U2. The liquid processing unit U1 applies a processing liquid for forming the upper layer film onto the resist film. The heat treatment unit U2 performs various heat treatments associated with the formation of the upper layer film.

[0016] The processing module 14 incorporates a liquid processing unit U1, a heat processing unit U2, and a transport device A3 for transporting workpieces W to these units. The processing module 14 performs development processing and associated heat processing of the exposure-treated resist film using the liquid processing unit U1 and the heat processing unit U2. The liquid processing unit U1 develops the resist film by supplying a developer solution onto the surface of the exposed workpiece W and then washing it off with a rinsing solution. The heat processing unit U2 performs various heat treatments associated with the development process. Specific examples of heat treatments include post-exposure bake (PEB) and post-bake (PB).

[0017] The coating and developing apparatus 2 has a measuring unit 18. The measuring unit 18 is a unit that measures a feature quantity (hereinafter referred to as "shape feature quantity") that represents the shape characteristics of the surface Wa of a workpiece W after it has been subjected to substrate processing for semiconductor manufacturing. The substrate processing for semiconductor manufacturing (a predetermined processing) may be a processing that forms a film of processing solution, or it may be a processing that includes developing and heating processing. The shape characteristics of the surface Wa of the workpiece W correspond to the shape characteristics of the film of processing solution (e.g., a resist film) formed on the surface Wa. The shape feature quantity is, for example, the thickness (film thickness) of the processing solution film on the surface Wa of the workpiece W after it has been subjected to processing that forms a film of processing solution such as a resist film. Alternatively, the shape feature quantity is the line width of the resist film on the surface Wa of the workpiece W after it has been subjected to processing that includes developing and heating processing. In the following, the contents of this disclosure will be explained using the case where the shape feature quantity is film thickness (thickness of the resist film) as an example.

[0018] The measuring unit 18 may measure the film thickness at multiple measurement points on the surface Wa of the workpiece W. For example, the measuring unit 18 may measure the film thickness at multiple measurement points set on a line passing through the center of the workpiece W on the surface Wa of the workpiece W (see Figure 10(a)). Alternatively, the measuring unit 18 may measure the film thickness at multiple measurement points set to be scattered across the entire surface Wa of the workpiece W. In this disclosure, the set of measured film thickness values ​​at multiple measurement points on the surface Wa of the workpiece W is referred to as the "film thickness distribution". The measuring unit 18 may measure the film thickness in any manner.

[0019] A shelf unit U10 is provided on the carrier block 4 side within the processing block 5. The shelf unit U10 is divided into multiple cells arranged vertically. A transport device A7, including a lifting arm, is provided near the shelf unit U10. The transport device A7 raises and lowers the workpiece W between the cells of the shelf unit U10.

[0020] A shelf unit U11 is provided on the interface block 6 side within the processing block 5. The shelf unit U11 is divided into multiple cells arranged vertically.

[0021] Interface block 6 handles the transfer of workpieces W to and from the exposure device 3. For example, interface block 6 incorporates a transport device A8, which includes a transfer arm, and is connected to the exposure device 3. Transport device A8 transfers workpieces W, which are placed on shelf unit U11, to the exposure device 3. Transport device A8 receives workpieces W from the exposure device 3 and returns them to shelf unit U11.

[0022] (Liquid processing unit) Next, an example of the liquid processing unit U1 of the processing module 12 will be described in detail. The liquid processing unit U1 performs processing that includes discharging a processing liquid toward the surface Wa of the workpiece W while rotating the workpiece W. The liquid processing unit U1 also performs processing that includes rotating the workpiece W after the coating process to dry the film of processing liquid on the surface Wa of the workpiece W. Processing that includes discharging the processing liquid and processing that includes drying may be referred to as "unit processing". The liquid processing performed by the liquid processing unit U1 includes various unit processing. As shown in Figure 3, the liquid processing unit U1 has a rotating holding unit 20 and a processing liquid supply unit 30.

[0023] The rotating holding unit 20 holds and rotates the workpiece W based on the operation instructions of the control device 100. The rotating holding unit 20 includes, for example, a holding unit 22 and a rotational drive unit 24. The holding unit 22 supports the center of the workpiece W, which is placed horizontally with its surface Wa facing upward, and holds the workpiece W by, for example, vacuum suction. The rotational drive unit 24 is an actuator including a power source such as an electric motor, and rotates the holding unit 22 around a vertical axis Ax. As a result, the workpiece W on the holding unit 22 rotates. The holding unit 22 may hold the workpiece W such that its center substantially coincides with the axis Ax. The holding unit 22 rotates the workpiece W at a rotational speed corresponding to the operation instructions of the control device 100.

[0024] The processing liquid supply unit 30 supplies processing liquid to the surface Wa of the workpiece W by discharging the processing liquid toward the surface Wa of the workpiece W based on the operation instructions of the control device 100. The processing liquid is a solution (resist) for forming a resist film. The processing liquid supply unit 30 includes, for example, a nozzle 32, a liquid supply unit 34, a supply passage 36, and an on / off valve 38. The nozzle 32 discharges processing liquid toward the surface Wa of the workpiece W held by the holding unit 22. The nozzle 32 is positioned, for example, above the workpiece W and discharges processing liquid downwards. The liquid supply unit 34 supplies processing liquid to the nozzle 32 via the supply passage 36 based on the operation instructions of the control device 100. The liquid supply unit 34 sends the processing liquid toward the nozzle 32 by, for example, a pump.

[0025] The on / off valve 38 is provided in the supply passage 36 between the nozzle 32 and the liquid supply unit 34. Based on the operation instructions of the control device 100, the on / off valve 38 switches the open / closed state of the supply passage 36 between an open state and a closed state. The on / off valve 38 is, for example, an air-operated valve. When the on / off valve 38 receives an open command from the control device 100, it transitions the open / closed state of the supply passage 36 from a closed state to an open state. This starts the discharge of the processed liquid from the nozzle 32. When the on / off valve 38 receives a close command from the control device 100, it transitions the open / closed state of the supply passage 36 from an open state to a closed state. This stops the discharge of the processed liquid from the nozzle 32.

[0026] The control device 100 is one or more computer devices that control the coating and developing apparatus 2. An input / output device 102 may be connected to the control device 100 (see Figure 2). The input / output device 102 is a device that inputs input information indicating instructions from a user such as an operator to the control device 100, and outputs information from the control device 100 to the user. The input / output device 102 may include a keyboard, operation panel, or mouse as an input device, and may include a monitor (e.g., a liquid crystal display) as an output device. The input / output device 102 may be a touch panel that integrates the input and output devices. The control device 100 and the input / output device 102 may be integrated.

[0027] The specific configuration of the substrate processing apparatus is not limited to the configuration of the substrate processing system 1 exemplified above. The substrate processing apparatus may be any apparatus that includes a processing unit for performing substrate processing for semiconductor manufacturing and a control device capable of controlling this processing unit.

[0028] [Coating and Developing Procedure] The control device 100 controls the coating and developing apparatus 2 to perform a coating and developing process on one workpiece W, for example, in the following procedure. First, the control device 100 controls the transport apparatus A1 to transport the workpiece W in the carrier C to the shelf unit U10, and then controls the transport apparatus A7 to place this workpiece W into a cell for the processing module 11.

[0029] Next, the control device 100 controls the transport device A3 to transport the workpiece W from the shelf unit U10 to the liquid treatment unit U1 and the heat treatment unit U2 in the processing module 11. The control device 100 also controls the liquid treatment unit U1 and the heat treatment unit U2 to form an underlayer film on the surface Wa of the workpiece W. After that, the control device 100 controls the transport device A3 to return the workpiece W with the underlayer film formed on it back to the shelf unit U10, and controls the transport device A7 to place the workpiece W into a cell for the processing module 12.

[0030] Next, the control device 100 controls the transport device A3 to transport the workpiece W from the shelf unit U10 to the liquid treatment unit U1 and the heat treatment unit U2 in the processing module 12. The control device 100 also controls the liquid treatment unit U1 and the heat treatment unit U2 to form a resist film on the underlying film of the workpiece W. After that, the control device 100 controls the transport device A3 to return the workpiece W to the shelf unit U10 and controls the transport device A7 to place the workpiece W into a cell for the processing module 13.

[0031] Next, the control device 100 controls the transport device A3 to transport the workpiece W from the shelf unit U10 to each unit in the processing module 13. The control device 100 also controls the liquid processing unit U1 and the heat processing unit U2 to form an upper layer film on the resist film of the workpiece W. After that, the control device 100 controls the transport device A3 to transport the workpiece W to the shelf unit U11.

[0032] Next, the control device 100 controls the transport device A8 to send the workpiece W from the shelf unit U11 to the exposure device 3. Subsequently, the control device 100 controls the transport device A8 to receive the exposed workpiece W from the exposure device 3 and place it in the cell for the processing module 14 in the shelf unit U11.

[0033] Next, the control device 100 controls the transport device A3 to transport the workpiece W from the shelf unit U11 to each unit in the processing module 14, and controls the liquid processing unit U1 and the heat processing unit U2 to develop the resist film on the workpiece W. As the resist film is developed, a resist pattern is formed on the surface of the workpiece W.

