Substrate processing system, schedule establishment method, storage medium, and schedule establishment program

Abstract

The present invention provides a substrate processing system capable of suppressing the usage of source power. The substrate processing system (1000) includes a substrate processing device (200) and a control device (300). The control device (300) establishes a schedule (SK) for the operation of the substrate processing device (200). The control device (300) includes a storage unit (303) and a control unit (304). The storage unit (303) stores a plurality of learned models for establishing a plurality of schedules (SK) with different usage of source power. The control unit (304) can establish a plurality of schedules (SK) based on the plurality of learned models. The plurality of learned models are constructed by performing reinforcement learning based on input data during learning. The input data during learning includes information on the number of pieces, process condition information, and source power information (RS). The source power information (RS) indicates the usage of source power used in each movement included in the steps of the operation of the substrate processing device (200).

Description

Technical Field

[0001] The present invention relates to a substrate processing system, a schedule establishment method, a storage medium, and a schedule establishment program. Background Art

[0002] A known schedule creation device arranges a plurality of blocks defining the contents of operations of a substrate processing apparatus along a time sequence, thereby creating a schedule defining the operations of the substrate processing apparatus along the time sequence (see, for example, Patent Document 1). The schedule creation device of Patent Document 1 outputs a schedule with the highest throughput.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2021-36582 Summary of the Invention

[0006] Problems to be solved by the invention

[0007] However, sometimes it is preferable to suppress power usage over production capacity. Situations where suppressing power usage over production capacity is preferable include situations where, for example, the supply of electricity or water is restricted at specified times in a location where substrate processing equipment is installed, or where there is a shortage of raw materials for processing fluids used for substrate processing (e.g., raw materials for chemical solutions or inert gases). While it is possible to adjust the various settings (values ​​of various parameters) of the substrate processing equipment to suppress power usage, this is time-consuming and labor-intensive. Furthermore, changing the settings may reduce the quality of substrate processing.

[0008] The present invention has been made in view of the above-mentioned problems, and an object thereof is to provide a substrate processing system, a schedule creation method, a storage medium, and a schedule creation program that can suppress the usage of source power.

[0009] Means used to solve problems

[0010] According to one embodiment of the present invention, a substrate processing system includes a substrate processing apparatus and a control apparatus. The substrate processing apparatus processes substrates using power. The control apparatus arranges a plurality of blocks defining the contents of the substrate processing apparatus's actions along a time series based on wafer count information, process condition information, and power information, thereby creating a schedule defining the substrate processing apparatus's actions along a time series. The wafer count information indicates the number of substrates to be processed by the substrate processing apparatus. The process condition information defines the steps of the substrate processing apparatus's actions. The power information indicates the amount of power used in each movement included in the steps of the substrate processing apparatus's actions. The control apparatus includes a storage unit and a control unit. The storage unit stores a plurality of learned models for creating a plurality of schedules with different amounts of power used. The control unit can create the plurality of schedules based on input data during schedule creation and the plurality of learned models. The input data during schedule creation includes the wafer count information, the process condition information, and the power information. The plurality of learned models are constructed by performing reinforcement learning based on the input data during learning. The learning input data includes the number of chips, the process condition information, and the source power information.

[0011] In one embodiment, the plurality of schedules include a first schedule and a second schedule, and the second schedule ends later than the first schedule.

[0012] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The second schedule has a smaller peak value of the source power usage than the first schedule.

[0013] In one embodiment, the plurality of schedules include a first schedule and a second schedule, wherein the second schedule uses less total amount of the source power than the first schedule.

[0014] In one embodiment, the source power includes at least one of a treatment fluid, electricity, and gas.

[0015] In one embodiment, the substrate processing apparatus includes a substrate processing unit for processing the substrate. The schedule creation input data and the learning input data further include information indicating a position of the substrate processing unit.

[0016] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The second schedule disperses timings at which the substrate processing apparatus starts processing the plurality of substrates compared to the first schedule.

[0017] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a substrate processing unit and a chemical tank. The substrate processing unit processes a first substrate using a first chemical. The substrate processing unit processes a second substrate using a second chemical. The chemical tank exclusively stores the first chemical and the second chemical for supply to the substrate processing unit. The second schedule involves fewer switching times between the first chemical and the second chemical in the chemical tank than in the first schedule.

[0018] In one embodiment, the substrate processing apparatus further includes a first substrate processing unit, a second substrate processing unit, a first chemical tank, and a second chemical tank. The first substrate processing unit processes the substrate using a first chemical tank or a second chemical tank. The second substrate processing unit processes the substrate using the first chemical tank or the second chemical tank. The first chemical tank is capable of exclusively storing the first chemical tank or the second chemical tank for supplying to the first substrate processing unit. The second chemical tank is capable of exclusively storing the first chemical tank or the second chemical tank for supplying to the second substrate processing unit. The plurality of schedules include a schedule for supplying the first chemical tank to the first substrate processing unit and a schedule for supplying the second chemical tank to the second substrate processing unit.

[0019] In one embodiment, the second chemical solution is of a different type from the first chemical solution.

[0020] In one embodiment, the second chemical solution has a different temperature or concentration than the first chemical solution.

[0021] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a plurality of substrate processing units that process the substrate. The second schedule causes a smaller number of substrate processing units to process the substrate than the first schedule.

[0022] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a substrate processing unit that processes the substrate. In the second schedule, a period during which the substrate processing unit does not process the substrate is shorter than in the first schedule.

[0023] In one embodiment, the substrate processing apparatus includes a substrate processing unit that processes the substrate. The substrate processing unit performs a specific process after a predetermined period of time has elapsed while not processing the substrate. The plurality of schedules includes a schedule for causing the substrate processing unit to continue processing the substrate after the predetermined period of time has elapsed.

[0024] In one embodiment, the substrate processing apparatus includes a substrate processing unit and a chemical tank. The substrate processing unit processes the substrate using a chemical tank. The chemical tank stores the chemical tank for supplying to the substrate processing unit. The substrate processing apparatus replaces the chemical tank after a predetermined period of time. The plurality of schedules include a schedule for causing the substrate processing unit to resume processing the substrate after the predetermined period of time has elapsed.

[0025] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a transport unit for transporting the substrate. The second schedule has fewer operations of the transport unit than the first schedule.

[0026] In one embodiment, the substrate processing apparatus includes a plurality of towers. The plurality of towers each include a plurality of substrate processing units for processing the substrate. The plurality of schedules include a schedule for causing the substrate processing unit included in a specific tower among the plurality of towers to process the substrate.

[0027] In one embodiment, the substrate processing apparatus includes a plurality of chemical tanks storing chemical solutions, and the plurality of schedules include a schedule for processing the substrate using the chemical solution stored in a specific chemical tank among the plurality of chemical tanks.

[0028] In one embodiment, the control unit creates a plurality of schedules based on a plurality of the learned models. The control unit obtains an end time for each of the plurality of schedules. The control unit selects a schedule from the plurality of schedules whose end time is within an allowable range and whose total amount of source power usage is minimized.

[0029] According to another embodiment of the present invention, a schedule creation method arranges a plurality of blocks defining the contents of the actions of a substrate processing apparatus that processes substrates using power along a time sequence, thereby creating a schedule defining the actions of the substrate processing apparatus along the time sequence. The schedule creation method includes a schedule creation step of creating at least one of a plurality of schedules having different amounts of power usage based on input data during schedule creation and at least one of a plurality of learned models. The schedule creation input data includes wafer count information, process condition information, and power information. The wafer count information indicates the number of substrates to be processed by the substrate processing apparatus. The process condition information defines the steps of the substrate processing apparatus's actions. The power information indicates the amount of power used for each movement included in the steps of the substrate processing apparatus's actions. The plurality of learned models are each constructed by performing reinforcement learning based on the learning input data. The learning input data includes the wafer count information, the process condition information, and the power information.

[0030] In one embodiment, the plurality of schedules include a first schedule and a second schedule, and the second schedule ends later than the first schedule.

[0031] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The second schedule has a smaller peak value of the source power usage than the first schedule.

[0032] In one embodiment, the plurality of schedules include a first schedule and a second schedule, wherein the second schedule uses less total amount of the source power than the first schedule.

[0033] In one embodiment, the source power includes at least one of a treatment fluid, electricity, and gas.

[0034] In one embodiment, the substrate processing apparatus includes a substrate processing unit for processing the substrate. The schedule creation input data and the learning input data further include information indicating a position of the substrate processing unit.

[0035] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The second schedule disperses timings at which the substrate processing apparatus starts processing the plurality of substrates compared to the first schedule.

[0036] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a substrate processing unit and a chemical tank. The substrate processing unit processes a first substrate using a first chemical. The substrate processing unit processes a second substrate using a second chemical. The chemical tank exclusively stores the first chemical and the second chemical for supply to the substrate processing unit. The second schedule involves fewer switching times between the first chemical and the second chemical in the chemical tank than in the first schedule.

[0037] In one embodiment, the substrate processing apparatus further includes a first substrate processing unit, a second substrate processing unit, a first chemical tank, and a second chemical tank. The first substrate processing unit processes the substrate using a first chemical tank or a second chemical tank. The second substrate processing unit processes the substrate using the first chemical tank or the second chemical tank. The first chemical tank is capable of exclusively storing the first chemical tank or the second chemical tank for supplying to the first substrate processing unit. The second chemical tank is capable of exclusively storing the first chemical tank or the second chemical tank for supplying to the second substrate processing unit. The plurality of schedules include a schedule for supplying the first chemical tank to the first substrate processing unit and a schedule for supplying the second chemical tank to the second substrate processing unit.

[0038] In one embodiment, the second chemical solution is of a different type from the first chemical solution.

[0039] In one embodiment, the second chemical solution has a different temperature or concentration than the first chemical solution.

[0040] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a plurality of substrate processing units that process the substrate. The second schedule causes a smaller number of substrate processing units to process the substrate than the first schedule.

[0041] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a substrate processing unit that processes the substrate. In the second schedule, a period during which the substrate processing unit does not process the substrate is shorter than in the first schedule.

[0042] In one embodiment, the substrate processing apparatus includes a substrate processing unit that processes the substrate. The substrate processing unit performs a specific process after a predetermined period of time has elapsed while not processing the substrate. The plurality of schedules includes a schedule for causing the substrate processing unit to continue processing the substrate after the predetermined period of time has elapsed.

[0043] In one embodiment, the substrate processing apparatus includes a substrate processing unit and a chemical tank. The substrate processing unit processes the substrate using a chemical tank. The chemical tank stores the chemical tank for supplying to the substrate processing unit. The substrate processing apparatus replaces the chemical tank after a predetermined period of time. The plurality of schedules include a schedule for causing the substrate processing unit to resume processing the substrate after the predetermined period of time has elapsed.

[0044] In one embodiment, the plurality of schedules include a first schedule and a second schedule. The substrate processing apparatus includes a transport unit for transporting the substrate. The second schedule has fewer operations of the transport unit than the first schedule.

[0045] In one embodiment, the substrate processing apparatus includes a plurality of towers. The plurality of towers each include a plurality of substrate processing units for processing the substrate. The plurality of schedules include a schedule for causing the substrate processing unit included in a specific tower among the plurality of towers to process the substrate.

[0046] In one embodiment, the substrate processing apparatus includes a plurality of chemical tanks storing chemical solutions, and the plurality of schedules include a schedule for processing the substrate using the chemical solution stored in a specific chemical tank among the plurality of chemical tanks.

[0047] In one embodiment, the schedule establishment process includes the following steps: establishing a plurality of the schedules based on a plurality of the learned models; obtaining the end time of each of the plurality of the schedules; and selecting a schedule from the plurality of the schedules whose end time is within an allowable range and whose total amount of source power usage is the least.

[0048] According to another embodiment of the present invention, a storage medium stores a schedule creation program for executing on a computer. The schedule creation program causes the computer to execute operations according to the above-mentioned schedule creation method.

[0049] According to another embodiment of the present invention, a schedule creation program is executed by a computer. The schedule creation program causes the computer to perform operations according to the above-mentioned schedule creation method.

[0050] Effects of the Invention

[0051] According to the substrate processing system, schedule creation method, storage medium, and schedule creation program of the present invention, the usage of source power can be suppressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a block diagram showing the structure of the learning device.

[0053] Figure 2 It is a plan view schematically showing an example of the configuration of a substrate processing apparatus included in the substrate processing system according to the first embodiment of the present invention.

[0054] Figure 3 It is a side view schematically showing an example of the configuration of a substrate processing apparatus included in the substrate processing system according to the first embodiment of the present invention.

[0055] Figure 4 This is a diagram showing an example of source power information.

[0056] Figure 5 This is a sequence diagram showing the flow of reinforcement learning processing.

[0057] Figure 6 This is a diagram showing the first example of the first schedule.

[0058] Figure 7 This is a diagram showing an example of the second schedule.

[0059] Figure 8 (a) and (b) are diagrams showing still another example of the second schedule.

[0060] Figure 9 This is a diagram showing the second example of the first schedule.

[0061] Figure 10(a) and (b) are diagrams showing the first example of the third schedule.

[0062] Figure 11 (a) is a diagram showing the third example of the first schedule, and (b) is a diagram showing the second example of the third schedule.

[0063] Figure 12 (a) is a diagram showing the fourth example of the first schedule. (b) is a diagram showing the first example of the fourth schedule.

