Substrate processing apparatus management system, management device, substrate processing apparatus, substrate processing apparatus management method, and substrate processing apparatus management program

Abstract

To prevent lowering of an operation rate of a substrate processing device.SOLUTION: A substrate processing device 1 includes an exhaust path EX1 to which predetermined force is applied, and substrate processing units WU1 to WU16 which are configured to be able to share the exhaust path EX1, where a management device 4, which corresponds to each of the substrate processing units WU, previously determines a group correlation as a correlation among a plurality of pieces of processing information indicating the operation or the state relating to air supply / exhaust, and manages the substrate processing device 1, acquires the plurality of pieces of processing information corresponding to each of the substrate processing units WU1 to WU16, and detects the state before the plurality of substrate processing units WU1 to WU16 are abnormal, on the basis of comparison information obtained by comparing the correlation among the plurality of pieces of processing information relating to the exhaust among the plurality of processing information with the group correlation.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a substrate processing apparatus management system, a management apparatus, a substrate processing apparatus, a substrate processing apparatus management method, and a substrate processing apparatus management program, and more particularly to a substrate processing apparatus management system for managing a substrate processing apparatus in which a plurality of substrate processing units share an exhaust path, a management apparatus included in the substrate processing apparatus management system, a substrate processing apparatus provided with the management apparatus, a substrate processing apparatus management method executed by the management apparatus, and a substrate processing apparatus management program for causing a computer to execute the substrate processing apparatus management method.

Background Art

[0002] A substrate processing apparatus for processing a substrate such as a semiconductor substrate (semiconductor wafer) has a plurality of chambers in which the substrate is processed, and includes an air supply and exhaust system for keeping the atmosphere inside each of the plurality of chambers clean. Japanese Unexamined Patent Application Publication No. 2021-136435 describes a substrate processing apparatus that discharges air in a plurality of chambers through a single exhaust pipe. In this substrate processing apparatus, the exhaust pressure of the chamber is adjusted by the opening degree of a damper.

[0003] Generally, the inside of a chamber of a substrate processing apparatus is maintained at a constant pressure to prevent the scattering of particles. In each of the plurality of chambers, the air supply pressure and the exhaust pressure are controlled so that the appropriate pressure inside the chamber is kept constant.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Because multiple chambers are connected to a single exhaust pipe, even if the internal pressure of the chambers remains constant, the damper opening in one or more chambers may reach its upper limit. Furthermore, in special circumstances, such as a decrease in the force applied to the exhaust pipe, if the damper opening increases in a chamber where the damper opening was previously below the upper limit, the exhaust pressure may decrease in one or more chambers where the damper opening was previously at the upper limit. If the exhaust pressure of a chamber decreases, an abnormality indicating a decrease in pressure inside the chamber is detected, and the substrate processing device stops.

[0006] The objective of the present invention is to provide a substrate processing apparatus management system, a management device, a substrate processing apparatus, a substrate processing apparatus management method, and a substrate processing apparatus management program that prevent a decrease in the operating rate of the substrate processing apparatus. [Means for solving the problem]

[0007] According to one aspect of this invention, the substrate processing apparatus management system is a substrate processing apparatus management system comprising a management device for managing the substrate processing apparatus and an information analysis device, wherein the substrate processing apparatus comprises an exhaust path to which a predetermined force is applied and a plurality of substrate processing units configured to share the exhaust path, and the information analysis device corresponds to each of the plurality of substrate processing units. Before detection The management device includes a model generation unit that generates group correlations as correlations between multiple processing information indicating operations or states related to air supply and exhaust, and the management device provides multiple processing information corresponding to each of the multiple substrate processing units. When detected The system includes a processing information acquisition unit and a detection unit that detects the state of multiple substrate processing units before they become abnormal, based on comparison information obtained by comparing the correlation between multiple processing information related to exhaust gases among the multiple processing information acquired by the processing information acquisition unit with a group correlation.

[0008] In this scenario, if the group correlation indicates the correlation of processing information when multiple substrate processing units are operating normally, the operation or state of exhaust of multiple substrate processing units sharing an exhaust path can be compared with the normal state. Therefore, it is possible to detect when the operation or state of exhaust in multiple substrate processing units sharing an exhaust path differs from the normal state. Consequently, it is possible to detect the state before an abnormality occurs in any of the multiple substrate processing units sharing an exhaust path. As a result, it is possible to provide a substrate processing device management system that can predict the occurrence of abnormalities in multiple substrate processing units and prevent a decrease in the operating rate of the substrate processing device.

[0009] According to another aspect of this invention, the management device is a management device for managing a substrate processing apparatus, wherein the substrate processing apparatus comprises an exhaust path to which a predetermined force is applied, and a plurality of substrate processing units configured to share the exhaust path, and a group correlation is predetermined as a correlation between a plurality of processing information indicating operations or states related to supply and exhaust, corresponding to each of the plurality of substrate processing units, and comprises a processing information acquisition unit that acquires a plurality of processing information corresponding to each of the plurality of substrate processing units, and a detection unit that detects the state of the plurality of substrate processing units before they become abnormal based on comparison information obtained by comparing the correlation between a plurality of processing information related to exhaust among the plurality of processing information acquired by the processing information acquisition unit with the group correlation.