[0034] Subsequently, the control device 100 controls the transport device A3 to return the workpiece W to the shelf unit U10, and controls the transport devices A7 and A1 to return the workpiece W to the carrier C. This completes the coating and developing process for one workpiece W. The control device 100 then causes the coating and developing device 2 to perform the same coating and developing process for each of the subsequent workpieces W.

[0035] (Liquid treatment procedure) Next, an example of a liquid processing procedure performed in the processing module 12 will be described with reference to Figures 4 and 5. In this liquid processing procedure (liquid processing method), with the workpiece W supported by the holding part 22 of the rotating holding part 20, the control device 100 executes step S01, as shown in Figure 4. In step S01, for example, the control device 100 controls the rotation drive unit 24 to change the rotation speed of the workpiece W held by the holding part 22. Figure 5 shows an example of controlling the rotation speed of the workpiece W after step S01. For example, as shown in Figure 5, the control device 100 controls the rotation drive unit 24 to accelerate the rotation of the stationary workpiece W to a rotation speed ω1. The rotation speed ω1 is determined before the liquid processing procedure is executed.

[0036] Next, the control device 100 executes step S02. In step S02, for example, the control device 100 controls the processing liquid supply unit 30 to start discharging the processing liquid (resist) from the nozzle 32. For example, the control device 100 outputs an open command to the on / off valve 38 to start discharging the processing liquid toward the surface Wa of the workpiece W which is rotating at a rotational speed ω1. Discharging of the processing liquid is started when the on / off valve 38 switches the open / closed state of the supply passage 36 from the closed state to the open state. The control device 100 may execute step S02 according to the discharge start timing determined before the execution of the liquid processing procedure.

[0037] Next, the control device 100 executes step S03. In step S03, for example, the control device 100 waits from the start of discharge of the processing liquid until a first set time ts1 has elapsed. As a result, for at least part of the first set time ts1, the workpiece W rotates at a rotational speed ω1 while the processing liquid is discharged toward the surface Wa of the workpiece W. The first set time ts1 (or the time for which discharge of the processing liquid continues) is set, for example, so that an amount of processing liquid is supplied that is sufficient to form a liquid film on the surface Wa of the workpiece W, and is determined before the execution of the liquid processing procedure.

[0038] Next (after the first set time ts1 has elapsed), the control device 100 executes step S04. In step S04, for example, the control device 100 controls the processing liquid supply unit 30 to stop the discharge of processing liquid from the nozzle 32. For example, the control device 100 outputs a closing command to the on / off valve 38 to stop the discharge of processing liquid toward the surface Wa of the workpiece W. The on / off valve 38 switches the open / closed state of the supply passage 36 from the open state to the closed state, thereby stopping the discharge of processing liquid.

[0039] Next, the control device 100 performs step S05. In step S05, for example, the control device 100 controls the rotation drive unit 24 to change the rotational speed of the workpiece W held in the holding unit 22. As shown in Figure 6, the control device 100 controls the rotation drive unit 24 to decelerate the rotation of the workpiece W from rotational speed ω1 to rotational speed ω2. The rotational speed ω2 is determined before the liquid processing procedure is performed, and its set value is different from the set value of rotational speed ω1. The rotational speed ω2 may be less than the rotational speed ω1.

[0040] Next, the control device 100 executes step S06. In step S06, for example, the control device 100 waits from the time the workpiece W starts rotating at rotational speed ω2 until a second set time ts2 has elapsed. As a result, the workpiece W rotates at rotational speed ω2 for the duration of the second set time ts2. The second set time ts2 may be different from or the same as the first set time ts1.

[0041] Next (after the second set time ts2 has elapsed), the control device 100 executes step S07. In step S07, for example, the control device 100 controls the rotation drive unit 24 to change the rotational speed of the workpiece W held in the holding unit 22. As shown in Figure 6, the control device 100 controls the rotation drive unit 24 to accelerate the rotation of the workpiece W from rotational speed ω2 to rotational speed ω3. The rotational speed ω3 is determined before the execution of the liquid processing procedure and may be the same as or different from rotational speed ω1.

[0042] Next, the control device 100 executes step S08. In step S08, for example, the control device 100 waits from the time the workpiece W starts rotating at rotational speed ω3 until a third set time ts3 has elapsed. As a result, the workpiece W rotates at rotational speed ω3 for the duration of the third set time ts3. The third set time ts3 is set to the extent that a film of the processing liquid supplied to the workpiece W forms on the surface Wa, and is determined before the execution of the liquid processing procedure.

[0043] When the third set time ts3 has elapsed, for example, the control device 100 controls the rotation holding unit 20 to stop the rotation of the workpiece W. With this, the series of liquid treatment procedures for the workpiece W to be processed is completed. After the execution of this liquid treatment procedure, the control device 100 controls the transport device A3 to transport the workpiece W to be processed to the heat treatment unit U2. Then, the control device 100 controls the heat treatment unit U2 to apply heat treatment to the workpiece W, which has a film of processing liquid (resist liquid) formed on it. As a result, a resist film is formed on the surface Wa of the workpiece W.

[0044] [Control device details] Next, an example of the control device 100 will be described in detail. The control device 100 (feature quantity estimation device) has a function to control the coating and developing apparatus 2 to perform substrate processing such as film formation processing including the liquid treatment procedure described above, as well as a function to estimate shape features such as film thickness. The control device 100 controls the coating and developing apparatus 2 to perform various substrate processing in the production stage in which workpieces W are produced. In the preparation stage before moving to the production stage, the control device 100 performs a process to estimate shape features such as film thickness in order to set (adjust) the control conditions related to substrate processing in the production stage. In the preparation stage, pre-processing corresponding to the substrate processing in the production stage (for example, film formation processing including the liquid treatment procedure described above) may be performed in the substrate processing system 1 in which the control device 100 is installed. Processing in the preparation stage is performed, for example, when the coating and developing apparatus 2 is started for the first time, or when operation is restarted after changing the control conditions.

[0045] The control device 100 has, in terms of its functional configuration (hereinafter referred to as "functional modules"), for example, as shown in Figure 6, a condition storage unit 112, an operation instruction unit 114, an input information acquisition unit 118, a data acquisition unit 122, a model generation unit 124, a model selection unit 126, an estimated value calculation unit 128, and a condition change unit 142. The processing performed by these functional modules corresponds to the processing performed by the control device 100.

[0046] The condition storage unit 112 is a functional module that stores control conditions for substrate processing on the workpiece W. For example, the condition storage unit 112 stores the control conditions for the liquid processing procedure on the workpiece W. Multiple parameters are set in the control conditions for the liquid processing procedure (the value of each of the multiple parameters is defined). Each of the multiple parameters defines at least a part of the control content in the liquid processing procedure. The multiple parameters that define the liquid processing procedure include, for example, the time for which the processing liquid is discharged (or the first set time ts1), the rotation speed ω1 of the workpiece W when the processing liquid is discharged, the rotation speed ω2 after the discharge of the processing liquid is completed, the time for rotating at rotation speed ω2 (second set time ts2), the rotation speed ω3 when drying, and the time for rotating at rotation speed ω3 (third set time ts3).

[0047] The operation instruction unit 114 is a functional module that controls the coating and developing apparatus 2 to perform substrate processing on the workpiece W according to the control conditions stored in the condition storage unit 112. For example, the operation instruction unit 114 controls the liquid processing unit U1 to execute the liquid processing procedure including the steps S01 to S08 described above, according to the control conditions for the liquid processing procedure stored in the condition storage unit 112. The input information acquisition unit 118 is a functional module that acquires input information indicating instructions from the user from the input / output device 102.

[0048] The data acquisition unit 122 is a functional module that acquires actual measurement data obtained by measuring the film thickness on the surface Wa of the pre-treated workpiece W during the preparation stage. The workpiece W used in the pre-treatment may be the same type as the workpiece W that undergoes substrate treatment during the production stage. During the preparation stage, pre-treatment corresponding to substrate treatment is performed according to each of the multiple pre-treatment conditions (according to each pre-treatment condition). In the pre-treatment corresponding to film formation treatment, which includes a liquid treatment procedure, the same treatment as in steps S01 to S08 described above and a heat treatment in the heat treatment unit U2 are performed.

[0049] Each of the multiple pre-treatment conditions corresponds to the control condition described above and has the same multiple parameters (multiple parameters related to the film formation process) as the control condition. In the multiple pre-treatment conditions, the setting values ​​of one or more of the multiple parameters are different from each other. The multiple pre-treatment conditions may be determined by the user before performing multiple pre-treatments. The data acquisition unit 122 may acquire actual measurement data of the film thickness distribution after pre-treatment from the measurement unit 18 for each of the multiple pre-treatments. The actual measurement data of the film thickness distribution includes the measured values ​​of the film thickness at multiple measurement points on the surface Wa of the workpiece W.

[0050] The model generation unit 124 is a functional module that generates multiple regression models based on the measured data acquired by the data acquisition unit 122 during the preparation phase. Each of the multiple regression models is a model that represents the relationship between the film thickness and one or more parameters (hereinafter referred to as "target parameters") arbitrarily selected from among multiple parameters included in the control conditions (pre-processing conditions). The one or more target parameters can be arbitrarily selected by the user from among multiple parameters. Each of the multiple regression models is a regression equation generated to output an estimated value of the film thickness according to the input value of one or more target parameters. The multiple regression models (regression equations) may include first-order models and higher-order models of second order or higher. In one example, the model generation unit 124 generates a first-order regression model, a second-order regression model, and a third-order regression model as multiple regression models. The model generation unit 124 generates multiple regression models for each of the multiple locations on the surface Wa of the workpiece W. The above multiple locations correspond to the multiple measurement locations where the film thickness is measured by the measurement unit 18.