[0064] Figure 13 (a) is a diagram showing the fifth example of the first schedule, and (b) is a diagram showing the second example of the fourth schedule.

[0065] Figure 14 It is a block diagram showing the configuration of a control device included in the substrate processing system according to the first embodiment of the present invention.

[0066] Figure 15 Mode information showing the relationship between the operation mode of the substrate processing apparatus included in the substrate processing system according to the first embodiment of the present invention and the scheduler is shown.

[0067] Figure 16 A diagram showing a mode selection screen.

[0068] Figure 17 This is a sequence diagram showing the flow of processing when creating a schedule.

[0069] Figure 18 It is a diagram showing a confirmation screen displayed on a display unit included in the substrate processing system according to the second embodiment of the present invention.

[0070] Figure 19 This is a flowchart showing the flow of processing executed by a control unit included in the substrate processing system according to the second embodiment of the present invention.

[0071] Figure 20 It is a diagram showing a confirmation screen displayed on a display unit included in the substrate processing system according to the third embodiment of the present invention.

[0072] Figure 21 This is a flowchart showing the flow of processing executed by a control unit included in the substrate processing system according to the third embodiment of the present invention.

[0073] Figure 22 A diagram showing a mode selection screen displayed on a display unit included in a substrate processing system according to a fourth embodiment of the present invention.

[0074] Figure 23 This is a flowchart showing the flow of processing executed by a control unit included in the substrate processing system according to the fifth embodiment of the present invention. DETAILED DESCRIPTION

[0075] Hereinafter, referring to the accompanying drawings ( Figures 1 to 23 ) to describe embodiments of the substrate processing system, schedule creation method, storage medium, and schedule creation program of the present invention. However, the present invention is not limited to the following embodiments and can be implemented in various embodiments without departing from the gist of the present invention. Furthermore, where the description overlaps, the description may be omitted as appropriate. In the figures, identical or corresponding portions are denoted by the same reference numerals, and description thereof will not be repeated.

[0076] The "substrate" in the embodiments of the present invention can be applied to various substrates, such as semiconductor wafers, glass substrates for photomasks, glass substrates for liquid crystal displays, glass substrates for plasma displays, substrates for FEDs (Field Emission Displays), substrates for optical disks, magnetic disks, and magneto-optical disks. The following describes the embodiments of the present invention primarily using a substrate processing system, scheduling method, storage medium, and scheduling program for processing disk-shaped semiconductor wafers as an example, but the present invention can also be similarly applied to the processing of various substrates exemplified above. Furthermore, various substrate shapes are applicable.

[0077] [Implementation Method 1]

[0078] First, refer to Figure 1 and Figure 2 The learning device 100 will be described. Figure 1 2 is a block diagram showing the configuration of the learning device 100 . Figure 2 It is a plan view schematically showing an example of the configuration of the substrate processing apparatus 200 included in the substrate processing system 1000 according to the present embodiment.

[0079] like Figure 2 As shown, the substrate processing system 1000 of this embodiment includes a substrate processing apparatus 200 and a control apparatus 300 . Figure 1 The learning device 100 shown here creates a schedule creation program through reinforcement learning. The schedule creation program is a program for creating a schedule SK that specifies the operations of the substrate processing apparatus 200 along a time series. The schedule creation program includes a learned model constructed through reinforcement learning.

[0080] Specifically, the learning device 100 includes a machine learning model M. The learning device 100 uses the machine learning model M to create a schedule SK. The machine learning model M is a reinforcement learning model. For example, the machine learning model M may be a deep reinforcement learning model. The learning device 100 repeatedly creates the schedule SK, thereby constructing a learned model.

[0081] The reinforcement learning environment is the substrate processing apparatus 200, and the learning apparatus 100 stores information indicating the configuration of the substrate processing apparatus 200. The movement subjects of the reinforcement learning are the components of the substrate processing apparatus 200, and the movement of the reinforcement learning is the movement of the components of the substrate processing apparatus 200.

[0082] Specifically, the learning device 100 stores schematic configuration information that schematically represents the configuration of the substrate processing apparatus 200. Specifically, Figure 2 As shown, the substrate processing apparatus 200 includes: a plurality of loading ports LP, an indexing robot IR, a delivery unit PASS, a central robot CR, a plurality of substrate processing units MC, and at least one chemical liquid cabinet CC. In this embodiment, the at least one chemical liquid cabinet CC includes a first chemical liquid cabinet CC1 and a second chemical liquid cabinet CC2. The schematic configuration information shows: one of the plurality of loading ports LP, the indexing robot IR, the delivery unit PASS, the central robot CR, and a plurality of substrate processing units MC. The moving entities of the reinforcement learning include the indexing robot IR, the central robot CR, and a plurality of substrate processing units MC, and the movement of the reinforcement learning includes the movement of the indexing robot IR, the movement of the central robot CR, and the movement of the plurality of substrate processing units MC. Furthermore, the schematic configuration information does not include information indicating the position of the substrate processing unit MC. Furthermore, the schematic configuration information does not include information indicating the connection relationship between the substrate processing unit MC and the chemical liquid cabinet CC.

[0083] Figure 2 The control device 300 shown stores Figure 1 The schedule creation program created by the learning device 100 is shown. When a substrate storage container C docks with any of the plurality of load ports LP, the control device 300 creates a schedule SK based on the schedule creation program. The schedule SK specifies the actions of the substrate processing apparatus 200 in a time-sequential manner when substrate processing is performed on the plurality of substrates W stored in the substrate storage container C docked with the load ports LP. Specifically, the schedule SK specifies the movement of the index robot IR, the movement of the center robot CR, and the movement of the plurality of substrate processing units MC in a time-sequential manner.

[0084] In the following description, substrate processing performed on a plurality of substrates W stored in the substrate storage container C docked at the load port LP may be referred to as “this project PJ”.

[0085] Reference Figure 1 , the learning device 100 is further described. Figure 1 As shown, the learning device 100 includes an input unit 101 , a display unit 102 , a storage unit 103 , and a processing unit 104 .

[0086] The input unit 101 is a user interface device operated by the operator. The input unit 101 inputs instructions (control signals) corresponding to the operator's operations into the processing unit 104. Furthermore, the input unit 101 inputs data corresponding to the operator's operations into the processing unit 104. The input unit 101 may include a keyboard and a mouse. The input unit 101 may also include a touch sensor superimposed on the display surface of the display unit 102. By superimposing the touch sensor on the display surface of the display unit 102, a graphical user interface may be formed.

[0087] For example, the operator can operate the input unit 101 to input the current state of the device, the content of the started process (the content of the current project PJ), the source power information RS, and the additional remuneration into the calculation processing unit 104 .

[0088] The current state of the device indicates the current state of each component of the substrate processing device 200. For example, the current state of the device indicates: the current state of the index robot IR, the current state of the center robot CR, and the current state of the plurality of substrate processing units MC. The operator assumes an arbitrary state as the current state of the device and inputs it. In the case where it is assumed that a substrate W remains inside the substrate processing device 200, the current state of the device further includes the position of the substrate W remaining inside the substrate processing device 200. For example, in the case where the index robot IR holds the substrate W, the current state of the device includes information indicating that the substrate W is held by the index robot IR.

[0089] The content of the processing to be started (the content of this project PJ) includes wafer count information and process condition information. The wafer count information indicates the number of substrates W to be processed by the substrate processing apparatus 200. Specifically, the wafer count information indicates the number of substrates W stored in the substrate storage container C docked with the load port LP. The operator assumes an arbitrary number of wafers and enters the wafer count information.

[0090] The process condition information defines the steps of the operation of the substrate processing apparatus 200 on the substrate W. Specifically, the process condition information defines the steps of the operation of the substrate processing apparatus 200 for the current project PJ. The operator assumes and inputs arbitrary contents as the contents of the process to be started.

[0091] More specifically, the process condition information includes: process conditions, pre-process conditions, post-process conditions, and flow process conditions. The process conditions include substrate treatment process conditions R and chemical solution process conditions KC.

[0092] The substrate processing process conditions R define the details of the processing performed on the substrate W. Specifically, the substrate processing process conditions R include the procedure for the processing performed on the substrate W and the conditions for the processing. The chemical process conditions KC define the details of the operation of the chemical cabinet CC. Specifically, the chemical process conditions KC include the procedure for the processing performed by the chemical cabinet CC and the conditions for the chemical solutions stored in the chemical cabinet CC.

[0093] The pre-process conditions define the content of the pre-processing performed by the substrate processing unit MC. The pre-processing is performed at the beginning of the current project PJ. For example, the pre-processing includes a pre-distribution process. The pre-distribution process means that the pre-distribution process is performed by the substrate processing unit MC. Figure 3 ) The process of discharging the liquid inside the piping for supplying the liquid from the nozzle 22.

[0094] The post-process conditions define the content of post-processing. Post-processing is performed at the end of this project PJ. For example, post-processing includes the following: Figure 3 The cleaning process for the chamber 2a may include cleaning the chuck pins holding the substrate W or cleaning other parts (eg, a processing cup) in the chamber 2a.

[0095] The process conditions define the execution order and execution times of control according to process conditions (process condition control), control according to pre-process conditions (pre-process condition control), and control according to post-process conditions (post-process condition control).

[0096] The source power information RS indicates the amount of source power used. For example, the source power information RS indicates at least one of the amount of processing liquid used, the amount of electricity used, and the amount of gas used. The operator can input information indicating the amount of source power used for each movement that can be performed by the substrate processing apparatus 200 as the source power information RS. Alternatively, the operator can input the amount of source power used in the current project PJ as the source power information RS. Specifically, the operator can input the amount of source power used for each movement included in the steps of the operation of the substrate processing apparatus 200 based on the current project PJ.

[0097] The additional reward is added to the reward assigned to the schedule SK by the learning device 100. Specifically, the operator uses the learning device 100 to create multiple schedule creation programs. Each of the multiple schedule creation programs creates a schedule SK. The operator adds the additional reward so that the multiple schedule creation programs create multiple schedules SK with different amounts of source power usage. Hereinafter, the reward assigned to the schedule SK by the learning device 100 may be referred to as the "initial reward." Alternatively, the reward obtained by adding the additional reward to the initial reward may be referred to as the "final reward."

[0098] Furthermore, the input unit 101 may further include an interface for accessing removable media. Removable media may include, for example, a memory card such as an SD card. Removable media may include, for example, a USB memory. Removable media may include, for example, an optical disc such as a CD (compact disc) or a DVD. The interface may include, for example, a slot for inserting a memory card. The interface may include, for example, a USB terminal. The interface may include, for example, a reader for reading data from an optical disc. A USB cable for a memory card reader may also be connected to the USB terminal.

[0099] The display unit 102 displays various screens or images. For example, the display unit 102 displays a screen for the operator to operate the learning device 100. Furthermore, the display unit 102 displays the schedule SK output by the machine learning model M. The operator confirms the schedule SK displayed on the display unit 102 and inputs additional rewards. The display unit 102 includes, for example, a liquid crystal display device or an organic EL (electroluminescence) display device.

[0100] The storage unit 103 includes a main storage device. The main storage device includes, for example, a semiconductor memory. The storage unit 103 further includes an auxiliary storage device. The auxiliary storage device includes, for example, at least one of a semiconductor memory and a hard disk drive. The storage unit 103 may include a removable medium.

[0101] The storage unit 103 stores various computer programs and various data. Various data include reference Figure 1 and Figure 2 The general structure information of the description. Various computer programs include programs for machine learning. Programs for machine learning include machine learning models M. More specifically, programs for machine learning include programs for reinforcement learning. The algorithm for reinforcement learning is not particularly limited, and for example, it can be an algorithm that follows Q learning, SARSA method, policy gradient method, Actor-Critic method, or Monte Carlo method. The algorithm for reinforcement learning constructs a prediction model as a learned model. Prediction model prediction reference Figure 5 The "evaluation value (Q value) of each move that can be executed next" is explained.

[0102] The prediction model includes, for example, a neural network. A neural network includes an input layer, a single or multiple intermediate layers, and an output layer. Specifically, neural networks include deep neural networks (DNNs), recurrent neural networks (RNNs), convolutional neural networks (CNNs), or quantum neural networks (QNNs), which perform deep learning. For example, a deep neural network includes an input layer, multiple intermediate layers (hidden layers), and an output layer.

[0103] The processing unit 104 includes a processor. The processing unit 104 may include, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Alternatively, the processing unit 104 may include a general-purpose computer, a dedicated computing unit, a GPU (Graphics Processing Unit), an NPU (Neural Network Processing Unit), or a quantum computer. A dedicated computing unit may include, for example, an ASIC (Application Specific Integrated Circuit).

[0104] The processing unit 104 executes various processes based on the various programs and data stored in the storage unit 103. For example, the processing unit 104 creates a schedule SK based on a reinforcement learning program. Specifically, the processing unit 104 uses the machine learning model M to create the schedule SK and obtains a reward (initial reward) for the created schedule SK. The processing unit 104 repeatedly creates the schedule SK to construct a learned model (prediction model). As a result, a schedule creation program that includes the learned model (prediction model) is generated.

[0105] Here, refer to Figure 2 and Figure 3 , describing the structure of the substrate processing apparatus 200. Figure 3 This is a side view schematically illustrating an example of the configuration of a substrate processing apparatus 200 included in the substrate processing system 1000 according to this embodiment. As already described, the substrate processing apparatus 200 includes a plurality of load ports LP, an index robot IR, a delivery unit PASS, a center robot CR, a plurality of substrate processing units MC, a first chemical tank CC1, and a second chemical tank CC2.