[0051] The model selection unit 126 is a functional module that, in the preparation stage, selects one regression model from among multiple regression models generated by the model generation unit 124 to be used for calculating the estimated film thickness. Hereinafter, the regression model selected by the model selection unit 126 will be referred to as the "selected model." The model selection unit 126 may select a selected model from among multiple regression models for each of the multiple locations (for each location). For each of the multiple regression models, the model selection unit 126 calculates an index value representing the reliability of the film thickness estimation result and evaluates the suitability of the multiple pre-processing condition setting methods. Then, based on the index value calculation result and the evaluation result of the suitability of the setting method, the model selection unit 126 selects a selected model from among the multiple regression models. Specific examples of the index value calculation procedure and the setting method suitability evaluation procedure will be described later.

[0052] The Estimate Calculation Unit 128 is a functional module that uses a selected model (one regression model) to calculate an estimated film thickness corresponding to any value of one or more target parameters. The selected model used by the Estimate Calculation Unit 128 to calculate the estimated film thickness is one regression model selected by the Model Selection Unit 126. Any value of one or more target parameters may be input by the user via the input / output device 102. The Estimate Calculation Unit 128 calculates the output result obtained by inputting any value of one or more target parameters into the selected model as the estimated film thickness. The Estimate Calculation Unit 128 may also calculate the estimated film thickness distribution result (hereinafter referred to as "estimated film thickness distribution") by calculating the estimated film thickness for each of multiple measurement points on the surface Wa of the workpiece W using the corresponding selected model. The Estimate Calculation Unit 128 may also estimate the estimated film thickness distribution for each of multiple arbitrary values ​​of one or more target parameters. The Estimate Calculation Unit 128 may output the estimated film thickness or the estimated film thickness distribution to the input / output device 102.

[0053] The condition change unit 142 is a functional module that changes (updates) the control conditions stored in the condition storage unit 112. The condition storage unit 112 may have reference values ​​for each of the multiple parameters that constitute the control conditions stored in advance. The condition change unit 142 changes the control conditions, for example, in response to instructions from the user via the input / output device 102. A user, such as an operator, may give instructions to change the control conditions after confirming the estimated film thickness distribution output by the estimated value calculation unit 128. In one example, the user extracts the estimated film thickness distribution that is best suited to the required performance (e.g., uniformity of film thickness) from among multiple estimated film thickness distributions with different target parameter values, and inputs the target parameter values ​​from which the extracted estimated film thickness distribution is obtained to the control device 100. Instead of confirmation by the user, the control device 100 may autonomously extract the estimated film thickness distribution that is best suited to the required performance from among multiple estimated film thickness distributions with different target parameter values, and output the target parameter values ​​from which that distribution is obtained as recommended values.

[0054] The control device 100 is comprised of one or more control computers. The control device 100 has, for example, the circuit 150 shown in Figure 7. The circuit 150 includes one or more processors 152, a memory 154, a storage 156, an input / output port 158, and a timer 162. The storage 156 has a storage medium that can be read by the computer, such as a hard disk. The storage medium stores a program that causes the control device 100 to execute the substrate processing method, which includes the liquid processing procedure and the feature estimation procedure described later. The storage medium may be a removable medium such as a non-volatile semiconductor memory, a magnetic disk, or an optical disk.

[0055] Memory 154 temporarily stores the program loaded from the storage medium of storage 156 and the calculation results by processor 152. Processor 152 works in cooperation with memory 154 to execute the above program (estimated program) and thereby configures each of the functional modules described above. Input / output ports 158 input and output electrical signals to and from the rotation holding unit 20, processing liquid supply unit 30, measurement unit 18, and input / output devices 102, etc., according to commands from processor 152.

[0056] If the control device 100 is composed of multiple control computers, each functional module may be implemented by an individual control computer. The control device 100 may consist of a control computer including a condition storage unit 112 and an operation instruction unit 114, and a control computer (feature estimation device) including a data acquisition unit 122, a model generation unit 124, a model selection unit 126, and an estimated value calculation unit 128. Alternatively, each of these functional modules may be implemented by a combination of two or more control computers. In these cases, the multiple control computers may be connected to each other in a manner that allows them to communicate with one another, and may coordinately execute the board processing method.

[0057] The hardware configuration of the control device 100 is not necessarily limited to configuring each functional module by program. For example, each functional module of the control device 100 may be composed of a dedicated logic circuit or an ASIC (Application Specific Integrated Circuit) that integrates such circuits.

[0058] [Feature Estimation Method] In the preparation phase, the control device 100 performs a series of processes (feature estimation methods) to estimate shape features such as film thickness. This series of processes includes, for example, a series of processes for generating a regression model from measured data obtained through multiple preprocessing steps, and a series of processes for estimating the film thickness distribution. First, an example of a series of processes for generating a regression model will be described. In the series of processes for generating a regression model, the user may pre-set one or more target parameters to check the change in film thickness distribution and set (adjust) the values ​​in the control conditions. Below, an example is given where there are two target parameters instead of one or more. For example, one target parameter (hereinafter referred to as the "first parameter") is the time for which the processing liquid is continuously discharged, and the other target parameter (hereinafter referred to as the "second parameter") is the rotational speed of the workpiece W during the discharge of the processing liquid.

[0059] (Generating regression models) Figure 8 is a flowchart of a series of processes for generating a regression model. In this series of processes, the control device 100 first executes step S21. In step S21, for example, the input information acquisition unit 118 acquires user instructions indicating the pre-processing conditions for each of the multiple workpieces W for pre-processing. The input information acquisition unit 118 acquires information indicating multiple pre-processing conditions for multiple pre-processing cycles. For each pre-processing condition, the values ​​of the first parameter and the second parameter are defined. In the pre-processing conditions, the values ​​of parameters other than the first and second parameters are set to the same values ​​as the control conditions used in the production stage.

[0060] In step S21, the user sets multiple preprocessing conditions such that at least one of the values ​​of the first parameter and the second parameter differs between the multiple preprocessing conditions. That is, in any two of the multiple preprocessing conditions, either one of the values ​​of the first parameter and the second parameter differs, or both the values ​​of the first parameter and the second parameter differ. Here, the value of the first parameter in the preprocessing performed on the kth time is defined as "p1k", and the value of the second parameter in the preprocessing performed on the kth time is defined as "p2k". k is an integer from 1 to N, and N is an integer greater than or equal to 2. The multiple preprocessing conditions include (p11, p21), (p12, p22), ..., and (p1N, p2N). N corresponds to the number of times the preprocessing is performed.

[0061] The values ​​of each p1k, and the setting range which is the difference between the minimum and maximum values ​​of p1k, can be arbitrarily set by the user. p1k contains two or more different values. p1k may be set to all different values, or some p1k may be set to the same value. The values ​​of each p2k, and the setting range which is the difference between the minimum and maximum values ​​of p2k, can be arbitrarily set by the user. p2k contains two or more different values. p2k may be set to all different values, or some p2k may be set to the same value. An example of the settings for p1k and p2k is shown in Figure 9.

[0062] Next, the control device 100 executes steps S22 and S23. In step S22, for example, the control device 100 performs initial settings for the target parameter values. In one example, the control device 100 sets the first parameter and the second parameter to (p11, p21). In step S23, for example, the operation instruction unit 114 controls the liquid treatment unit U1 so that liquid treatment is performed on the workpiece W for pretreatment according to the pretreatment conditions set in step S22. In the first pretreatment, the liquid treatment in the liquid treatment unit U1 is performed with the first parameter set to p11 and the second parameter set to p21. In step S23, a series of processes similar to those in steps S01 to S08 are executed, except for the values ​​of the first and second parameters.

[0063] Next, the control device 100 executes steps S24 and S25. In step S24, for example, the operation instruction unit 114 controls the heat treatment unit U2 to perform heat treatment on the pre-treatment workpiece W, which has undergone the liquid treatment in step S23, under the same conditions as the heat treatment in the production stage. In step S25, for example, the data acquisition unit 122 acquires the measurement results as actual measurement data from the measurement unit 18, which are the results of measuring the film thickness distribution of the resist film formed on the surface Wa of the pre-treatment workpiece W after the heat treatment in step S24.

[0064] Next, the control device 100 executes step S26. In step S16, for example, the control device 100 determines whether or not pre-processing has been completed for all pre-processing conditions. If it is determined in step S26 that pre-processing has not been completed for all pre-processing conditions (step S26: NO), the control device 100 proceeds to step S27. In step S27, for example, the control device 100 changes the value of the target parameter. After the first pre-processing is performed, the control device 100 sets the first and second parameters to (p12, p22).

[0065] Thereafter, the control device 100 repeatedly executes the series of processes in steps S23 to S27 on multiple workpieces W for pretreatment until N pretreatments have been performed. As described above, the control device 100 applies pretreatment to the workpieces W for pretreatment, corresponding to the liquid treatment procedure and the film formation process including heat treatment, for each pretreatment condition. The control device 100 then acquires measured data of the line width distribution on the surface Wa of the workpiece W after pretreatment for each pretreatment condition (for each pretreatment).

[0066] In step S26, if it is determined that pre-treatment has been completed under all pre-treatment conditions (step S26: YES), the control device 100 proceeds to step S28. In step S28, for example, the model generation unit 124 generates multiple regression models to predict the film thickness distribution on the surface Wa of the workpiece W after the film formation treatment in which the resist film is formed, based on the measured film thickness distribution data obtained by repeating step S25. The model generation unit 124 generates multiple regression models for each measurement location of the film thickness by the measurement unit 18.

[0067] In the series of processes shown in Figure 8, the control device 100 performs multiple (N) preprocessing steps while changing the settings of the first and second parameters. The data acquisition unit 122 of the control device 100 acquires the measurement results of the film thickness distribution from the measurement unit 18 each time it performs a preprocessing step. Since the values ​​of at least one of the first and second parameters differ between multiple preprocessing steps, the actual measurement data of the film thickness distribution obtained will differ. Figure 9 shows a table of examples of the values ​​of the first and second parameters for each preprocessing step and an example of the measurement results of the film thickness distribution. After preprocessing is performed according to each of the multiple preprocessing conditions, the model generation unit 124 may generate multiple regression models such that each model represents the relationship between one or more pre-set target parameters and the film thickness.

[0068] Figures 10(a) and 10(b) show schematic diagrams illustrating examples of multiple regression models. The model generation unit 124 generates multiple regression models capable of calculating predicted values ​​of film thickness at each of the multiple measurement points on the surface Wa of the workpiece W. When focusing on one measurement point, the multiple regression models are constructed to calculate an estimated value of film thickness at that measurement point. The orders of the multiple regression models associated with a single measurement point are different from each other. The model generation unit 124 generates a regression model for each measurement point in the film thickness distribution measured after pretreatment (after the film formation process that forms the resist film performed in the preparation stage), based on the measured values ​​(multiple measured values) of film thickness at that measurement point.

[0069] In Figure 10(a), "Pi" indicates each measurement point. Each measurement point, represented by Pi, is set on a straight line L passing through the center CP of the workpiece W on its surface Wa. Multiple measurement points are set at equal intervals along line L. In "Pi", "i" represents a number from 1 to M (where M is an integer greater than or equal to 2). In the example shown in Figure 10(a), i is from 1 to 13, and the 7th measurement point coincides with the center CP. Note that the number of measurement points set on line L is not limited to 13; it can be set to a number that allows observation of the trend in film thickness change on the surface Wa.

[0070] In Figure 10(b), "F1i(X1,X2)" represents a first-order regression model corresponding to each measurement point Pi. "F2i(X1,X2)" represents a second-order regression model corresponding to each measurement point Pi, and "F3i(X1,X2)" represents a third-order regression model corresponding to each measurement point Pi. "X1" represents the first parameter, and "X2" represents the second parameter. The model generation unit 124 generates a first-order regression equation, a second-order regression equation, and a third-order regression equation for each measurement point Pi that represent the change in film thickness (predicted value of film thickness) according to the first and second parameters.

[0071] In the graph shown in Figure 10(b), multiple measurements of film thickness taken for each pretreatment at a single measurement point Pi are indicated by black circles. The model generation unit 124 generates a regression equation (regression plane) for each measurement point Pi by performing regression analysis based on the multiple measurements of film thickness at that measurement point Pi and the values ​​of the first and second parameters (p1k, p2k) from which those measurements were obtained. The model generation unit 124 may perform the regression analysis using any method. For example, the model generation unit 124 generates multiple regression models with different degrees by ridge regression (ridge regression analysis).

[0072] (Estimation of film thickness distribution) Figure 11 is a flowchart showing a series of processes for estimating the film thickness distribution. In this series of processes, for example, the control device 100 executes step S31. In step S31, for example, the input information acquisition unit 118 acquires the values ​​of the first and second parameters for calculating the estimated film thickness distribution as prediction conditions. In one example, the user determines the values ​​for which they want to check the estimated film thickness distribution for each of the first and second parameters from within the range set by multiple pre-processing conditions (range from minimum to maximum value), and inputs these values ​​to the control device 100 via the input / output device 102.

[0073] Next, the control device 100 executes step S32. In step S32, for example, the model selection unit 126 evaluates the suitability of the setting method for the multiple preprocessing conditions obtained in step S21 described above. The setting method for multiple preprocessing conditions refers to the setting method for multiple values ​​of the target parameters in the multiple preprocessing conditions. In evaluating the suitability of the setting method, the property of being able to accurately calculate estimated values ​​within the setting range of the target parameters of the multiple preprocessing conditions using the regression model obtained by the regression analysis (machine learning) described above is evaluated. When evaluating the suitability of the setting method, the model selection unit 126 may evaluate the degree to which the setting method is appropriate in multiple stages. In one example, when evaluating the suitability of the setting method, the model selection unit 126 may determine whether or not the setting method is appropriate. Whether or not the setting method is appropriate means whether or not estimated values ​​can be accurately calculated within the setting range of the target parameters of the multiple preprocessing conditions using the regression model. For example, the model selection unit 126 determines whether the selection of multiple values ​​for the first parameter and multiple values ​​for the second parameter in the multiple preprocessing conditions used when generating the regression model allows for accurate estimation using the regression model. The setting range for the target parameter (the range from the lower limit to the upper limit for which the film thickness distribution is to be estimated) may be determined by the user. The following explanation will describe an example of determining whether the setting method is appropriate in the evaluation of the suitability of the setting method.

[0074] Figure 12 shows three sets of data where the values ​​of the first and second parameters are set differently under multiple pretreatment conditions. In the graph shown in Figure 12, each plotted black circle represents the values ​​of the first and second parameters under one pretreatment condition. In the two-dimensional graph shown in (I) in Figure 12, the data is evenly distributed throughout the entire setting range, and the data set is symmetrical (symmetrical with respect to the midline of the X1 and X2 axes). On the other hand, in the two-dimensional graphs shown in (II) and (III), there are blank areas where no data exists in certain regions within the setting range, compared to the graph shown in (I). If a regression model is generated using a data set with blank areas within the setting range, and the film thickness is to be estimated using that regression model in those blank areas, the estimated value may deviate significantly from the measured value.

[0075] In one example, if two or more target parameters are set, the model selection unit 126 performs the following three evaluations (evaluation 1, evaluation 2, and evaluation 3) to determine whether the above setting method is appropriate. The model selection unit 126 may determine that the above setting method is not appropriate if the result of any one of the three evaluations does not meet the evaluation criteria. Alternatively, the model selection unit 126 may determine that the above setting method is appropriate if the results of all three evaluations meet the evaluation criteria. Is the number of combinations of target parameter values ​​with a rating of 1:2 or higher greater than or equal to a predetermined number? Evaluation 2: For each of the target parameters, do the evaluation values ​​representing data bias meet the specified conditions? Evaluation 3: Does the evaluation value representing the regularity between target parameters of 2 or higher satisfy the predetermined conditions?

[0076] In evaluation 1 above, the model selection unit 126 determines whether the number of combinations of setting values ​​for two or more target parameters (N in the example above) is greater than or equal to a predetermined number. The model selection unit 126 determines that the evaluation criteria are met if the number of combinations is greater than or equal to the predetermined number, and determines that the evaluation criteria are not met if the number of combinations is less than the predetermined number. The predetermined number is set in advance and is 7 to 15 in one example.

[0077] In evaluation 2 above, the model selection unit 126 calculates an evaluation value representing the degree of data bias for each target parameter, based on multiple setting values ​​for that parameter under multiple preprocessing conditions. Hereinafter, for convenience of explanation, the evaluation value representing the degree of data bias will be referred to as "first evaluation value a1". The model selection unit 126 then determines whether the setting method is appropriate based on the calculation result of the first evaluation value a1. In one example, the model selection unit 126 determines that the evaluation criteria are met if the first evaluation value a1 is less than or equal to a predetermined first threshold, and determines that the evaluation criteria are not met if the first evaluation value a1 is greater than the first threshold.

[0078] Figure 13 shows histograms for the second parameter (X2) corresponding to the three types of data sets shown in Figure 12. In the histograms shown in Figure 13, the horizontal axis represents the second parameter, and the vertical axis represents the number of data points. The model selection unit 126 may calculate the first evaluation value a1 by subtracting the average number of data points on the histogram from the maximum number of data points on the histogram. In Figure 13, in the histogram shown in (I), the maximum number of data points is 3, and the average number of data points is 2.6 (=13 / 5). In the histogram shown in (II), the maximum number of data points is 8, and the average number of data points is approximately 2.2 (=13 / 6). In the histogram shown in (III), the maximum number of data points is 5, and the average number of data points is 1.8 (=9 / 5).