[0106] A substrate storage container C is docked at the loading port LP. Specifically, a substrate storage container C for storing unprocessed substrates W is docked at a portion of the multiple loading ports LP. An empty substrate storage container C is docked at the remaining loading ports LP. The processed substrates W are stored in the empty substrate storage container C. The substrate storage container C stores a plurality of substrates W in a stacked state. Specifically, a plurality of substrates W are stacked in the vertical direction in a horizontal posture with intervals between them in the substrate storage container C. Here, the so-called horizontal posture refers to a state in which the thickness direction of the substrate W is along the vertical direction. The substrate storage container C may be, for example, a FOUP (Front Opening Unified Pod), a SMIF (Standard Mechanical InterFace) box, or an OC (Open Cassette).

[0107] The index robot IR transfers unprocessed substrates W from the substrate storage container C to the delivery interface PASS, and transfers processed substrates W from the delivery interface PASS to the substrate storage container C. The index robot IR is an example of a "transfer unit." Specifically, the index robot IR includes a base unit 8, a multi-jointed arm 9, and two hands 10A and 10B.

[0108] The base portion 8 supports the base end portion of the multi-joint arm 9. More specifically, the base portion 8 supports the base end portion of the multi-joint arm 9 so that it can be freely rotated and raised and lowered. Specifically, the base portion 8 has an arm rotation mechanism and an arm lifting mechanism. The arm rotation mechanism rotates the multi-joint arm 9 around a vertical axis. More specifically, the arm rotation mechanism rotates the multi-joint arm 9 in both forward and reverse directions. The arm lifting mechanism lifts and lowers the multi-joint arm 9. Therefore, the multi-joint arm 9 can rotate freely relative to the base portion 8. Furthermore, the multi-joint arm 9 can be freely raised and lowered relative to the base portion 8.

[0109] The multi-jointed arm 9 is capable of flexible flexion and extension in the horizontal direction. Specifically, the multi-jointed arm 9 comprises a plurality of arms, a plurality of joints, and a plurality of independent rotation mechanisms. Two arms are connected to each joint. Each arm is capable of free rotation relative to the connected joint. The independent rotation mechanisms are provided at the joints to rotate the corresponding arms along the horizontal plane. Specifically, the independent rotation mechanisms rotate the corresponding arms in both forward and reverse directions.

[0110] The hands 10A and 10B are supported by the front end of the multi-jointed arm 9. The multi-jointed arm 9 further includes a hand rotation mechanism that allows the hands 10A and 10B to rotate independently about a vertical axis, and a hand advance and retreat mechanism that allows the hands 10A and 10B to advance and retreat independently in the horizontal direction. Therefore, the hands 10A and 10B can rotate freely and independently relative to the multi-jointed arm 9. Specifically, the hands 10A and 10B can rotate freely in both the forward and reverse directions. Furthermore, the hands 10A and 10B can advance and retreat freely and independently relative to the multi-jointed arm 9.

[0111] The hands 10A and 10B each hold a substrate W. Furthermore, the hands 10A and 10B can be arranged in an overlapping manner. Figure 2 In FIG. 1 , for clarity, the hands 10A and 10B are shown offset in a direction parallel to the paper surface (horizontal direction).

[0112] Next, the delivery interface PASS is described. The delivery interface PASS includes a plurality of racks (not shown) for supporting substrates W. Specifically, the delivery interface PASS includes at least one rack (not shown) for supporting unprocessed substrates W, and at least one rack (not shown) for supporting processed substrates W. In this embodiment, the delivery interface PASS includes two racks (not shown) for supporting unprocessed substrates W, and two racks (not shown) for supporting processed substrates W.

[0113] Next, the central robot CR will be described. The central robot CR transports unprocessed substrates W from the delivery unit PASS to the substrate processing unit MC, and transports processed substrates W from the substrate processing unit MC to the delivery unit PASS. The central robot CR is an example of a "transport unit." Specifically, the central robot CR includes a base unit 11, a multi-jointed arm 12, and two hands 13A and 13B.

[0114] The base portion 11 supports the base end portion of the multi-joint arm 12. More specifically, the base portion 11 supports the base end portion of the multi-joint arm 12 so that it can be rotated and raised and lowered freely. Specifically, the base portion 11 has an arm rotation mechanism and an arm lifting mechanism. The arm rotation mechanism rotates the multi-joint arm 12 around a vertical axis. More specifically, the arm rotation mechanism rotates the multi-joint arm 12 in both forward and reverse directions. The arm lifting mechanism lifts and lowers the multi-joint arm 12. Therefore, the multi-joint arm 12 can rotate freely relative to the base portion 11. Furthermore, the multi-joint arm 12 can be raised and lowered freely relative to the base portion 11.

[0115] The multi-jointed arm 12 is capable of flexible flexion and extension in the horizontal direction. Specifically, the multi-jointed arm 12 comprises a plurality of arms, a plurality of joints, and a plurality of independent rotation mechanisms. Two arms are connected to each joint. Each arm is capable of free rotation relative to the connected joint. The independent rotation mechanisms are provided at the joints to rotate the corresponding arms along the horizontal plane. Specifically, the independent rotation mechanisms rotate the corresponding arms in both forward and reverse directions.

[0116] The hands 13A and 13B are supported by the front end of the multi-jointed arm 12. The multi-jointed arm 12 further includes a hand rotation mechanism that allows the hands 13A and 13B to rotate independently about a vertical axis, and a hand advance and retreat mechanism that allows the hands 13A and 13B to advance and retreat independently in the horizontal direction. Therefore, the hands 13A and 13B can rotate freely and independently relative to the multi-jointed arm 12. Specifically, the hands 13A and 13B can rotate freely in both the forward and reverse directions. Furthermore, the hands 13A and 13B can advance and retreat freely and independently relative to the multi-jointed arm 12.

[0117] The hands 13A and 13B each hold a substrate W. Furthermore, the hands 13A and 13B can be arranged in an overlapping manner. Figure 2 In FIG. 1 , for clarity, the hands 13A and 13B are shown offset in a direction parallel to the paper surface (horizontal direction).

[0118] Next, the substrate processing unit MC will be described. Figure 2 and Figure 3 The illustrated substrate processing apparatus 200 is a single-wafer apparatus, with a substrate processing unit MC processing each substrate W individually. In this embodiment, the substrate processing unit MC supplies a processing liquid to the substrates W to process them. The processing liquid is not particularly limited; it can be any liquid that comes into contact with the substrates W. In this embodiment, the processing liquid includes a chemical solution and a rinse liquid.

[0119] The chemical solution may be, for example, dilute hydrofluoric acid (DHF), hydrofluoric acid (HF), nitric acid fluoride (a mixture of hydrofluoric acid and nitric acid (HNO3)), buffered hydrofluoric acid (BHF), ammonium fluoride, HFEG (a mixture of hydrofluoric acid and ethylene glycol), phosphoric acid (H3PO4), sulfuric acid, acetic acid, nitric acid, hydrochloric acid, ammonia water, hydrogen peroxide, organic acid (e.g., citric acid, oxalic acid), organic base (e.g., TMAH: tetramethylammonium hydroxide), sulfuric acid hydrogen peroxide mixture (SPM), ammonia hydrogen peroxide mixture (SC1), hydrochloric acid hydrogen peroxide mixture (SC2), isopropyl alcohol (IPA), surfactant, or corrosion inhibitor.

[0120] The rinse liquid is, for example, pure water (eg, deionized water), carbonated water, electrolytic ionized water, hydrogen water, ozone water, or hydrochloric acid water with a diluted concentration (eg, about 10 ppm to 100 ppm).

[0121] like Figure 3 As shown, the substrate processing unit MC may include, for example, a chamber 2a, a spin chuck 21, and a nozzle 22. The chamber 2a has a roughly box-like shape and houses the spin chuck 21 and the nozzle 22. Furthermore, the chamber 2a houses the substrate W that has been transferred into the chamber 2a. The spin chuck 21 holds and rotates a single substrate W in a horizontal position. The nozzle 22 sprays a processing liquid onto the substrate W held by the spin chuck 21. The nozzle 22 can spray the processing liquid onto the rotating substrate W. Furthermore, the nozzle 22 can supply gas to the substrate W.

[0122] like Figure 2 and Figure 3 As shown, the plurality of substrate processing units MC form a plurality of towers TW arranged so as to surround the central robot CR in a plan view. In this embodiment, the plurality of towers TW include a first tower TW1 through a fourth tower TW4 (four towers TW). Each tower TW includes a plurality of substrate processing units MC (here, three substrate processing units MC) stacked one above the other.

[0123] Next, the chemical liquid cabinet CC is described. The chemical liquid cabinet CC stores chemical liquid. The chemical liquid cabinet CC supplies chemical liquid to the substrate processing unit MC. Specifically, Figure 2 As shown, the first chemical tank CC1 supplies chemical liquid to the substrate processing parts MC constituting the first tower TW1 and the second tower TW2 , and the second chemical tank CC2 supplies chemical liquid to the substrate processing parts MC constituting the third tower TW3 and the fourth tower TW4 .

[0124] In detail, a piping P1 is connected to the chemical cabinet CC. The piping P1 allows the chemical liquid to flow from the power source equipment of the factory where the substrate processing apparatus 200 is installed to the chemical cabinet CC. When the chemical liquid is an aqueous solution, a piping P2 is further connected to the chemical cabinet CC. The piping P2 allows pure water (for example, deionized water) to flow from the power source equipment of the factory where the substrate processing apparatus 200 is installed to the chemical cabinet CC. The substrate processing apparatus 200 further includes an on-off valve (not shown) arranged on the piping P1. The control device 300 controls the on-off state of the on-off valve, controls the supply of the chemical liquid to the chemical cabinet CC, and stops the supply of the chemical liquid. When the chemical liquid is an aqueous solution, the substrate processing apparatus 200 further includes an on-off valve (not shown) arranged on the piping P2. The control device 300 controls the on-off state of the on-off valve, controls the supply of pure water to the chemical cabinet CC, and stops the supply of pure water.

[0125] like Figure 3 As shown, the chemical tank CC may include a storage unit 3, a temperature control unit 4, and a pump 5. The storage unit 3 stores chemical liquid. The temperature control unit 4 controls the temperature of the chemical liquid. The pump 5 delivers the chemical liquid to the substrate processing unit MC.

[0126] Next, the substrate processing apparatus 200 will be further described. Figure 3 As shown, the substrate processing apparatus 200 may further include a drain pipe 6 and an on-off valve 7. One end of the drain pipe 6 is connected to the storage unit 3. The on-off valve 7 is disposed on the drain pipe 6.

[0127] The control device 300 controls the open / close state of the on / off valve 7. For example, the control device 300 controls the open / close state of the on / off valve 7 when replacing the chemical solution in the storage unit 3 with a new solution.

[0128] Specifically, the control device 300 changes the on-off valve 7 from the closed state to the open state, and discharges the liquid in the storage unit 3 through the drain pipe 6. When the storage unit 3 becomes empty, the control device 300 changes the on-off valve 7 from the open state to the closed state. Figure 2 The on-off valve (not shown) of the pipe P1 shown in FIG. 1 is changed from the closed state to the open state. As a result, the chemical liquid is supplied from the power source equipment of the factory to the storage unit 3. When the chemical liquid is an aqueous solution, the control device 300 further causes the valve 300 to be configured at Figure 2 The on-off valve (not shown) of the pipe P2 shown is switched from the closed state to the open state. As a result, pure water is supplied from the power source equipment of the factory to the storage unit 3.

[0129] Next, refer to Figures 1 to 4 , the movement of each component of the substrate processing apparatus 200 and the source power information RS will be described. Figure 4 : is a diagram showing an example of source power information RS. Figure 4 As shown, the source power information RS associates the movement of each component of the substrate processing apparatus 200 with the type of source power and the usage amount of the source power.

[0130] First, the movement of each component of the substrate processing apparatus 200 will be described. Figure 4 As shown, the movement of each component of the substrate processing apparatus 200 may include, for example, movements AC1 to AC10.

[0131] Movement AC1 indicates the movement of the index robot IR unloading an unprocessed substrate W from the substrate storage container C. Movement AC2 indicates the movement of the index robot IR loading an unprocessed substrate W into the delivery interface PASS. Movement AC3 indicates the movement of the center robot CR unloading an unprocessed substrate W from the delivery interface PASS. Movement AC4 indicates the movement of the center robot CR loading an unprocessed substrate W into the substrate processing section MC. Movement AC5 indicates the movement of the center robot CR unloading a processed substrate W from the substrate processing section MC. Movement AC6 indicates the movement of the center robot CR loading a processed substrate W into the delivery interface PASS. Movement AC7 indicates the movement of the index robot IR unloading a processed substrate W from the delivery interface PASS. Movement AC8 indicates the movement of the index robot IR loading a processed substrate W into the substrate storage container C. Movement AC9 indicates the movement of the substrate processing section MC processing a substrate W based on process conditions A. Movement AC10 indicates the movement of the substrate processing section MC processing a substrate W based on process conditions B.

[0132] Furthermore, when the substrate processing unit MC processes the substrate W based on process conditions A, the chemical solution A and the rinse liquid are used. Furthermore, when the substrate processing unit MC processes the substrate W based on process conditions B, the chemical solution B, the rinse liquid, and nitrogen gas are used.

[0133] Next, the source power information RS will be described. The source power information RS indicates the type and amount of source power used in the movement unit of each component of the substrate processing apparatus 200. That is, the source power information RS indicates the type and amount of source power used in each movement. Figure 4 As shown, the source power used by the index robot IR and the center robot CR is electricity (movements AC1 to AC8). When the substrate processing unit MC processes substrates W under process conditions A, the source power used is electricity, chemical solution A, and rinse liquid (movement AC9). When the substrate processing unit MC processes substrates W under process conditions B, the source power used is electricity, chemical solution B, rinse liquid, and nitrogen gas (movement AC10). Source power information RS can be input to the processing unit 104, for example, from a removable medium.

[0134] Here, the method of creating the schedule SK is described. The operation processing unit 104 creates the schedule SK by arranging a plurality of blocks BL that define the contents of the operations of the substrate processing apparatus 200 along a time series. The blocks BL correspond to the movements of the components of the substrate processing apparatus 200. For example, the operation processing unit 104 arranges the blocks BL along a time series. Figure 4 The schedule SK is created by aligning the blocks BL corresponding to the movements AC1-AC10 shown. Specifically, the processing unit 104 allocates each block BL to the schedule. As already explained, the blocks BL correspond to the movements of each component. Therefore, the source power information RS indicates the source power usage for each block BL.

[0135] Next, refer to Figures 1 to 5 , the process of reinforcement learning processing is explained. Figure 5 This is a time sequence diagram showing the process of reinforcement learning. The calculation processing unit 104 performs reinforcement learning based on the input data during learning, including the current state of the device, the content of the started process (the content of this project PJ), and the source power information RS, and constructs a learned model. Figure 5 As shown, in reinforcement learning, the processes of step S1 to step S12 are repeatedly executed.

[0136] In detail, the operation processing unit 104 creates a list of all movements required for this project PJ based on the current state of the device and the content of the processing to be started (the content of this project PJ) (step S1). In this embodiment, the list of all movements includes: all movements of the index robot IR, all movements of the center robot CR, and all movements of the substrate processing unit MC. Here, all movements until the end of this project PJ are included. For example, in this project PJ, when processing N substrates W accommodated in the substrate storage container C, all movements of the index robot IR include N movements AC1 (refer to Figure 4 ).

[0137] After creating the list of all moves, the processing unit 104 refers to the current device state and the list of all moves to create a list of moves that can be executed next (step S2). Hereinafter, the list of moves that can be executed next may be referred to as a "list of subsequent moves."

[0138] For example, when the target substrate W to be transported by the index robot IR is stored in the substrate storage container C, the index robot IR does not hold the substrate W, and the target substrate W to be transported by the index robot IR is not placed on the delivery portion PASS, the list of all movements includes movements AC1, AC2, AC7, and AC8 (see Figure 4 ), the arithmetic processing unit 104 obtains only movement AC1 among movements AC1, AC2, AC7, and AC8 as the next executable movement.

[0139] When creating a list of subsequent moves, the processing unit 104 refers to the source power information RS to obtain the type and amount of source power used in each of the next executable moves. Hereinafter, the type and amount of source power used in each of the next executable moves may be referred to as "source power of the subsequent move."

[0140] After creating the list of subsequent moves, the processing unit 104 performs pre-processing on the input data (step S3), including the next possible moves, the current device state, and the driving force behind the subsequent moves. This pre-processing obtains feature quantities from the input data. For example, the pre-processing may include at least one of dimensionality reduction and convolution.

[0141] The processing unit 104 inputs the pre-processed input data into the neural network model (step S4). As a result, the neural network model outputs the Q value of each move that can be executed next. The neural network model is an example of a machine learning model M (reinforcement learning model). By repeatedly learning through the neural network model, a prediction model (learned model) is constructed. Figure 5 As shown, a neural network model can include an input layer, a feature acquisition layer (intermediate layer), a combination layer (intermediate layer), and an output layer. The feature acquisition layer acquires features corresponding to a plurality of predetermined evaluation criteria. The combination layer combines the Q values ​​of the evaluation criteria.

[0142] The calculation processing unit 104 performs post-processing on the Q value output from the neural network model and calculates the evaluation value of each move that can be executed next (step S5). The evaluation value reflects the Q value of a plurality of evaluation items.

[0143] The processing unit 104 determines the next move to be executed from the next possible moves (step S6). The processing unit 104 may determine the next move to be executed based on, for example, the ε-greedy method. In this case, the next move to be executed is randomly selected during the learning process. Specifically, when the value of ε (Epsilon) is greater than a specified value, the processing unit 104 randomly selects the next move to be executed. When the value of ε (Epsilon) is less than a specified value, the processing unit 104 selects the move with the highest evaluation value among the next possible moves as the next move to be executed.

[0144] After determining the next movement to be executed, the calculation processing unit 104 arranges the blocks BL corresponding to the determined movement in the timetable along the time series and updates the timetable. As a result, the determined movement becomes the completed movement.

[0145] After updating the schedule, the processing unit 104 updates the current device state to the state after the determined movement is executed (step S7). After updating the schedule, the processing unit 104 deletes the completed movement from the list of all movements required for the current project PJ and updates the list of all movements (step S8).

[0146] The calculation processing unit 104 repeats the process of steps S1 to S8 until all moves are deleted from the list of all moves required for the current project PJ. As a result, a schedule SK is created (step S9). The calculation processing unit 104 displays the created schedule SK on the display unit 102.

[0147] Furthermore, the calculation processing unit 104 calculates the remuneration (initial remuneration) to be given to the created schedule SK (step S10). Specifically, the calculation processing unit 104 calculates the initial remuneration based on the evaluation value calculated in the post-processing (step S5).

[0148] The operator confirms the schedule SK displayed on the display unit 102 and operates the input unit 101 to input the additional remuneration. The final remuneration is calculated by the processing unit 104 (step S11). The additional remuneration is input by the operator based on the purpose of the schedule creation program. For example, the operator can determine the value of the additional remuneration based on the evaluation items set for calculating the Q value in the neural network model.

[0149] After calculating the final reward, the processing unit 104 adjusts the weighting coefficients of the neural network model based on the final reward (step S12). As a result, the weighting coefficients of the evaluation items set in the neural network model for calculating the Q value are adjusted. Specifically, the processing unit 104 adjusts the weighting coefficients of the evaluation items so that the evaluation value of each move calculated in post-processing (step S5) matches the final reward.

[0150] After adjusting the weighting coefficients, the calculation processing unit 104 executes the processes of steps S1 to S12 to adjust the weighting coefficients again. Specifically, the calculation processing unit 104 repeats the processes of steps S1 to S12 until the evaluation value of each move calculated in the post-processing (step S5) reaches a value consistent with the initial reward.

[0151] Next, refer to Figures 1 to 5 , the evaluation items given to the machine learning model M are explained. Figure 5 In the example shown, the evaluation items assigned to the machine learning model M are evaluation items set in the neural network model to calculate the Q value. The evaluation items assigned to the machine learning model M are changed according to the purpose of the scheduler.

[0152] Specifically, the operator generates a plurality of schedule creation programs using the learning device 100. Each of the plurality of schedule creation programs creates a schedule SK with different source power usage. In this embodiment, the operator generates first to fourth schedule creation programs using the learning device 100.

[0153] The first schedule creation process creates a schedule SK that prioritizes production capacity. The second schedule creation process creates a schedule SK that disperses peak power usage. The third schedule creation process creates a schedule SK that uses less power than the first schedule creation process. The fourth schedule creation process creates a schedule SK that uses less power than the first schedule creation process.

[0154] Hereinafter, the schedule SK for creating the first schedule creation program may be referred to as “first schedule SK1.” Similarly, the schedules SK for creating the second to fourth schedule creation programs may be referred to as “second schedule SK2” to “fourth schedule SK4,” respectively.

[0155] When creating the first schedule creation program, the only evaluation factor is production capacity. Therefore, the earlier the end time of the schedule SK, the greater the evaluation value and reward (initial reward). Furthermore, when creating the first schedule creation program, the operator does not need to enter additional rewards.

[0156] The evaluation items used when creating the second schedule creation program include production capacity and peak power usage. The more dispersed the peak power usage of the operators, the greater the additional remuneration. As a result, the weighting coefficient for the peak power usage increases, and the more dispersed the peak power usage, the greater the evaluation value and remuneration (initial remuneration). Therefore, the peak power usage of the second schedule SK2 is smaller than that of the first schedule SK1. Furthermore, it is feasible that the greater the weighting coefficient for the peak power usage of the operator, the smaller the weighting coefficient for production capacity.

[0157] The evaluation items for creating the third schedule creation program include production capacity, the number of times the index robot IR transfers substrates W, and the number of times the center robot CR transfers substrates W. Hereinafter, the number of times the index robot IR transfers substrates W and the number of times the center robot CR transfers substrates W are sometimes referred to as "the number of transfers by the transfer unit."

[0158] The fewer times the transport unit makes transfers, the more the operator increases the additional remuneration. As a result, the weighting coefficient for the number of transfers by the transport unit increases. The fewer times the transport unit makes transfers, the greater the evaluation value and remuneration (initial remuneration). Therefore, schedule 3 SK3 involves fewer movements of the index robot IR and the center robot CR compared to schedule 1 SK1. By reducing the number of transfers of substrates W by the index robot IR and the center robot CR, power usage can be suppressed. Furthermore, it is feasible that the greater the weighting coefficient for the number of transfers by the transport unit, the more the processing unit 104 reduces the weighting coefficient for production capacity.

[0159] The evaluation items when creating the third schedule creation program may further include the number of times the chemical solution is switched in the chemical solution cabinet CC. Hereinafter, the number of times the chemical solution is switched in the chemical solution cabinet CC may be referred to as the "number of chemical solution switching times."

[0160] The fewer times the chemical solution is switched, the more the operator increases the additional reward. As a result, the weighting factor for the number of chemical solution switches increases. The fewer times the chemical solution is switched, the greater the evaluation value and reward (initial reward) are. By reducing the number of chemical solution switches, chemical solution usage can be reduced. Therefore, schedule 3 SK3 can reduce source power usage compared to schedule 1 SK1. Furthermore, it is feasible that the greater the weighting factor for the number of chemical solution switches, the smaller the weighting factor for production capacity applied by the processing unit 104.

[0161] The evaluation items when creating the fourth schedule creation program include production capacity and the number of parallel units. Here, the number of parallel units represents the number of substrate processing units MC that process substrates W in the current project PJ.

[0162] The fewer substrate processing units MC used in this project PJ, the more the operator increases the additional remuneration. As a result, the weighting coefficient for the number of parallel units increases. The fewer the number of parallel units, the greater the evaluation value and remuneration (initial remuneration). Therefore, the fourth schedule SK4 uses fewer substrate processing units MC to process substrates W than the first schedule SK1. By reducing the number of substrate processing units MC used, the number of substrate processing units MC performing pre-processing and post-processing based on pre-process conditions and post-process conditions is reduced, thereby reducing electricity and pure water usage. Furthermore, it is feasible that the larger the weighting coefficient for the number of parallel units, the smaller the weighting coefficient for production capacity by the arithmetic processing unit 104.

[0163] The evaluation items when creating the fourth schedule creation program may include the length of a period during which the substrate processing unit MC does not process a substrate W. Hereinafter, the length of a period during which the substrate processing unit MC does not process a substrate W may be referred to as a "suspended period of the substrate processing unit MC."

[0164] When the idle period of the substrate processing unit MC is shorter than a certain time, the operator is awarded additional remuneration. As a result, the weighting factor for the idle period of the substrate processing unit MC increases, and the evaluation value and remuneration (initial remuneration) increase when the idle period of the substrate processing unit MC is shorter than the certain time. Therefore, in the fourth schedule SK4, the idle period of the substrate processing unit MC is more likely to be shorter than the certain time compared to the first schedule SK1. By shortening the idle period of the substrate processing unit MC to a certain time, the use of source power can be reduced.

[0165] Specifically, when the idle period of a substrate processing unit MC exceeds a certain time, the control device 300 processes dummy substrates for that substrate processing unit MC. This results in increased power usage. Therefore, by shortening the idle period of a substrate processing unit MC to less than the certain time, the number of dummy substrates processed can be reduced, thereby suppressing power usage. Furthermore, it is feasible that the greater the weighting factor for the idle period of a substrate processing unit MC, the smaller the weighting factor for production capacity applied by the arithmetic processing unit 104.

[0166] The evaluation items when creating the fourth schedule creation program may include the number of cleaning processes in the chamber 2a. Hereinafter, the cleaning process in the chamber 2a may be referred to as "chamber cleaning." Chamber cleaning is an example of a "specific process."

[0167] Specifically, the substrate processing unit MC repeats chamber cleaning every first predetermined time (a certain period of time). Specifically, after the first predetermined time has passed since the last chamber clean, the substrate processing unit MC performs a subsequent chamber clean while not processing the substrate W. Therefore, even after the first predetermined time has passed since the last chamber clean, the substrate processing unit MC does not perform a subsequent chamber clean while continuing to process the substrate W. Therefore, the number of chamber cleans can be determined by whether the substrate processing unit MC continues to process the substrate W after the first predetermined time has passed since the last chamber clean. Hereinafter, a substrate processing unit MC after the first predetermined time has passed since the last chamber clean may be referred to as a "substrate processing unit MC to be cleaned."