[0079] The model selection unit 126 determines whether the first evaluation value a1 for the second parameter is less than or equal to the first threshold. The first threshold is predetermined, and in one example it is 0.8 to 2.0. Similarly, the model selection unit 126 calculates the first evaluation value a1 for the first parameter and determines whether it is less than or equal to the first threshold. The model selection unit 126 may determine that the evaluation criteria are met if the first evaluation value a1 for both the first and second parameters is less than or equal to the first threshold. The model selection unit 126 may also determine that the evaluation criteria are not met if the first evaluation value a1 for either the first or second parameter is greater than the first threshold. In the examples shown in Figures 12 and 13, the data group (I) is determined to meet the evaluation criteria in evaluation 2, and the data groups (II) and (III) are determined not to meet the evaluation criteria in evaluation 2.

[0080] In evaluation 3 above, the model selection unit 126 calculates an evaluation value of the regularity between two or more target parameters for each of the multiple setting values ​​of two or more target parameters in the multiple pre-processing conditions. For example, the model selection unit 126 calculates an evaluation value of the regularity between the first parameter and the second parameter for each of the multiple setting values ​​of the first parameter and the second parameter in the multiple pre-processing conditions. Hereinafter, for the sake of explanation, the evaluation value of the regularity between two or more target parameters will be referred to as the "second evaluation value a2". The second evaluation value a2 represents the degree of regularity (degree of correlation) between the two or more target parameters. Then, based on the calculation result of the second evaluation value a2, the model selection unit 126 determines whether the setting method is appropriate or not.

[0081] The second evaluation value a2 represents the degree of symmetry when multiple settings for each of two or more target parameters are plotted on a graph within the setting range of those parameters. The less regularity (less correlation), the higher the degree of symmetry. The model selection unit 126 calculates the second evaluation value a2 by creating a correlation matrix between two or more target parameters and then finding the determinant of that correlation matrix. The model selection unit 126 only needs to calculate the determinant of the correlation matrix, and this calculation includes creating the correlation matrix. The second evaluation value a2 is the result of calculating the determinant of the correlation matrix.

[0082] If there are two or more target parameters, namely the first and second parameters, the second evaluation value a2 is calculated by performing the determinant of the correlation matrix represented by the following equation (1).

number

[0083] In the determinant (correlation matrix) included in equation (1) above, r12 and r21 are the correlation coefficients between multiple settings for the first parameter and multiple settings for the second parameter. The greater the degree of symmetry with respect to the multiple settings for each of the first and second parameters, the closer the correlation coefficient approaches 0. As a result, the greater the degree of symmetry, the closer the second evaluation value a2 approaches 1. Conversely, the smaller the degree of symmetry, the closer the correlation coefficient approaches 1, and the closer the second evaluation value a2 approaches 0.

[0084] The model selection unit 126 determines whether the second evaluation value a2 is greater than or equal to a predetermined second threshold. The second threshold is predetermined and, in one example, is 0.9 to 1.0. If the second threshold is set to 1.0, the model selection unit 126 determines whether the second evaluation value a2 is 1. The model selection unit 126 may also determine that the evaluation criteria are met if the second evaluation value a2 is greater than or equal to the second threshold. The model selection unit 126 may also determine that the evaluation criteria are not met if the second evaluation value a2 is less than the second threshold. In the examples shown in Figures 12 and 13, the data groups (I) and (III) are determined to meet the evaluation criteria in evaluation 3, and the data group (II) is determined not to meet the evaluation criteria in evaluation 3.

[0085] In the example shown in Figure 12, summarizing the judgments in evaluations 1 to 3 above, in the data set (I), it is determined that the evaluation criteria are met in all of evaluations 1 to 3, so the model selection unit 126 determines that the method of setting multiple preprocessing conditions is appropriate. In the data set (II), it is determined that the evaluation criteria are not met in evaluations 2 and 3, so the model selection unit 126 determines that the method of setting multiple preprocessing conditions is not appropriate. In the data set (III), it is determined that the evaluation criteria are not met in evaluation 3, so the model selection unit 126 determines that the method of setting multiple preprocessing conditions is not appropriate.

[0086] When one target parameter is set (when generating a regression model that represents the relationship between one target parameter and film thickness), the model selection unit 126 may determine whether the setting method is appropriate by performing the determinations in evaluation 1 and evaluation 2 without performing the determination in evaluation 3. In the determination in evaluation 1 when one target parameter is set, the predetermined number used for determining whether the determination is possible or not may be 3 to 6.

[0087] In evaluation 2, when one target parameter is set, the model selection unit 126 may determine whether the setting method is appropriate or not according to the following determination method, unlike the determination method described above (determination method using the first evaluation value a1). The model selection unit 126 may determine that the setting method is appropriate if all of the minimum value, 1 / 4 value, median value, 3 / 4 value, and maximum value are included among the multiple setting values ​​of the target parameter. The model selection unit 126 may determine that the setting method is inappropriate if none of the minimum value, 1 / 4 value, median value, 3 / 4 value, and maximum value are included among the multiple setting values ​​of the target parameter. The 1 / 4 value is the value obtained by adding (maximum value - minimum value) / 4 to the minimum value. The 3 / 4 value is the value obtained by adding 3 × (maximum value - minimum value) / 4 to the minimum value.

[0088] Figure 14 shows three sets of data where the setting method for the second parameter value differs under multiple pre-processing conditions. In the example shown in Figure 14, only the second parameter is set as the target parameter. In Figure 14, the data sets shown in (I) and (II) all include the minimum value, quarter value, median, three-quarter value, and maximum value. Therefore, the model selection unit 126 determines that the setting method is appropriate. On the other hand, the data set shown in (III) does not include the quarter value or three-quarter value. Therefore, the model selection unit 126 determines that the setting method is not appropriate. In Figure 14, data that corresponds to any of the minimum value, quarter value, median, three-quarter value, and maximum value are represented by black circles, and data that does not correspond to any of the above five types of values ​​are represented by white circles.

[0089] The model selection unit 126 may determine that the setting method is appropriate if it determines that the criteria are met in at least two of the evaluations from evaluation 1, evaluation 2, and evaluation 3. The method for determining whether the setting method for multiple preprocessing conditions is appropriate is not limited to the example described above. When two or more target parameters are set, the model selection unit 126 may determine whether the setting method is appropriate using at least one or two of the evaluation methods from evaluation 1, evaluation 2, and evaluation 3.

[0090] Returning to Figure 11, after step S32 is executed, the control device 100 executes step S33. In step S33, for example, the model selection unit 126 determines whether the evaluation result in step S32 is appropriate or not. If it is determined that the evaluation result is not appropriate (step S33: NO), the control device 100 proceeds to step S34. In step S34, for example, the model selection unit 126 narrows down from the multiple regression models generated in step S28 to one or more candidate models that can be used as the regression model (the selected model above) for estimating the film thickness.

[0091] The number of candidate models narrowed down in step S34 is less than the number of regression models generated in step S28. In this case, the model selection unit 126 performs a process in step S34 to select one or more candidate models from among the multiple regression models, which is less than the number of multiple regression models. For example, the model selection unit 126 excludes models with a high degree of 1 or higher from among the multiple regression models, and then sets the remaining models as one or more candidate models. In one example, the model selection unit 126 excludes the second-order and third-order regression models from among the multiple regression models, and then sets the first-order regression model as a candidate model. The model selection unit 126 narrows down the multiple regression models to one or more candidate models for each measurement point (for example, setting only the first-order regression model as a candidate model).

[0092] On the other hand, if the evaluation result is determined to be appropriate in step S33 (step S33: YES), the control device 100 proceeds to step S35. If the process proceeds to step S35, the model selection unit 126 sets the multiple regression models as they are, into one or more candidate models. In this case, the model selection unit 126 selects all models from the multiple regression models as one or more candidate models. In one example, the model selection unit 126 sets the first-order, second-order, and third-order regression models as three candidate models. For each measurement point, the model selection unit 126 sets the multiple regression models as multiple candidate models without narrowing them down from the multiple regression models.

[0093] In steps S32 to S34 described above, the model selection unit 126 selects one or more candidate models from among multiple regression models to be candidates for the selected model, based on the determination result of whether the setting method for the multiple preprocessing conditions is appropriate or not. If the model selection unit 126 determines that the setting method is not appropriate, it excludes one or more models from the multiple regression models and sets the remaining models as one or more candidate models. If the model selection unit 126 determines that the setting method is not appropriate in the determination result of whether the setting method is appropriate or not, it may exclude models with a high degree of 1 or higher from the multiple regression models and set the remaining models as one or more candidate models. For example, if the model selection unit 126 determines that the setting method is not appropriate in the determination result of whether the setting method is appropriate, it may exclude the second-order and third-order regression models from the multiple regression models and set the first-order regression model as a candidate model. Alternatively, if the model selection unit 126 determines that the setting method is appropriate, it may set the multiple regression models (for example, multiple regression models of first, second, and third order) as they are and set them as one or more candidate models.

[0094] Next, the control device 100 executes step S35. In step S35, for example, the model selection unit 126 calculates an index value representing the confidence level of the shape feature estimation results for each of the one or more candidate models. The model selection unit 126 calculates the index value for each measurement location (for each of the multiple measurement locations). If step S34 has been executed, the model selection unit 126 calculates the index value for each of the one or more narrowed-down candidate models for each measurement location. If step S34 has not been executed, the model selection unit 126 calculates the index value for each of the multiple regression models generated in step S28 for each measurement location.

[0095] The index value calculated in step S35 is, for example, AIC (Akaike Information Criterion). The AIC index value is calculated for each candidate model based on the mean squared error (MSE) between the predicted value and the measured value in the actual data when multiple setting values ​​for the target parameters of multiple preprocessing conditions are input into the candidate model, and the order of the model. The higher the order, the better the model fits to multiple setting values ​​(the mean squared error becomes smaller), but the risk of overfitting increases. The AIC is calculated by comprehensively considering the goodness of the model's fitting to multiple setting values ​​in multiple preprocessing conditions and the risk of overfitting associated with an increasing order. In AIC, a smaller value means that the model has a higher reliability of the estimation result. Note that AIC is just one example of an index value, and the model selection unit 126 may use other model selection criteria such as finite modified AIC (c-AIC) instead of AIC.

[0096] Next, the control device 100 executes step S36. In step S36, for example, the model selection unit 126 selects one regression model as the selected model from among the multiple regression models generated in step S28, based on the results of the index value calculation in step S35 and the determination result of whether the setting method in step S32 is appropriate. The model selection unit 126 determines the selected model for each measurement location (for each of multiple measurement locations) based on the results of the index value calculation and the determination result of whether the setting method is appropriate. The model selection unit 126 selects a selected model from one or more candidate models based on the results of the index value calculation in step S35. For example, the model selection unit 126 selects the regression model that is judged to have the highest reliability based on the index value from among the one or more candidate models selected by the determination result in step S32. In one example, the model selection unit 126 selects the regression model with the smallest AIC from among the one or more candidate models as the selected model.

[0097] Figure 15 schematically shows how one model is selected from among first-order, second-order, and third-order regression models for each measurement point. The calculation results of the index values ​​for each of the multiple regression models may differ between any two or more measurement points, and the order of the selected models may differ from one another. Note that if one candidate model is selected (narrowed down) in step S34, the processing in step S35 may be omitted. Furthermore, the model selection unit 126 may set the single candidate model as the selected model in step S36.

[0098] Next, the control device 100 executes step S37. In step S37, for example, the estimation unit 128 calculates an estimated film thickness corresponding to the prediction conditions (values ​​of one or more target parameters) obtained in step S31, using the regression model selected in step S36. For each of the multiple measurement locations, the estimation unit 128 calculates an estimated film thickness by inputting the values ​​of one or more target parameters defined in the prediction conditions (for example, the values ​​of the first parameter and the second parameter) into the selected model determined at that location. This provides an estimated result of the film thickness distribution representing the variation in film thickness on the surface Wa of the workpiece W.

[0099] Next, the control device 100 executes step S38. In step S38, for example, the estimated value calculation unit 128 outputs the estimated film thickness distribution calculated in step S37 to the input / output device 102. By executing step S38, the user can confirm the estimated film thickness distribution corresponding to the prediction conditions in step S31. This completes the series of processes for estimating the film thickness distribution. The user may have the control device 100 repeatedly execute the series of processes from steps S31 to S38 while changing the prediction conditions until the desired film thickness distribution is obtained. The estimated value calculation unit 128 may further calculate an index of in-plane uniformity at the surface Wa in the predicted film thickness distribution and output that index to the input / output device 102. The estimated value calculation unit 128 may extract the values ​​of one or more target parameters from among the multiple prediction conditions that yield the predicted film thickness distribution with the highest index of in-plane uniformity and output them to the input / output device 102.

[0100] After the series of processes in steps S31 to S38 have been executed, the user or the control device 100 itself may set the values ​​of one or more target parameters used in the liquid processing procedure at the production stage. For example, the condition change unit 142 changes the control conditions stored in the condition storage unit 112 in response to user input via the input / output device 102, or according to the values ​​of the target parameters extracted by the estimated value calculation unit 128.

[0101] [Validation results of model selection] Figure 16 shows a graph representing the verification results of the estimated film thickness distribution, calculated by selecting a regression model from among several regression models using an index value representing the reliability of the estimation results. In the verification of the estimated film thickness distribution, as a comparative example, results were prepared in which the estimated film thickness distribution was calculated using one of the following regression models: first-order, second-order, or third-order, at all measurement locations. Then, for each measurement location, the estimated film thickness distribution was calculated using the model with the smallest AIC from among the first-order, second-order, and third-order regression models, and compared with the results in the comparative example. The first-order, second-order, and third-order regression models were generated by ridge regression analysis from the same multiple pre-processing conditions. The predicted film thickness distribution was calculated for each of the three conditions (Condition 1, Condition 2, and Condition 3) in which at least one of the first and second parameters differed, and then compared with the measured film thickness distribution. In the comparison between the predicted film thickness distribution and the measured film thickness distribution, the average of the difference between the predicted and measured values ​​at multiple measurement locations was calculated as the mean error.

[0102] In the graph shown in Figure 16, "ΔRidge_1" shows the average error calculated after obtaining the predicted film thickness distribution using a first-order regression model at all measurement points. "ΔRidge_2" shows the average error calculated after obtaining the predicted film thickness distribution using a second-order regression model at all measurement points. "ΔRidge_3" shows the average error calculated after obtaining the predicted film thickness distribution using a third-order regression model at all measurement points. "ΔAIC" shows the average error calculated after obtaining the predicted film thickness distribution using a model with a small AIC for each measurement point.

[0103] The verification results in Figure 16 show that when a model with a small AIC is selected for each measurement location to calculate the predicted film thickness distribution, the average error is generally smaller compared to when the same-order regression model is used for all measurement locations. Note that in condition 2, the average error is smaller when a first-order regression model is used for all measurement locations, but in conditions 1 and 3, the average error is larger. Therefore, it can be seen that selecting a model using AIC for each measurement location consistently results in a smaller average error.

[0104] [Differentiation] The series of processes shown in Figures 8 and 11 are examples and can be modified as appropriate. In the above series of processes, the control device 100 may execute one step and the next step in parallel, or it may execute each step in a different order than the example above. The control device 100 may execute a process in any of the steps that is different from the example above.

[0105] The control device 100 may, instead of predicting the film thickness, predict the distribution of line widths on the surface Wa of the workpiece W after a predetermined process including development and heat treatment associated with development, as the distribution of shape features. In this case, the multiple parameters constituting the control conditions (pre-treatment conditions) include, for example, the timing of the start of dispensing the developer solution during the development process, the time for which the dispensing of the developer solution is continued, the time for which the paddle is maintained during static development, the timing of the start of dispensing of the adjusting solution other than the developer solution supplied during development, and the time for which the dispensing of the adjusting solution is continued.

[0106] In the selection results for each measurement location, if there is a large difference in the order of the models selected at adjacent measurement locations, the selected models may be adjusted. The model selection unit 126 may calculate the difference (hereinafter referred to as the "order difference") between the order of the model selected at one measurement location (first measurement location) and the order of the model selected at the adjacent measurement location (second measurement location). Based on the order difference, the model selection unit 126 may determine whether it is necessary to modify the model selected at either of the adjacent measurement locations in order to reduce the difference.

[0107] For example, when first-order, second-order, and third-order regression models are generated, the model selection unit 126 determines that model modification is necessary if the order difference is 2, and determines that model modification is not necessary if the order difference is 0 or 1. If the model selection unit 126 determines that model modification is necessary, it may output the determination result to the input / output device 102. In this case, the model selection unit 126 may change the model selected at any measurement point based on user input after the output of the determination result. If the model selection unit 126 determines that model modification is necessary, it may change the model with the higher order (e.g., a third-order regression model) to the model with the lower order (e.g., a second-order regression model) among a pair of models selected at adjacent measurement points.

[0108] The control device 100 may calculate complementary conditions to complement the multiple pre-processing conditions so that the setting methods for the multiple pre-processing conditions are appropriate. The control device 100 determines whether there is a region with a relatively small distribution density within the setting range of multiple setting values ​​for any one of the target parameters (first parameter) in the multiple pre-processing conditions. If there is a region with a relatively small distribution density, the control device 100 calculates the values ​​or ranges included in that region as complementary conditions so that the setting method is determined to be appropriate.

[0109] Even if it is determined that the setting method for multiple pre-processing conditions is not appropriate, if the prediction condition (the value of the target parameter that the user wants to estimate) is included in a specific area among the multiple pre-processing conditions, the regression model does not need to be narrowed down. For example, the control device 100 determines whether the value of the target parameter that the user wants to estimate is in an area where the distribution density is relatively large within the setting range of multiple setting values ​​for that target parameter in the multiple pre-processing conditions. Then, if the value of the symmetric parameter to be estimated is included in an area where the distribution density is relatively large, the control device 100 may select a model from among first-order, second-order, and third-order regression models based on the index value for each measurement location and calculate the estimated film thickness distribution.

[0110] The control device 100 may acquire measured shape feature data obtained by performing multiple pre-processing operations according to multiple pre-processing conditions in a coating and developing apparatus 2 separate from the one being controlled, along with the settings of multiple implementation processing conditions. A computer including a data acquisition unit 122, a model generation unit 124, a model selection unit 126, and an estimated value calculation unit 128 may constitute a feature estimation device. The computer constituting the feature estimation device may be provided independently of the substrate processing system 1 and may not have the function of controlling the coating and developing apparatus 2.