[0168] The operator is given additional remuneration when the substrate processing unit MC, which is the target of cleaning, continues processing the substrate W. As a result, the weighting coefficient for the number of chamber cleans increases. The fewer chamber cleans, the greater the evaluation value and remuneration (initial remuneration). Therefore, in the fourth schedule SK4, the number of chamber cleans is reduced compared to the first schedule SK1. Since power usage increases during chamber cleans, reducing the number of chamber cleans can reduce power usage. Furthermore, it is feasible that the greater the weighting coefficient for the number of chamber cleans, the smaller the weighting coefficient for production capacity by the processing unit 104.

[0169] Next, the evaluation items used when creating the fourth schedule creation program will be further described. Additional evaluation items can be added during the creation of the fourth schedule creation program. Specifically, additional evaluation items can be added by including information indicating the connection between the substrate processing unit MC and the chemical tank CC, information about the tower TW to which the substrate processing unit MC belongs, or location information of the substrate processing unit MC in the input data during learning.

[0170] Specifically, when the input data during learning includes the connection relationship between the substrate processing unit MC and the chemical cabinet CC, the number of chemical liquid replacements can be added to the evaluation items. In detail, the control device 300 performs the chemical liquid replacement process every time the second specified time (a certain time) has passed. The chemical liquid replacement process means the process of replacing the chemical liquid in the chemical liquid cabinet CC with a new liquid. Hereinafter, the chemical liquid cabinet CC that has passed the second specified time since the last chemical liquid replacement process is sometimes recorded as the "chemical liquid cabinet CC to be replaced with chemical liquid". In addition, the substrate processing unit MC connected to the chemical liquid cabinet CC to be replaced with chemical liquid is sometimes recorded as the "substrate processing unit MC to be replaced with chemical liquid".

[0171] The control device 300 replaces the chemical solution in the chemical cabinet CC of the target chemical cabinet CC with fresh solution while the substrate processing unit MC connected to the target chemical cabinet CC is not processing substrates W. Therefore, even after the second predetermined time has passed since the last chemical solution replacement, the subsequent chemical solution replacement process will not be performed while the target substrate processing unit MC is still processing substrates W. Therefore, the number of chemical solution replacements can be determined by whether the target substrate processing unit MC is still processing substrates W.

[0172] When a substrate processing unit MC, targeted for liquid chemical replacement, continues processing a substrate W, the operator assigns additional compensation. As a result, the weighting coefficient for the number of liquid chemical replacements increases. The fewer liquid chemical replacements, the greater the evaluation value and compensation (initial compensation). Therefore, schedule 4 SK4 reduces the number of liquid chemical replacements compared to schedule 1 SK1. Because performing liquid chemical replacements increases power usage, reducing the number of liquid chemical replacements can help suppress power usage. Furthermore, it is feasible that the greater the weighting coefficient for the number of liquid chemical replacements, the smaller the weighting coefficient for production capacity applied by the processing unit 104.

[0173] When the input data during learning includes the connection relationship between the substrate processing unit MC and the chemical tank CC, whether only a specific chemical tank CC is used can be added as an evaluation item. Specifically, whether only one of the first chemical tank CC1 and the second chemical tank CC2 is used can be added as an evaluation item.

[0174] Whether only a specific chemical tank CC is used can be evaluated by determining whether the substrate W is processed only with the chemical stored in the specific chemical tank CC. In other words, the evaluation can be performed by determining whether the substrate W is processed only by the substrate processing unit MC connected to the specific chemical tank CC.

[0175] Specifically, whether only a specific chemical tank CC is used can be evaluated by determining whether the substrate W is processed using only the chemical stored in one of the first chemical tank CC1 and the second chemical tank CC2. In other words, the evaluation can be performed by determining whether the substrate W is processed only by the substrate processing unit MC connected to one of the first chemical tank CC1 and the second chemical tank CC2.

[0176] When an operator processes a substrate W using only the chemical solution stored in one of the first chemical tank CC1 and the second chemical tank CC2, an additional remuneration is assigned. In other words, when an operator processes a substrate W using only the substrate processing unit MC connected to either the first chemical tank CC1 or the second chemical tank CC2, an additional remuneration is assigned. As a result, the evaluation value and remuneration (initial remuneration) increase when only a specific chemical tank CC is used. By using only one of the first chemical tank CC1 and the second chemical tank CC2, power usage can be reduced. Therefore, the fourth schedule SK4 can reduce power usage compared to the first schedule SK1. Furthermore, it is feasible that the greater the weighting coefficient for the use of a specific chemical tank CC, the smaller the weighting coefficient for production capacity assigned by the processing unit 104.

[0177] When the input data during learning includes information about the tower TW to which the substrate processing unit MC belongs, whether only the specific tower TW is used can be added as an evaluation item. Specifically, whether substrates W are processed only by the substrate processing unit MC included in a specific tower TW among the four towers TW (1st tower TW1 to 4th tower TW4) can be added as an evaluation item.

[0178] When an operator processes a substrate W only in the substrate processing unit MC included in a specific tower TW, an additional remuneration is granted. As a result, when only the specific tower TW is used, the evaluation value and remuneration (initial remuneration) become larger. By using only a specific tower TW among the four towers TW, the use of power can be reduced. Therefore, the fourth schedule SK4 can reduce the use of power compared to the first schedule SK1. Furthermore, it is feasible that the greater the weighting coefficient for the use of a specific tower TW, the smaller the weighting coefficient for production capacity by the calculation processing unit 104.

[0179] When the position information of the substrate processing unit MC is included in the input data during learning, whether the substrate W is processed by the substrate processing unit MC located lower in the vertical direction may be added as an evaluation item.

[0180] When an operator processes a substrate W at a lower substrate processing unit MC, the additional remuneration value increases. Consequently, the evaluation value and remuneration (initial remuneration) increase as the lower substrate processing unit MC is used. By using lower substrate processing units MC, the amount of power supplied to the temperature control unit 4 or pump 5 can be reduced. Therefore, the fourth schedule SK4 can reduce the amount of source power used compared to the first schedule SK1. Furthermore, it is feasible that the greater the weighting coefficient for the location of the substrate processing unit MC used, the smaller the weighting coefficient for production capacity applied by the arithmetic processing unit 104.

[0181] Next, refer to Figures 6 to 13 , an example of the first schedule SK1 to the fourth schedule SK4 is described. Figures 6 to 13 In the figure, "W" indicates the movement of simultaneously transporting two substrates W. "P" indicates the movement of the central robot CR carrying a substrate W into the substrate processing unit MC. "X" indicates the movement of exchanging a substrate W held by the central robot CR with a substrate W in the substrate processing unit MC. "WX" indicates the movement of exchanging two unprocessed substrates W with two processed substrates W at the transfer unit PASS. "G" indicates the movement of the central robot CR carrying a substrate W out of the substrate processing unit MC.

[0182] Figure 6 This is a diagram showing a first example of the first schedule SK1. Figure 7 This is a diagram showing an example of the second schedule SK2. Figure 8 (a) and Figure 8 (b) is a diagram showing another example of the second schedule SK2. Figure 6 、 Figure 7 、 Figure 8 (a) and Figure 8 In (b), the substrate processing section PR in which the substrate processing section MC supplies chemical solution to the substrate W and processes the substrate W is indicated by “shading”. Figure 6 、 Figure 7 、 Figure 8 (a) and Figure 8 In (b), the transport path of the substrate W at the end of this project PJ is omitted.

[0183] exist Figure 6 、 Figure 7 、 Figure 8 (a) and Figure 8 In the example shown in (b), four substrate processing units MC (first substrate processing unit MC1 to fourth substrate processing unit MC4) are used to process the substrate W. Figure 6 、 Figure 7 、 Figure 8 (a) and Figure 8In the case of the first schedule SK1 and the second schedule SK2 shown in (b), the input data during learning does not include: information indicating the connection relationship between the substrate processing unit MC and the chemical solution cabinet CC, information indicating the tower TW to which the substrate processing unit MC belongs, and position information of the substrate processing unit MC.

[0184] like Figure 6 As shown, since the production capacity is prioritized in the first schedule SK1, the substrate processing sections PR of the first substrate processing section MC1 to the fourth substrate processing section MC4 are likely to overlap. Figure 6 In the example shown, three substrate processing sections PR overlap, and the peak value of the chemical liquid usage becomes large (time t10 to time t12 and time t20 to time t22).

[0185] On the other hand, Figure 7 As shown, since the second schedule SK2 prioritizes the dispersion of peak values ​​of the source power usage, the substrate processing sections PR of the first to fourth substrate processing sections MC1 to MC4 are less likely to overlap compared to the first schedule SK1.

[0186] Specifically, the timing of starting the processing of the substrate W in the second schedule SK2 is dispersed compared to the first schedule SK1. Figure 6 As shown, in the first schedule SK1, the substrate processing period PR starts at time t7, time t8, time t10, and time t11. Figure 7 As shown, in the second schedule SK2, the substrate processing period PR starts at time t7, time t8, time t11, and time t12.

[0187] As a result, the peak amount of the chemical liquid usage in the second schedule SK2 is smaller than that in the first schedule SK1. Figure 7 In the example shown, only two substrate processing sections PR overlap at most (time t8 to time t15 and time t18 to time t25), and the peak value of the chemical liquid usage is smaller than that of the first schedule SK1 in which three substrate processing sections PR overlap.

[0188] Furthermore, if Figure 6 and Figure 7 As shown, since the second schedule SK2 does not prioritize production capacity, the end time of this project PJ is later than that of the first schedule SK1.

[0189] Next, refer to Figure 8 (a) and Figure 8 (b) shows another example of the second schedule SK2. Figure 8 (a) and Figure 8In (b), the period during which the substrate W is on standby in the substrate processing unit MC (standby period WT) is indicated by “dot hatching”. Figure 8 The second schedule SK2 of (b) indicates Figure 8 (a) Continuation of the second schedule SK2.

[0190] exist Figure 8 In the example shown, with Figure 7 Compared to the second schedule SK2 shown in FIG. 1 , the timings of starting the processing of the substrates W (the starting timings of the substrate processing sections PR) are more dispersed. Specifically, Figure 8 In the example shown, no overlap of substrate processing zones PR occurs. Figure 8 In the second schedule SK2 shown, Figure 7 Compared with the second schedule SK2 shown, the peak value of the chemical liquid usage is further reduced.

[0191] However, in Figure 8 In the second schedule SK2 shown, a waiting period WT occurs. Furthermore, a vacant period VT occurs when no substrates W are loaded into the substrate processing unit MC. This results in a significant reduction in production capacity and a significant delay in the end time of the current project PJ. In this case, the operator adjusts the value of the additional remuneration to reduce the weighting factor for the peak chemical usage. This results in a second schedule creation procedure that is less likely to generate waiting periods WT or vacant periods VT.

[0192] Next, refer to Figure 9 、 Figure 10 (a) and Figure 10 (b) of the present invention will explain the third schedule SK3. Figure 9 This is a diagram showing a second example of the first schedule SK1. Figure 10 (a) and Figure 10 (b) is a diagram showing the first example of the third schedule SK3. Figure 9 、 Figure 10 (a) and Figure 10 In (b), "S" represents the movement of transporting one substrate W. Figure 9 、 Figure 10 (a) and Figure 10 In (b), the standby period WT is indicated by “dot hatching”. Figure 10 The third schedule SK3 of (b) indicates Figure 10 (a) Continuation of the 3rd schedule SK3.

[0193] exist Figure 9 、 Figure 10 (a) and Figure 10In the example shown in (b), five substrate processing units MC (first substrate processing unit MC1 to fifth substrate processing unit MC5) are used to process the substrate W. Figure 9 、 Figure 10 (a) and Figure 10 In the case of the first schedule SK1 and the third schedule SK3 shown in (b), the input data during learning does not include: information indicating the connection relationship between the substrate processing unit MC and the chemical solution cabinet CC, information indicating the tower TW to which the substrate processing unit MC belongs, and position information of the substrate processing unit MC.

[0194] like Figure 9 As shown, when the production capacity is prioritized, a period (for example, time t7 to time t10) during which the load port LP and the central robot CR transport one substrate W occurs. Figure 10 (a) and Figure 10 As shown in (b), in the third schedule SK3, the transport frequency of the transport unit is prioritized, so the load port LP and the central robot CR always transport two substrates W at the same time.

[0195] like Figure 9 、 Figure 10 (a) and Figure 10 As shown in (b), by giving priority to the simultaneous transport of two substrates W, the number of operations of the index robot IR and the center robot CR is reduced. Therefore, the third schedule SK3 can reduce the amount of power used compared to the first schedule SK1.

[0196] However, in the third schedule SK3, productivity is not prioritized. The more times two substrates W are simultaneously transported, the greater the evaluation value and reward (initial reward). Therefore, compared to the first schedule SK1, a waiting period WT is more likely to occur. For example, in the fourth substrate processing unit MC4, the waiting period WT lasts from time t18 to time t24.

[0197] Next, refer to Figure 11 (a) and Figure 11 (b) of the present invention will explain the third schedule SK3. Figure 11 (a) is a diagram showing the third example of the first schedule SK1. Figure 11 (b) is a diagram showing the second example of the third schedule SK3. Figure 11 (a) and Figure 11 In the case of the first schedule SK1 and the third schedule SK3 shown in (b), the input data during learning does not include: information indicating the connection relationship between the substrate processing unit MC and the chemical solution cabinet CC, information indicating the tower TW to which the substrate processing unit MC belongs, and position information of the substrate processing unit MC.

[0198] exist Figure 11 (a) and Figure 11 In the example shown in (b), the process conditions include: a first substrate processing process condition R1, a second substrate processing process condition R2, a first chemical process condition KC1, and a second chemical process condition KC2. Consequently, based on the process conditions, substrate processing is performed under the first substrate processing process condition R1 and the first chemical process condition KC1, and substrate processing is performed under the second substrate processing process condition R2 and the second chemical process condition KC2.

[0199] Specifically, in the substrate processing under the first substrate processing process condition R1, the first chemical liquid is used. Before the substrate processing under the first substrate processing process condition R1 is performed, the first chemical liquid is stored in the chemical liquid cabinet CC based on the first chemical liquid process condition KC1. In the substrate processing under the second substrate processing process condition R2, the second chemical liquid is used. Before the substrate processing under the second substrate processing process condition R2 is performed, the second chemical liquid is stored in the chemical liquid cabinet CC based on the second chemical liquid process condition KC2. The second chemical liquid may be of a different type from the first chemical liquid, for example. Alternatively, the temperature or concentration of the second chemical liquid may be different from that of the first chemical liquid. Furthermore, in the following description, the substrate processing under the first substrate processing process condition R1 is sometimes described as "first substrate processing". Also, the substrate processing under the second substrate processing process condition R2 is sometimes described as "second substrate processing".

[0200] like Figure 11 (a) and Figure 11 As shown in (b), the first substrate processing unit MC1 and the second substrate processing unit MC2 perform the first substrate processing, and the second substrate processing unit MC3 and the fourth substrate processing unit MC4 perform the second substrate processing.

[0201] like Figure 11 As shown in (a), in the first schedule SK1, substrate processing is performed in the order of first substrate processing, second substrate processing, first substrate processing, and second substrate processing. Therefore, two switches occur from the first substrate processing to the second substrate processing, each time the chemical process conditions KC are switched from the first chemical process conditions KC1 to the second chemical process conditions KC2. Furthermore, one switch occurs from the second substrate processing to the first substrate processing. In this case, the chemical process conditions KC are switched from the second chemical process conditions KC2 to the first chemical process conditions KC1.

[0202] When the chemical process condition KC is switched from the first chemical process condition KC1 to the second chemical process condition KC2, the chemical in the chemical tank CC is switched from the first chemical to the second chemical. Also, when the chemical process condition KC is switched from the second chemical to the first chemical, the chemical in the chemical tank CC is switched from the second chemical to the first chemical. Figure 11In the first schedule SK1 shown in (a), the number of times the chemical solution is switched is 3.

[0203] On the other hand, in the third schedule SK3, the fewer times the chemical solution is switched, the greater the evaluation value and reward (initial reward). Figure 11 As shown in (b), the number of times the chemical solution is switched is reduced compared to the first schedule SK1. Specifically, the order of substrate processing is changed as a result of giving priority to reducing the number of chemical solution switches. For example, Figure 11 In the third schedule SK3 shown in (b), substrate processing is performed in the order of first substrate processing, first substrate processing, second substrate processing, and second substrate processing. Therefore, switching from the first substrate processing to the second substrate processing only requires one operation, reducing the number of chemical solution switching operations compared to the first schedule SK1. Thus, the third schedule SK3 can reduce source power usage compared to the first schedule SK1.

[0204] Furthermore, in this embodiment, the substrate processing apparatus 200 includes two chemical tanks CC (a first chemical tank CC1 and a second chemical tank CC2). Therefore, the learning input data entered when creating the third schedule creation program may include information indicating the connection relationship between the substrate processing unit MC and the chemical tanks CC. In this case, the third schedule creation program creates the following schedule (third schedule SK3): the first chemical liquid is stored in the first chemical tank CC1, the second chemical liquid is stored in the second chemical tank CC2, the substrate processing unit MC supplied with chemical liquid (the first chemical liquid) from the first chemical tank CC1 performs the first substrate process, and the substrate processing unit MC supplied with chemical liquid (the second chemical liquid) from the second chemical tank CC2 performs the second substrate process. As a result, the number of chemical liquid switching operations is reduced.

[0205] Next, refer to Figure 12 (a) and Figure 12 (b) of the present invention will explain the fourth schedule SK4. Figure 12 (a) is a diagram showing the fourth example of the first schedule SK1. Figure 12 (b) is a diagram showing the first example of the fourth schedule SK4. Figure 12 (a) and Figure 12 In the case of the first schedule SK1 and the fourth schedule SK4 shown in (b), the input data during learning does not include: information indicating the connection relationship between the substrate processing unit MC and the chemical solution cabinet CC, information indicating the tower TW to which the substrate processing unit MC belongs, and position information of the substrate processing unit MC.

[0206] exist Figure 12 (a) and Figure 12 In the example shown in (b), pre-processing and post-processing are performed based on pre-process conditions and post-process conditions. Figure 12As shown in (a), in the first schedule SK1, the first to fourth substrate processing units MC1 to MC4 perform pre-processing and post-processing. In contrast, in the fourth schedule SK4, the number of parallel units is reduced because the evaluation value and reward (initial reward) increase as the number of parallel units decreases.

[0207] For example, in Figure 12 In the fourth schedule SK4 shown in (b), only the first and second substrate processing units MC1 to MC4 perform substrate processing. Therefore, only the first and second substrate processing units MC1 to MC2 perform pre-processing and post-processing. As a result, compared to the first schedule SK1, the amount of source power used is reduced.

[0208] Next, refer to Figure 13 (a) and Figure 13 (b) of the present invention will explain the fourth schedule SK4. Figure 13 (a) is a diagram showing the fifth example of the first schedule SK1. Figure 13 (b) is a diagram showing the second example of the fourth schedule SK4. Figure 13 In the case of the fourth schedule SK4 shown in (b), the input data during learning includes information indicating the tower TW to which the substrate processing unit MC belongs and position information of the substrate processing unit MC.

[0209] exist Figure 13 (a) and Figure 13 In the example shown in (b), the 1st substrate processing part MC1 to the 4th substrate processing part MC4 belong to the same tower TW, and are arranged from bottom to top in the order of the 1st substrate processing part MC1, the 2nd substrate processing part MC2, the 3rd substrate processing part MC3, and the 4th substrate processing part MC4.

[0210] like Figure 13 As shown in (a), in the first schedule SK1, the first substrate processing unit MC1 to the fourth substrate processing unit MC4 perform substrate processing. In contrast, in the fourth schedule SK4, the lower the substrate processing unit MC is used, the greater the evaluation value and reward (initial reward). Therefore, in Figure 13 In the fourth schedule SK4 shown in (b), only the first substrate processing unit MC1 and the second substrate processing unit MC2 among the first to fourth substrate processing units MC1 to MC4 perform substrate processing. As a result, the amount of source power used is reduced compared to the first schedule SK1.

[0211] Next, refer to Figure 2 and Figure 3The substrate processing system 1000 of this embodiment will be further described. The control device 300 controls the operation of the various components of the substrate processing apparatus 200. For example, the control device 300 controls the operation of the load port LP, the index robot IR, the central robot CR, the plurality of substrate processing units MC, and the chemical tank CC. Furthermore, the control device 300 controls the operation of the substrate storage container C via the load port LP.

[0212] Specifically, the control apparatus 300 stores a schedule creation program created by the learning apparatus 100. The control apparatus 300 creates a schedule SK based on the schedule creation program, and controls the operation of each component of the substrate processing apparatus 200 based on the created schedule SK.

[0213] More specifically, the control device 300 stores a plurality of schedule creation programs created by the learning device 100. The control device 300 can create a plurality of schedules SK with different amounts of source power usage based on the plurality of schedule creation programs. Figures 1 to 13 Regarding the first to fourth schedule creation programs described above, for example, the control device 300 creates a first schedule SK1 based on the first schedule creation program.

[0214] Then, based on Figure 2 、 Figure 3 and Figure 14 , the control device 300 included in the substrate processing system 1000 of this embodiment will be described. Figure 14 This is a block diagram showing the configuration of the control device 300 included in the substrate processing system 1000 according to this embodiment.

[0215] like Figure 14 As shown, the control device 300 includes an input unit 301 , a display unit 302 , a storage unit 303 , and a control unit 304 .

[0216] The input unit 301 is a user interface device operated by the user. The input unit 301 inputs instructions (control signals) corresponding to the user's operation to the control unit 304. The input unit 301 also inputs data corresponding to the user's operation to the control unit 304. The input unit 301 may include a keyboard and a mouse. The input unit 301 may also include a touch sensor superimposed on the display surface of the display unit 302. By superimposing the touch sensor on the display surface of the display unit 302, a graphical user interface can be formed. The input unit 301 may further include an interface for accessing removable media.

[0217] For example, the user operates the input unit 301 to select an operation mode of the substrate processing apparatus 200. The user also operates the input unit 301 to input source power information RS. The source power information RS may be input to the control unit 304 from a removable medium.

[0218] The display unit 302 displays various screens or images. For example, the display unit 302 displays a screen for selecting an operating mode of the substrate processing apparatus 200. Furthermore, the display unit 302 can display the schedule SK created by the schedule creation program. The display unit 302 includes, for example, a liquid crystal display device or an organic EL display device.

[0219] The storage unit 303 includes a primary storage device. The primary storage device includes, for example, a semiconductor memory. The storage unit 303 further includes an auxiliary storage device. The auxiliary storage device includes, for example, at least one of a semiconductor memory and a hard disk drive. The storage unit 303 may include removable media.

[0220] The storage unit 303 stores various computer programs and data. The various computer programs include the first to fourth schedule creation programs. The various data include multiple pieces of process condition information. Specifically, the storage unit 303 associates the multiple pieces of process condition information with process condition identification information and stores them. Furthermore, the various data include source power information RS. Specifically, the source power information RS input via the input unit 301 is stored in the storage unit 103.

[0221] The control unit 304 includes a processor. For example, the control unit 304 includes a CPU or an MPU. Alternatively, the control unit 304 may include a general-purpose computer, a special-purpose computing unit, a GPU, an NPU, or a quantum computer. For example, a special-purpose computing unit includes an ASIC.

[0222] The control unit 304 executes various processes based on the various programs and data stored in the storage unit 303. For example, the control unit 304 creates a first schedule SK1 based on a first schedule creation program. Specifically, when a substrate storage container C is docked at the load port LP, the control unit 304 receives information indicating the current state of the substrate processing apparatus 200. Furthermore, wafer count information and process condition identification information are input from the substrate storage container C via the substrate processing apparatus 200 to the control unit 304. The control unit 304 identifies process condition information associated with the process condition identification information read from the substrate storage container C. The control unit 304 then creates a schedule SK based on the current state of the apparatus, the identified process condition information, the source power information RS, and the schedule creation program.

[0223] Specifically, the control unit 304 inputs scheduling input data, including information indicating the current state of the equipment, specific process condition information, and source power information RS, into a learned model included in the schedule creation program. As a result, the learned model outputs a schedule SK. The learned model creates the schedule SK by arranging a plurality of blocks BL defining the contents of the operations of the substrate processing apparatus 200 in a time series.

[0224] Furthermore, the input data for schedule creation may further include information indicating the connection between the substrate processing unit MC and the chemical tank CC, information indicating the tower TW to which the substrate processing unit MC belongs, and location information of the substrate processing unit MC. This information may be pre-stored in the storage unit 303 or input by the user via the input unit 301.

[0225] Next, refer to Figure 15 , the operation mode of the substrate processing apparatus 200 is described. Figure 15 1 is a diagram showing mode information MP indicating the relationship between the operation mode of the substrate processing apparatus 200 included in the substrate processing system 1000 of this embodiment and the scheduler. The mode information MP is stored in the storage unit 303 .

[0226] like Figure 15 As shown, in this embodiment, the operation modes of the substrate processing apparatus 200 include a standard mode, a peak distribution mode, a conservation mode (weak), and a conservation mode (strong). The standard mode is associated with the first schedule creation procedure. The peak distribution mode is associated with the second schedule creation procedure. The conservation mode (weak) is associated with the third schedule creation procedure. The conservation mode (strong) is associated with the fourth schedule creation procedure.

[0227] like Figure 15 As shown, the first schedule creation process includes a first learned model. The first learned model is a prediction model constructed when the first schedule creation process is created. Similarly, the second through fourth schedule creation processes include second through fourth learned models, respectively. The second through fourth learned models are prediction models constructed when the second through fourth schedule creation processes are created, respectively.

[0228] Next, refer to Figures 14 to 16 , the mode selection screen G1 is explained. Figure 16 FIG. 1 is a diagram showing a mode selection screen G1. Figure 16 As shown, the display unit 302 displays the mode selection screen G1. For example, when a substrate storage container C is docked at the load port LP, the control unit 304 displays the mode selection screen G1 on the display unit 302. The mode selection screen G1 is a screen for the user to select an operation mode of the substrate processing apparatus 200.