[0111] The control device 100 may evaluate the suitability of the setting method in three or more stages, instead of determining whether it is suitable or not. The evaluation results of three or more stages of suitability may include an evaluation (determination) that it is not suitable. In this case, a method for narrowing down the regression model may be defined for each evaluation stage. In one of the various examples described above, at least some of the matters described in the other examples may be applied.

[0112] [Effects of the Embodiment] The feature estimation method described above is a method for estimating shape features that represent the shape characteristics of the surface Wa of a workpiece W after it has undergone a predetermined process for semiconductor manufacturing. This feature estimation method includes generating multiple regression models that represent the relationship between one or more target parameters included in the multiple parameters and the shape features, based on measured data of the shape features of the surface Wa of a workpiece W that has undergone a predetermined process according to each of multiple preprocessing conditions having multiple parameters related to the predetermined process; calculating an index value representing the reliability of the shape feature estimation result for each of the multiple regression models; evaluating the suitability of the method for setting the multiple preprocessing conditions; selecting one regression model as the selected model from among the multiple regression models based on the calculation result of the index value and the evaluation result of the suitability of the setting method; and using the selected model to calculate an estimated value of the shape features corresponding to an arbitrary value of one or more target parameters.

[0113] One possible method for estimating shape features using a regression model is to estimate them using a single regression model specified by the user, regardless of how multiple preprocessing conditions are set. However, with this method, the estimation accuracy of the regression model may be low depending on the location on the surface Wa of the workpiece W where the shape features are calculated, or the value of the parameter to be estimated. In contrast, the above method evaluates the suitability of the settings for multiple preprocessing conditions, calculates an index value representing the reliability of the estimation results, and then selects a model based on these results. The selected model is then used to calculate the estimated value. Therefore, if the suitability of the setting method is expected to result in low estimation accuracy by the model, the available models can be limited, and a model with high reliability of the estimation results can be used. Consequently, the above method is useful for improving the estimation accuracy of shape features on the surface Wa of the workpiece W.

[0114] In the feature estimation method described above, the measured data may include measured values ​​of shape features at multiple measurement points on the surface Wa of the workpiece W. In the above method, for each measurement point included in the multiple measurement points, multiple regression models may be generated, index values ​​may be calculated, and a selection model may be made based on the index value calculation results and the evaluation results of the suitability of the setting method. It is also conceivable to calculate estimated values ​​of shape features using one type of regression model specified by the user for all measurement points. However, for each measurement point, the type of regression model that minimizes the error with the measured value may differ depending on the characteristics of that measurement point. In the above method, since the model is selected for each measurement point based on the evaluation results of the suitability of the setting method and the index value calculation results, it is possible to select a regression model that matches the characteristics of each measurement point. Therefore, this is even more useful for improving the accuracy of estimating shape features on the surface Wa of the workpiece W.

[0115] In the feature estimation method described above, selecting a model may include selecting one or more candidate models from among multiple regression models based on the evaluation results of the suitability of the setting method, and selecting a model from among one or more candidate models based on the calculation results of the index value. In this case, if the suitability of the setting method is assumed to result in low estimation accuracy by the model, the models used when calculating the estimated value are limited by selecting only some models from among multiple regression models. Therefore, if the suitability of the setting method is assumed to result in low estimation accuracy by the model, the use of models that are expected to produce larger errors can be excluded. Consequently, this is even more useful for improving the estimation accuracy of shape features on the surface Wa of workpiece W.

[0116] In the feature estimation method described above, the degrees of the multiple regression models may differ from each other. Selecting one or more candidate models may include, if the evaluation of the suitability of the setting method determines that the setting method is not appropriate, excluding models with a degree of 1 or higher from among the multiple regression models, and then setting the remaining models as one or more candidate models. In this case, if the setting method is not appropriate, the available models are limited to models with a lower degree. Models with a higher degree tend to produce larger errors between their estimated values ​​and actual values ​​compared to models with a lower degree when the setting method is not appropriate. The above method is even more useful for improving the accuracy of estimating shape features on the surface Wa of the workpiece W because it can exclude the use of models with a higher degree when the setting method is not appropriate.

[0117] In the feature estimation method described above, evaluating the suitability of the setting method may include calculating an evaluation value representing the degree of data bias for each parameter included in one or more target parameters, for multiple setting values ​​of that parameter under multiple preprocessing conditions, and determining whether the setting method is appropriate based on the calculation results of the evaluation value. Selecting one or more candidate models may include, if it is determined that the setting method is not appropriate, excluding one or more models from the multiple regression models and setting the remaining models as one or more candidate models, and if it is determined that the setting method is appropriate, setting the multiple regression models as they are as one or more candidate models. If the degree of data bias is large for multiple setting values ​​of the target parameter, under certain prediction conditions, the error between the estimated value by the regression model and the actual value may become large due to that bias. In the above method, it is possible to restrict the types of models that can be used when the degree of data bias is large, without restricting the types of models that can be used when the degree of data bias is small. Therefore, it is possible to exclude the use of models that are expected to produce large errors when the data is heavily biased or the setup method is not appropriate, which further improves the accuracy of estimating the shape features of the surface Wa of the workpiece W.

[0118] In the feature estimation method described above, the one or more target parameters may be two or more target parameters. Evaluating the suitability of the setting method may include calculating an evaluation value of the regularity between the two or more target parameters for each of the multiple setting values ​​of the two or more target parameters under multiple preprocessing conditions, and determining whether the setting method is appropriate based on the calculation results of the evaluation value. Selecting one or more candidate models may include, if the setting method is determined to be inappropriate, excluding one or more models from the multiple regression models and setting the remaining models as one or more candidate models, and if the setting method is determined to be appropriate, setting the multiple regression models as they are as one or more candidate models. If, for the multiple setting values ​​of two or more target parameters, the setting value of one parameter has a regularity such as a strong correlation with the setting values ​​of other parameters, then under certain prediction conditions, the error between the estimated value by the regression model and the actual value may become large due to that regularity. In the above method, the types of models that can be used are not restricted when there is no regularity between the two or more target parameters, but the types of models that can be used are restricted when there is regularity. Therefore, since it is possible to exclude the use of models that exhibit regularity and are expected to produce large errors when the setup method is not appropriate, it is even more useful in improving the accuracy of estimating shape features on the surface Wa of the workpiece W.

[0119] In the feature estimation method described above, the degrees of multiple regression models may differ from each other. The feature estimation method may further include determining whether modification of the model selected at the first or second measurement location is necessary to reduce the difference between the degree of the model selected at the first measurement location and the degree of the model selected at the second measurement location adjacent to the first measurement location, based on the result of selecting a model at each measurement location. If a pair of regression models with a large difference in degree are used at adjacent measurement locations, the error in the estimated values ​​at those measurement locations may be large. In the above method, if the difference in degree is large, it can be determined that the model needs to be modified, and the model that was selected can be changed to reduce the difference in degree. Therefore, it is even more useful for improving the accuracy of estimating the shape features of the surface Wa of the workpiece W.

[0120] In the feature estimation method described above, if there is a region with a relatively small distribution density among the ranges of multiple setting values ​​for the first parameter in multiple pre-processing conditions for one or more target parameters, the value or range of the first parameter included in that region may be calculated as a complementary condition. In this case, by performing additional pre-processing according to the calculated complementary condition, the regression model generated by the condition setting method determined to be appropriate can be updated. Therefore, this is even more useful for improving the estimation accuracy of shape features on the surface Wa of the workpiece W.

[0121] In the feature estimation method described above, generating multiple regression models may include generating multiple regression models that represent the relationship between one or more pre-defined target parameters and shape features after pre-processing has been performed according to each of the multiple pre-processing conditions. In this case, multiple regression models are generated for a specific pre-defined parameter. Therefore, the user can understand the estimated value of the shape feature corresponding to any value of that specific parameter.