[0229] In this embodiment, the mode selection screen G1 displays first to fourth selection buttons B1, B4, and an OK button B5. The first selection button B1 is a soft button for selecting the standard mode. The second selection button B2 is a soft button for selecting the peak dispersion mode. The third selection button B3 is a soft button for selecting the low-power mode. The fourth selection button B4 is a soft button for selecting the high-power mode. The OK button B5 is a soft button for confirming the selected operation mode.

[0230] The user operates the input unit 301, presses any one of the first selection button B1 to the fourth selection button B4, and then presses the confirmation button B5, thereby setting the operation mode of the substrate processing device 200 to any one of the standard mode, peak distribution mode, saving mode (weak) and saving mode (strong).

[0231] Specifically, when the user presses the confirm button B5 after pressing any of the first selection button B1 to the fourth selection button B4 via the input unit 301, the control unit 304 references the mode information MP and creates a schedule SK based on the schedule creation program associated with the selected operational mode. For example, if the user presses the first selection button B1 via the input unit 301 and then presses the confirm button B5, the control unit 304 creates a first schedule SK1 based on the first schedule creation program.

[0232] Next, refer to Figure 17 , the processing flow when establishing schedule SK is explained. Figure 17 This is a sequence diagram showing the flow of processing when creating the schedule SK.

[0233] When creating the schedule SK, the control unit 304 creates the schedule SK based on the schedule creation input data including the current device status, the content of the started process (the content of the current project PJ), and the source power information RS, and the schedule creation program selected through the mode selection screen G1. Figure 17 As shown, when the schedule SK is created, the processes of steps S21 to S29 are executed.

[0234] Due to the processing of steps S21 to S29 and the reference Figure 5 Since the processes of steps S1 to S9 are substantially the same, detailed descriptions are omitted here. However, in the process of step S26, the move with the highest evaluation value among the moves that can be executed next is selected as the move to be executed next.

[0235] Figure 17 The neural network model shown is a learned model (prediction model). The weighted coefficients of the neural network model are as shown in Figures 1 to 13 As explained above, the scheduling model is adjusted based on its intended purpose. For example, the weighting coefficients of the neural network model included in the first scheduling program are adjusted to prioritize production capacity. Therefore, when the user selects the standard mode, a production capacity-prioritized schedule SK (the first schedule SK1) is created.

[0236] Above, refer to Figures 1 to 17 Embodiment 1 of the present invention has been described. According to this embodiment, the amount of source power used can be reduced simply by selecting the operating mode of the substrate processing apparatus 200. Therefore, compared to adjusting various setting values ​​(various parameter values) of the substrate processing apparatus 200 to reduce source power usage, the operator's time and effort can be saved.

[0237] Furthermore, according to this embodiment, the amount of source power used can be selected in stages. Therefore, the amount of source power used can be changed in stages according to external factors, thereby improving user convenience.

[0238] [Implementation Method 2]

[0239] Next, refer to Figures 1 to 19 , Embodiment 2 of the present invention will be described. However, matters different from Embodiment 1 will be described, and descriptions of matters similar to Embodiment 1 will be omitted. Embodiment 2 differs from Embodiment 1 in the processing executed by the control unit 304 after the schedule SK is created.

[0240] Figure 18 FIG. 1 is a diagram showing a confirmation screen G2 displayed on the display unit 302 included in the substrate processing system 1000 of this embodiment. After the control unit 304 creates the schedule SK, the confirmation screen G2 is displayed on the display unit 302. Figure 18 As shown, the confirmation screen G2 displays the end time D1 of the schedule SK, an OK button B11 , and a return button B12 .

[0241] The end time D1 of the schedule SK indicates the end time of the current project PJ. The confirmation button B11 is a soft button for confirming the schedule SK. The return button B12 is a soft button for returning to the mode selection screen G1.

[0242] Figure 19 This is a flowchart showing the flow of processing executed by the control unit 304 included in the substrate processing system 1000 of this embodiment. Figure 19 The illustrated process is started by selecting an operation mode of the substrate processing apparatus 200 on the mode selection screen G1.

[0243] like Figure 19As shown, when the operation mode of the substrate processing apparatus 200 is selected in the mode selection screen G1 , the control unit 304 creates a schedule SK (step S31 ).

[0244] After creating the schedule SK, the control unit 304 displays a confirmation screen G2 on the display unit 302 (step S32 ).

[0245] After displaying the confirmation screen G2 on the display unit 302, the control unit 304 determines whether the soft button pressed by the user via the input unit 301 is the OK button B11 or the return button B12 (step S33). For example, if the user views the confirmation screen G2 and confirms the end time D1 of the schedule SK, and if the end time D1 is within the acceptable range, the user can operate the input unit 301 and press the OK button B11. Alternatively, if the end time D1 is outside the acceptable range, the user can operate the input unit 301 and press the return button B12.

[0246] When the user presses the OK button B11 (A in step S33), the control unit 304 causes the substrate processing apparatus 200 to start processing the substrate W (step S34). As a result, the process ends. Figure 19 At this time, the control unit 304 controls the operation of the substrate processing apparatus 200 based on the established schedule SK.

[0247] On the other hand, when the user presses the return button B12 (B in step S33), the control unit 304 causes the display unit 302 to display the mode selection screen G1 (step S35), and ends the process. Figure 19 The processing shown.

[0248] Above, refer to Figures 1 to 19 , Embodiment 2 of the present invention has been described. According to this embodiment, when the end time of the schedule SK is outside the permissible range, the user can reselect the operation mode of the substrate processing apparatus 200 to adjust the end time of the schedule SK to within the permissible range. This improves user convenience.

[0249] [Implementation Method 3]

[0250] Next, refer to Figures 1 to 17 、 Figure 20 and Figure 21 , Embodiment 3 of the present invention will be described. However, the matters that differ from Embodiments 1 and 2 will be described, and the description of the matters that are the same as Embodiments 1 and 2 will be omitted. Embodiment 3 differs from Embodiments 1 and 2 in the processing performed by the control unit 304 after the schedule SK is created.

[0251] Figure 20FIG. 3 is a diagram showing a confirmation screen G3 displayed on the display unit 302 included in the substrate processing system 1000 of this embodiment. After the schedule SK is created, the control unit 304 displays the confirmation screen G3 on the display unit 302. Figure 20 As shown, the confirmation screen G3 displays: the message M1, the end time D11 of the schedule SK, the confirm button B21, and the return button B22.

[0252] The confirmation screen G3 further includes the information M1. Figure 18 The confirmation screen G2 shown is different. Message M1 notifies that the end time D11 of schedule SK exceeds the permissible end time. In other words, message M1 notifies that the end time D11 of schedule SK is outside the permissible range. The user can confirm by viewing message M1 that the end time D11 of schedule SK is outside the permissible range. The permissible end time is pre-stored in storage unit 303. Specifically, the user operates input unit 301 to input the permissible end time.

[0253] Figure 21 This is a flowchart showing the flow of processing executed by the control unit 304 included in the substrate processing system 1000 of this embodiment. Figure 21 The illustrated process is started by selecting an operation mode of the substrate processing apparatus 200 on the mode selection screen G1.

[0254] like Figure 21 As shown, when the operation mode of the substrate processing apparatus 200 is selected in the mode selection screen G1 , the control unit 304 creates a schedule SK (step S41 ).

[0255] After creating the schedule SK, the control unit 304 determines whether the end time of the schedule SK exceeds the allowable end time (step S42).

[0256] When the control unit 304 determines that the end time of the schedule SK exceeds the permissible end time (Yes in step S42 ), it displays the confirmation screen G3 on the display unit 302 (step S43 ).

[0257] After displaying the confirmation screen G3 on the display unit 302, the control unit 304 determines whether the soft button pressed by the user through the input unit 301 is the confirm button B21 or the return button B22 (step S44). For example, if the user still wishes to execute the substrate processing even though the end time D11 is outside the allowable range, the user may press the confirm button B21 through the input unit 301. Alternatively, if the end time D11 is outside the allowable range, the user may press the return button B22 through the input unit 301.

[0258] When the user presses the OK button B21 (A in step S44), the control unit 304 causes the substrate processing apparatus 200 to start processing the substrate W (step S45). As a result, the process ends. Figure 21 At this time, the control unit 304 controls the operation of the substrate processing apparatus 200 based on the established schedule SK. Alternatively, if the control unit 304 determines that the end time of the schedule SK has not exceeded the allowable end time (No in step S42), the control unit 304 causes the substrate processing apparatus 200 to start processing the substrate W (step S45).

[0259] On the other hand, when the user presses the return button B22 (B in step S44), the control unit 304 causes the display unit 302 to display the mode selection screen G1 (step S46), and ends the process. Figure 21 The processing shown.

[0260] Above, refer to Figures 1 to 17 、 Figure 20 and Figure 21 , Embodiment 3 of the present invention has been described. According to this embodiment, when the end time of the schedule SK is outside the allowable range, the user can reselect the operation mode of the substrate processing apparatus 200 to adjust the end time of the schedule SK to within the allowable range. This improves user convenience.

[0261] [Implementation 4]

[0262] Next, refer to Figures 1 to 15 、 Figure 17 ,and Figure 22 , Embodiment 4 of the present invention will be described. However, only matters different from Embodiments 1 to 3 will be described, and descriptions of matters common to Embodiments 1 to 3 will be omitted. Embodiment 4 differs from Embodiments 1 to 3 in the processing performed by the control unit 304 when creating the schedule SK.

[0263] Figure 22 The diagram shows a mode selection screen G4 displayed on the display unit 302 included in the substrate processing system 1000 of this embodiment. For example, the control unit 304 displays the mode selection screen G4 on the display unit 302 when a substrate storage container C is docked at the load port LP.

[0264] Specifically, when a substrate storage container C is docked at the load port LP, the control unit 304 creates the first through fourth schedules SK1 through SK4 based on the first through fourth schedule creation procedures and obtains the end time of each of the first through fourth schedules SK1 through SK4. After obtaining the end time of each of the first through fourth schedules SK1 through SK4, the control unit 304 displays the mode selection screen G4 on the display unit 302.

[0265] like Figure 22 As shown, the mode selection screen G4 displays the first selection button B31 to the fourth selection button B34, the confirmation button B35, and the end time notification bar TS. The mode selection screen G4 is different from the mode selection screen G4 in that it further includes the end time notification bar TS. Figure 16 The mode selection screen G1 shown is different. The end time notification bar TS notifies the end time of the schedule SK when the substrate processing apparatus 200 operates in standard mode, the end time of the schedule SK when the substrate processing apparatus 200 operates in peak distribution mode, the end time of the schedule SK when the substrate processing apparatus 200 operates in saving mode (weak), and the end time of the schedule SK when the substrate processing apparatus 200 operates in saving mode (strong). In other words, the end time notification bar TS notifies the end time of each of the first to fourth schedules SK1 to SK4.

[0266] Above, refer to Figures 1 to 15 、 Figure 17 ,and Figure 22 , Embodiment 4 of the present invention has been described. According to this embodiment, since the end time of each operation mode is displayed on the mode selection screen G4, the user can select an operation mode after confirming the end time of each operation mode. Therefore, user convenience is improved.

[0267] [Implementation 5]

[0268] Next, refer to Figures 1 to 15 、 Figure 17 ,and Figure 23 , Embodiment 5 of the present invention will be described. However, only matters different from Embodiments 1 to 4 will be described, and descriptions of matters common to Embodiments 1 to 4 will be omitted. Embodiment 5 differs from Embodiments 1 to 4 in the processing performed by the control unit 304 when creating the schedule SK.

[0269] Figure 23 This is a flowchart showing the process flow of the control unit 304 included in the substrate processing system 1000 of this embodiment. For example, when the substrate storage container C is docked at the load port LP, the control unit 304 starts Figure 23 The processing shown.

[0270] The control unit 304 starts Figure 23 After the processing shown, the first to fourth schedules SK1 to SK4 are created based on the first to fourth schedule creation programs (step S51 ), and the end time of each of the first to fourth schedules SK1 to SK4 is acquired (step S52 ).

[0271] After obtaining the end time of each of the first to fourth schedules SK1 to SK4, the control unit 304 selects a schedule SK having an end time within the allowable range and the smallest total amount of source power usage from among the first to fourth schedules SK1 to SK4 (step S53).

[0272] After selecting the schedule SK, the control unit 304 causes the substrate processing apparatus 200 to start processing the substrate W (step S54), and ends the process. Figure 23 The control unit 304 controls the operation of the substrate processing apparatus 200 based on the schedule SK selected in step S53.

[0273] Above, refer to Figures 1 to 15 、 Figure 17 ,and Figure 23 , Embodiment 5 of the present invention has been described. According to this embodiment, from among the schedules SK (operation modes) whose end times are within the permissible range, the schedule SK (operation mode) with the lowest total source power usage is selected. Therefore, source power usage can be suppressed. Furthermore, according to this embodiment, the user does not need to select an operation mode for the substrate processing apparatus 200; the operation mode with the lowest total source power usage is selected. This improves user convenience.

[0274] Above, with reference to the accompanying drawings ( Figures 1 to 23 ) The embodiments of the present invention are described. However, the present invention is not limited to the above-mentioned embodiments, and can be implemented in various embodiments without departing from the gist thereof. In addition, the plurality of components disclosed in the above-mentioned embodiments can be appropriately changed. For example, a component among all the components shown in a certain embodiment can be added to the components of other embodiments, or some components among all the components shown in a certain embodiment can be deleted from the embodiment.

[0275] The accompanying drawings schematically illustrate the components to facilitate understanding of the present invention. However, the thickness, length, number, spacing, etc. of the components shown in the drawings may differ from the actual components due to the convenience of drawing the drawings. Furthermore, the configurations of the components shown in the above-described embodiments are merely examples and are not particularly limited. Various modifications may be made without departing from the scope of the present invention.

[0276] For example, the substrate processing performed by the substrate processing apparatus 200 (substrate processing unit MC) is not particularly limited, and may be, for example, etching, cleaning, brush cleaning, photosensitive film coating, developing, annealing, or drawing.

[0277] Again, in reference Figures 1 to 23While the substrate processing system 1000 in the illustrated embodiment stores four types of schedule creation programs (first through fourth schedule creation programs), the number of schedule creation programs is not limited to four as long as schedules SK with different amounts of source power usage can be created. The substrate processing system 1000 may store two, three, or five or more schedule creation programs with different amounts of source power usage.

[0278] Again, in reference Figures 1 to 23 While the substrate processing apparatus 200 is a single-wafer type in the illustrated embodiment, the substrate processing apparatus 200 may also be a batch type.

[0279] Again, in reference Figures 1 to 23 While the substrate processing apparatus 200 includes two chemical liquid tanks CC in the described embodiment, the substrate processing apparatus 200 may include one, or three or more chemical liquid tanks CC.

[0280] Industrial applicability

[0281] The present invention is useful for an apparatus for processing a substrate and has industrial applicability.

[0282] Description of Reference Numerals

[0283] 200: Substrate processing equipment

[0284] 300: Control device

[0285] 303: Storage

[0286] 304: Control Department

[0287] 1000: Substrate processing system

[0288] CC: Chemical Cabinet

[0289] CC1: No. 1 Chemical Cabinet

[0290] CC2: Second Chemical Cabinet

[0291] CR: Center Robot

[0292] D1: Ending time

[0293] D11: Ending Moment

[0294] IR: Indexing Robot

[0295] KC: Chemical process conditions

[0296] KC1: First liquid process conditions

[0297] KC2: Second liquid process conditions

[0298] M: Machine Learning Model

[0299] MC: Substrate Processing Department

[0300] MC1: 1st substrate processing unit

[0301] MC2: Second substrate processing unit

[0302] MC3: 3rd substrate processing unit

[0303] MC4: 4th substrate processing unit

[0304] MC5: 5th substrate processing unit

[0305] PJ: Project

[0306] PR: Substrate processing area

[0307] R: Substrate processing conditions

[0308] R1: 1st substrate processing condition

[0309] R2: Second substrate processing conditions

[0310] RS: Source Power Information

[0311] SK: Scheduling

[0312] SK1: Schedule 1

[0313] SK2: Schedule 2

[0314] SK3: Schedule 3

[0315] SK4: Schedule 4

[0316] TW: Tower

[0317] TW1: Tower 1

[0318] TW2: Tower 2

[0319] TW3: Tower 3

[0320] TW4: Tower 4

[0321] W: substrate

Claims

1. A substrate processing system, wherein: Include: A substrate processing apparatus that processes a substrate using a source power, and a control device that arranges a plurality of blocks defining contents of the operation of the substrate processing apparatus in a time series based on number information indicating the number of substrates to be processed by the substrate processing apparatus, process condition information defining steps of the operation of the substrate processing apparatus, and source power information defining the amount of source power used in each movement included in the steps of the operation of the substrate processing apparatus, and creates a schedule defining the operation of the substrate processing apparatus in a time series; The schedule includes a first schedule and a second schedule having different usage amounts of the source power. The control device establishes the first schedule based on schedule creation input data including the wafer count information, the process condition information, and the source power information, and a first learned model, and establishes the second schedule based on the schedule creation input data and a second learned model different from the first learned model; The first learned model is constructed by performing reinforcement learning to adjust weighting coefficients so that the end time of the first schedule is earlier based on learning input data including the number of pieces information, the process condition information, and the source power information. The second learned model is constructed by performing reinforcement learning based on the learning input data to adjust weighting coefficients so that the peak value of the source power usage in the second schedule is smaller or the total amount of the source power usage in the second schedule is smaller.

2. The substrate processing system according to claim 1, wherein: The second schedule ends later than the first schedule.

3. The substrate processing system according to claim 1 or 2, wherein: The second schedule has a smaller peak value of the source power usage than the first schedule.

4. The substrate processing system according to claim 1 or 2, wherein: The second schedule uses a smaller total amount of the source power than the first schedule.

5. The substrate processing system according to claim 1 or 2, wherein: The source power includes at least one of a processing fluid, electricity, and gas.

6. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a substrate processing portion for processing the substrate. The schedule creation input data and the learning input data further include information indicating a position of the substrate processing section.

7. The substrate processing system according to claim 1 or 2, wherein: The second schedule disperses timings at which the substrate processing apparatus starts processing the plurality of substrates compared to the first schedule.

8. The substrate processing system according to claim 1 or 2, wherein: The substrate processing device comprises: a substrate processing unit that processes a first substrate using a first chemical solution and processes a second substrate using a second chemical solution; and a chemical liquid cabinet that exclusively stores the first chemical liquid and the second chemical liquid supplied to the substrate processing section; and The second schedule is less frequent in switching between the first chemical solution and the second chemical solution in the chemical solution cabinet than the first schedule.

9. The substrate processing system according to claim 8, wherein: The second chemical solution is of a different type from the first chemical solution.

10. The substrate processing system according to claim 8, wherein: The second chemical solution has a different temperature or concentration from the first chemical solution.

11. The substrate processing system according to claim 1 or 2, wherein: The substrate processing device comprises: a first substrate processing unit and a second substrate processing unit for processing the substrate using a chemical solution; a first chemical liquid tank storing the chemical liquid supplied to the first substrate processing section; and a second chemical liquid tank storing the chemical liquid supplied to the second substrate processing section; The second schedule supplies the first chemical solution from the first chemical solution tank to the first substrate processing section, and supplies the second chemical solution from the second chemical solution tank to the second substrate processing section.

12. The substrate processing system according to claim 11, wherein: The second chemical solution is of a different type from the first chemical solution.

13. The substrate processing system according to claim 11, wherein: The second chemical solution has a different temperature or concentration from the first chemical solution.

14. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a plurality of substrate processing units for processing the substrate. The second schedule causes a smaller number of the substrate processing units to process the substrate compared to the first schedule.

15. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a substrate processing portion for processing the substrate. In the second schedule, compared with the first schedule, a period during which the substrate processing section does not process the substrate is shorter.

16. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a substrate processing portion for processing the substrate. The substrate processing unit performs a specific process during a period when the substrate is not processed after a certain period of time. The second schedule causes the substrate processing section to continue processing the substrate after the predetermined time has elapsed.

17. The substrate processing system according to claim 1 or 2, wherein: The substrate processing device comprises: a substrate processing unit that processes the substrate using a chemical solution; and a chemical liquid cabinet for storing the chemical liquid supplied to the substrate processing section; and The substrate processing apparatus performs a replacement process of the chemical solution stored in the chemical solution tank after a certain period of time. The second schedule causes the substrate processing unit to continue processing the substrate after the predetermined time has elapsed.

18. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a conveying unit for conveying the substrate. In the second schedule, the number of operations of the transport unit is smaller than that of the first schedule.

19. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a plurality of towers, The plurality of towers respectively include a plurality of substrate processing parts for processing the substrates. The second schedule causes the substrate processing section included in a specific tower among the plurality of towers to process the substrate.

20. The substrate processing system according to claim 1 or 2, wherein: The substrate processing apparatus includes a plurality of chemical liquid tanks for storing chemical liquids. The second schedule processes the substrate using the chemical solution stored in a specific chemical solution tank among the plurality of chemical solution tanks.

21. The substrate processing system according to claim 1 or 2, wherein: The schedule includes a plurality of schedules having different usage amounts of the source power. The control device obtains end times of each of the plurality of schedules, and selects a schedule whose end time is within an allowable range and whose total amount of source power usage is the smallest from the plurality of schedules.

22. A method for establishing a schedule, comprising: arranging a plurality of blocks defining the contents of an operation of a substrate processing apparatus along a time sequence, and establishing a schedule defining the operations of the substrate processing apparatus along the time sequence, wherein: The substrate processing apparatus processes the substrate using source power, The schedule includes a first schedule and a second schedule having different usage amounts of the source power. The schedule establishment method includes a schedule establishment process of establishing at least one of the first schedule and the second schedule. The first schedule is established based on schedule establishment input data including sheet number information indicating the number of substrates to be processed by the substrate processing apparatus, process condition information defining steps of the operation of the substrate processing apparatus, and source power information indicating the usage amount of the source power used in each movement included in the steps of the operation of the substrate processing apparatus, and a first learned model. The second schedule is established based on input data when the schedule is established and a second learned model different from the first learned model. The first learned model is constructed by performing reinforcement learning to adjust weighting coefficients so that the end time of the first schedule is earlier based on learning input data including the number of pieces information, the process condition information, and the source power information. The second learned model is constructed by performing reinforcement learning based on the learning input data to adjust weighting coefficients so that the peak value of the source power usage in the second schedule is smaller or the total amount of the source power usage in the second schedule is smaller.

23. The schedule creation method according to claim 22, wherein: The second schedule ends later than the first schedule.

24. The schedule creation method according to claim 22 or 23, wherein: The second schedule has a smaller peak value of the source power usage than the first schedule.

25. The schedule creation method according to claim 22 or 23, wherein: The second schedule uses a smaller total amount of the source power than the first schedule.

26. The schedule creation method according to claim 22 or 23, wherein: The source power includes at least one of a processing fluid, electricity, and gas.

27. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a substrate processing portion for processing the substrate. The schedule creation input data and the learning input data further include information indicating a position of the substrate processing section.

28. The schedule creation method according to claim 22 or 23, wherein: The second schedule disperses timings at which the substrate processing apparatus starts processing the plurality of substrates compared to the first schedule.

29. The schedule creation method according to claim 22 or 23, wherein: The substrate processing device comprises: a substrate processing unit that processes a first substrate using a first chemical solution and processes a second substrate using a second chemical solution; and a chemical liquid tank that exclusively stores the first chemical liquid and the second chemical liquid to be supplied to the substrate processing section; The second schedule is less frequent in switching between the first chemical solution and the second chemical solution in the chemical solution cabinet than the first schedule.

30. The schedule creation method according to claim 29, wherein: The second chemical solution is of a different type from the first chemical solution.

31. The schedule creation method according to claim 29, wherein: The second chemical solution has a different temperature or concentration from the first chemical solution.

32. The schedule creation method according to claim 22 or 23, wherein: The substrate processing device comprises: a first substrate processing unit and a second substrate processing unit for processing the substrate using a chemical solution; a first chemical liquid tank storing the chemical liquid supplied to the first substrate processing section, and a second chemical liquid tank storing the chemical liquid supplied to the second substrate processing section; The second schedule supplies the first chemical solution from the first chemical solution tank to the first substrate processing section, and supplies the second chemical solution from the second chemical solution tank to the second substrate processing section.

33. The schedule creation method according to claim 32, wherein: The second chemical solution is of a different type from the first chemical solution.

34. The schedule creation method according to claim 32, wherein: The second chemical solution has a different temperature or concentration from the first chemical solution.

35. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a plurality of substrate processing units for processing the substrate. The second schedule causes a smaller number of the substrate processing units to process the substrate compared to the first schedule.

36. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a substrate processing portion for processing the substrate. In the second schedule, compared with the first schedule, a period during which the substrate processing section does not process the substrate is shorter.

37. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a substrate processing portion for processing the substrate. The substrate processing unit performs a specific process during a period when the substrate is not processed after a certain period of time. The second schedule causes the substrate processing section to continue processing the substrate after the predetermined time has elapsed.

38. The schedule creation method according to claim 22 or 23, wherein: The substrate processing device comprises: a substrate processing unit that processes the substrate using a chemical solution; and a chemical liquid cabinet for storing the chemical liquid supplied to the substrate processing section; and The substrate processing apparatus performs a replacement process of the chemical solution stored in the chemical solution tank after a certain period of time. The second schedule causes the substrate processing unit to continue processing the substrate after the predetermined time has elapsed.

39. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a conveying unit for conveying the substrate. In the second schedule, the number of operations of the transport unit is smaller than that of the first schedule.

40. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a plurality of towers, The plurality of towers respectively include a plurality of substrate processing parts for processing the substrates. The second schedule causes the substrate processing section included in a specific tower among the plurality of towers to process the substrate.

41. The schedule creation method according to claim 22 or 23, wherein: The substrate processing apparatus includes a plurality of chemical liquid tanks for storing chemical liquids. The second schedule processes the substrate using the chemical solution stored in a specific chemical solution tank among the plurality of chemical solution tanks.

42. The schedule creation method according to claim 22 or 23, wherein: The schedule includes a plurality of schedules having different usage amounts of the source power. The schedule creation process includes the following steps: establishing a plurality of said schedules; Obtaining the end time of each of the plurality of schedules; A schedule is selected from the plurality of schedules, wherein the end time is within an allowable range and the total amount of the source power usage is the smallest.

43. A non-transitory computer-readable storage medium storing a schedule creation program for execution by a computer, wherein: The schedule creation program causes the computer to execute calculations according to the schedule creation method according to claim 22 or 23.

Images