[0122] [Note] This disclosure includes the methods described in Appendix 1 to 9 and the program described in Appendix 10. <Note 1> A feature estimation method for estimating shape feature quantities that represent the shape characteristics of the surface of a substrate after it has undergone a predetermined process for semiconductor manufacturing, Based on measured data of the shape features on the surface of a substrate that has undergone pre-processing corresponding to the predetermined processing, according to each of the multiple pre-processing conditions having multiple parameters for the predetermined processing, a plurality of regression models representing the relationship between one or more target parameters included in the plurality of parameters and the shape features are generated. For each of the aforementioned regression models, an index value representing the confidence level of the estimation result of the shape feature is calculated, To evaluate the suitability of the method for setting the aforementioned multiple pre-treatment conditions, Based on the calculation results of the aforementioned index values ​​and the evaluation results of the suitability of the setting method, one regression model is selected from among the multiple regression models as the selected model. A feature estimation method comprising: using the selection model described above to calculate an estimated value of the shape feature corresponding to any value of the one or more target parameters described above. <Note 2> The measured data includes the measured values ​​of the shape feature quantities at multiple measurement locations on the surface of the substrate. The feature estimation method described in Appendix 1, wherein for each measurement location included in the plurality of measurement locations, the generation of the plurality of regression models, the calculation of the index value, and the selection of the selected model based on the calculation result of the index value and the evaluation result of the suitability of the setting method are performed. <Note 3> The degrees of the aforementioned multiple regression models are different from each other. The feature estimation method described in Appendix 2 further includes determining, based on the difference between the order of the model selected at the first measurement location and the order of the model selected at the second measurement location adjacent to the first measurement location, whether or not it is necessary to modify the model selected at the first measurement location or the second measurement location in order to reduce the difference, in the selection result of the selected model at each measurement location. <Note 4> Selecting the aforementioned model means Based on the evaluation results of the suitability of the setting method, one or more candidate models are selected from the plurality of regression models to be candidates for the selected model. A feature estimation method according to any one of the appendices 1 to 3, comprising selecting the selected model from the one or more candidate models based on the calculation results of the aforementioned index value. <Note 5> The degrees of the aforementioned multiple regression models are different from each other. The feature estimation method described in Appendix 4, wherein the selection of one or more candidate models includes, if the evaluation result of the suitability of the setting method determines that the setting method is not suitable, excluding one or more models with a high degree from among the multiple regression models, and then setting the remaining models as one or more candidate models. <Note 6> Evaluating the suitability of the aforementioned setting method is For each of the one or more target parameters mentioned above, an evaluation value representing the degree of data bias is calculated for each of the multiple setting values ​​of that parameter in the multiple pre-processing conditions. This includes determining whether the setting method is appropriate based on the calculation result of the evaluation value, Selecting one or more candidate models as described above is If the aforementioned setting method is determined to be inappropriate, one or more of the aforementioned regression models will be excluded, and the remaining models will be set as the one or more candidate models. The feature estimation method according to Appendix 4 or 5, which includes setting the plurality of regression models as they are to the one or more candidate models if the setting method is determined to be appropriate. <Note 7> The above 1 or more target parameters are 2 or more target parameters, Evaluating the suitability of the aforementioned setting method is For each of the two or more target parameters in the aforementioned multiple pre-processing conditions, an evaluation value of the regularity between the two or more target parameters is calculated. This includes determining whether the setting method is appropriate based on the calculation result of the evaluation value, Selecting one or more candidate models as described above is If the aforementioned setting method is determined to be inappropriate, one or more of the aforementioned regression models will be excluded, and the remaining models will be set as the one or more candidate models. A feature estimation method according to any one of the appendices 4 to 6, which includes setting the multiple regression models as they are to the one or more candidate models if the setting method is determined to be appropriate. <Note 8> A feature estimation method according to any one of the appendices 1 to 7, further comprising, with respect to the first parameter included in the one or more target parameters, if there is a region in the range of multiple setting values ​​for the first parameter in the multiple preprocessing conditions in which the distribution density is relatively small, calculating the value or range of the first parameter included in the said region as a complementary condition. <Note 9> The feature estimation method according to any one of the appendices 1 to 8, wherein generating the plurality of regression models includes generating the plurality of regression models to represent the relationship between the predetermined one or more target parameters and the shape features after the preprocessing according to each of the plurality of preprocessing conditions has been performed. <Note 10> An estimation program that causes a computer to execute one of the feature estimation methods described in one of the appendices 1 to 9. [Explanation of symbols]

[0123] 1...Substrate processing system, W...Workpiece, 2...Coating and developing apparatus, U1...Liquid processing unit, U2...Heat treatment unit, 100...Control device, 122...Data acquisition unit, 124...Model generation unit, 126...Model selection unit, 128...Estimated value calculation unit.

Claims

1. A feature estimation method performed by a feature estimation device for estimating shape feature quantities that represent the shape characteristics of the surface of a substrate after a predetermined process for semiconductor manufacturing has been applied, The model generation unit of the feature estimation device generates a plurality of regression models that represent the relationship between one or more target parameters included in the plurality of parameters and the shape features, based on measured data of the shape features on the surface of a substrate that has undergone preprocessing corresponding to the predetermined processing, according to each of the plurality of preprocessing conditions having a plurality of parameters related to the predetermined processing. The model selection unit of the feature estimation device calculates an index value representing the reliability of the estimation result of the shape feature for each of the plurality of regression models, The model selection unit evaluates the suitability of the method for setting the multiple pre-processing conditions, The model selection unit selects one regression model from among the multiple regression models as the selected model based on the calculation results of the index values ​​and the evaluation results of the suitability of the setting method. A feature estimation method comprising the feature estimation device calculating an estimated value of the shape feature corresponding to any value of the one or more target parameters using the selected model.

2. The measured data includes the measured values ​​of the shape feature quantities at multiple measurement locations on the surface of the substrate. The feature estimation method according to claim 1, wherein for each measurement location included in the plurality of measurement locations, the model generation unit generates the plurality of regression models, the model selection unit calculates the index value, and the model selection unit selects the selected model based on the calculation result of the index value and the evaluation result of the suitability of the setting method.

3. The model selection unit selects the selected model, Based on the evaluation results of the suitability of the setting method, one or more candidate models are selected from the plurality of regression models to be candidates for the selected model. The feature estimation method according to claim 1, comprising selecting the selected model from the one or more candidate models based on the calculation result of the index value.

4. The degrees of the aforementioned multiple regression models are different from each other. The feature estimation method according to claim 3, wherein the model selection unit selects one or more candidate models, and if the evaluation result of the suitability of the setting method determines that the setting method is not suitable, it excludes one or more models with a high degree from the plurality of regression models and sets the remaining models as one or more candidate models.

5. The model selection unit evaluates the suitability of the setting method, For each of the one or more target parameters mentioned above, an evaluation value representing the degree of data bias is calculated for each of the multiple setting values ​​of that parameter in the multiple pre-processing conditions. This includes determining whether the setting method is appropriate based on the calculation result of the evaluation value, The model selection unit selects one or more candidate models. If the aforementioned setting method is determined to be inappropriate, one or more of the aforementioned regression models will be excluded, and the remaining models will be set as the one or more candidate models. The feature estimation method according to claim 3 or 4, further comprising setting the plurality of regression models as they are in the one or more candidate models if it is determined that the setting method is appropriate.

6. The aforementioned one or more target parameters are two or more target parameters, The model selection unit evaluates the suitability of the setting method, For each of the two or more target parameters in the aforementioned multiple pre-processing conditions, an evaluation value of the regularity between the two or more target parameters is calculated. This includes determining whether the setting method is appropriate based on the calculation result of the evaluation value, The model selection unit selects one or more candidate models. If the aforementioned setting method is determined to be inappropriate, one or more of the aforementioned regression models will be excluded, and the remaining models will be set as the one or more candidate models. The feature estimation method according to claim 3 or 4, further comprising setting the plurality of regression models as they are in the one or more candidate models if it is determined that the setting method is appropriate.

7. The degrees of the aforementioned multiple regression models are different from each other. The feature estimation method according to claim 2, further comprising the model selection unit determining, based on the difference between the order of the model selected at the first measurement location and the order of the model selected at the second measurement location adjacent to the first measurement location, whether or not it is necessary to modify the model selected at the first measurement location or the second measurement location in order to reduce the difference, in the selection result of the selected model for each measurement location.

8. The feature estimation method according to any one of claims 1 to 4, further comprising the feature estimation device calculating, with respect to the one or more target parameters, the value or range of the first parameter included in the region as a complementary condition when there is a region in the setting range of multiple setting values ​​for the first parameter in the multiple preprocessing conditions in which the distribution density is relatively small.

9. The feature estimation method according to any one of claims 1 to 4, wherein the model generation unit generates the plurality of regression models, which includes generating the plurality of regression models to represent the relationship between the one or more pre-set target parameters and the shape features after the pre-processing according to each of the plurality of pre-processing conditions has been performed.

10. An estimation program for causing a computer to execute the feature estimation method described in any one of claims 1 to 4.

11. A feature quantity estimation device for estimating shape feature quantities that represent the shape characteristics of the surface of a substrate after it has undergone a predetermined process for semiconductor manufacturing, A model generation unit generates a plurality of regression models that represent the relationship between one or more target parameters included in the plurality of parameters and the shape feature quantities, based on measured data of the shape feature quantities on the surface of a substrate that has been subjected to pre-processing corresponding to the predetermined processing, according to each of a plurality of pre-processing conditions having a plurality of parameters for the predetermined processing. A model selection unit that selects one regression model from among the aforementioned multiple regression models, The system includes an estimation unit that uses the selection model to calculate an estimated value of the shape feature corresponding to any value of the one or more target parameters, The aforementioned model selection unit is For each of the aforementioned regression models, an index value representing the confidence level of the estimation result of the shape feature is calculated, To evaluate the suitability of the method for setting the aforementioned multiple pre-treatment conditions, A feature estimation device that performs the following: selecting the selected model based on the calculation results of the aforementioned index values ​​and the evaluation results of the suitability of the setting method.

12. The measured data includes the measured values ​​of the shape feature quantities at multiple measurement locations on the surface of the substrate. The model generation unit generates the plurality of regression models for each measurement location included in the plurality of measurement locations. The feature estimation device according to claim 11, wherein the model selection unit calculates the index value for each measurement location and selects the selected model for each measurement location based on the calculation result of the index value and the evaluation result of the suitability of the setting method.

13. The aforementioned model selection unit is Based on the evaluation results of the suitability of the setting method, one or more candidate models are selected from the plurality of regression models to be candidates for the selected model. A feature estimation device according to claim 11 or 12, which performs the following: selecting the selected model from the one or more candidate models based on the calculation result of the index value.

Images