[0010] In this scenario, if the group correlation indicates the correlation of processing information when multiple substrate processing units are operating normally, the operation or state of exhaust of multiple substrate processing units sharing an exhaust path can be compared to the normal state. Therefore, it is possible to detect when the operation or state of exhaust in multiple substrate processing units sharing an exhaust path differs from the normal state.

[0011] Preferably, the type of processing information obtained from each of the multiple substrate processing units belonging to the group is the same.

[0012] In this scenario, the same type of processing information is acquired among multiple substrate processing units belonging to the group, allowing the correlation between the same type of processing information to be compared with the normal state. Therefore, the correlation between the same type of processing information can be compared among multiple substrate processing units.

[0013] Preferably, the system further includes a deviation degree acquisition unit that acquires deviation degree information for each of the multiple substrate processing units belonging to the group, which indicates the degree of deviation between the predicted value predicted using the group correlation relationship from the processing information acquired by the processing information acquisition unit corresponding to the substrate processing unit and the processing information acquired by the processing information acquisition unit corresponding to the substrate processing unit.

[0014] In this scenario, processing information corresponding to the substrate processing unit is used to predict values ​​using group correlations, allowing the correlation of the operation or state related to the exhaust of the substrate processing unit to be compared with a normal state. Since deviation information indicating the degree of discrepancy between the predicted value and the processing information is obtained, the difference between the correlation of the operation or state related to exhaust and a normal state can be shown.

[0015] Preferably, each of the multiple substrate processing units includes an air supply unit that supplies gas into the substrate processing unit, a first pressure gauge that measures the pressure of the gas supplied into the substrate processing unit, a first control unit that controls the air supply unit so that the pressure measured by the first pressure gauge becomes a first target value, an exhaust unit that adjusts the size of an opening that connects the internal space of the substrate processing unit to an exhaust path, a second pressure gauge that measures the pressure of the gas discharged from the substrate processing unit, and a second control unit that controls the exhaust unit so that the pressure measured by the second pressure gauge becomes a second target value. The processing information includes a first pressure value measured by the first pressure gauge, an operating amount of the air supply unit, an operating amount output by the first control unit to the air supply unit, a second pressure value measured by the second pressure gauge, and an opening degree indicating the size of the opening output by the second control unit to the exhaust unit, with the processing information related to exhaust being the opening degree.

[0016] In this scenario, the correlation between the first pressure value, the amount of operation of the air supply unit, the amount of operation of the air supply unit, the second pressure value, and the size of the opening of the exhaust unit is defined as a group correlation, and the correlation of the opening degrees between multiple substrate processing devices is compared with the normal state. Therefore, the exhaust state can be compared with the normal state between multiple substrate processing units.

[0017] Preferably, each of the multiple substrate processing units processes the substrate according to one of several processing recipes, and the processing information further includes information that identifies the processing recipe to which the substrate is processed by the substrate processing unit.

[0018] Following this approach, it becomes possible to distinguish between processing recipes and compare their correlations to those of the normal state.

[0019] Preferably, the exhaust path has a plurality of distinct divided paths, the exhaust unit includes a switching unit that switches to one of the plurality of divided paths, and the processing information further includes information that identifies the divided path switched by the switching unit.

[0020] Following this approach, it is possible to distinguish between multiple divided paths and compare the correlation between multiple substrate processing devices with that of a normal state.

[0021] Preferably, the substrate processing apparatus has a plurality of exhaust paths, and the plurality of substrate processing units are classified into any of a plurality of groups corresponding to the plurality of exhaust paths. In each of the plurality of groups, the plurality of substrate processing units classified into the group are configured to share the exhaust path corresponding to the group among the plurality of exhaust paths. For each of the plurality of groups, a group correlation relationship is predefined as a correlation relationship between a plurality of process information items corresponding to the plurality of substrate processing units classified into the group and indicating operations or states related to air supply and exhaust. The detection unit, for each of the plurality of groups, detects a state before the plurality of substrate processing units classified into the group become abnormal based on comparison information obtained by comparing the correlation relationship between a plurality of process information items related to exhaust among the plurality of process information items obtained by the process information acquisition unit corresponding to the plurality of substrate processing units classified into the group with the group correlation relationship corresponding to the group.

[0022] According to this aspect, for each of the plurality of exhaust paths, it can be detected that the operations or states related to exhaust in the plurality of substrate processing units sharing the exhaust path are different from the normal state.

[0023] According to another aspect of this invention, the substrate processing apparatus includes the above-described management device.

[0024] According to still another aspect of this invention, a substrate processing apparatus management method is a substrate processing apparatus management method for managing a substrate processing apparatus. The substrate processing apparatus includes an exhaust path to which a predetermined force is applied and a plurality of substrate processing units configured to share the exhaust path. A group correlation relationship is predefined as a correlation relationship between a plurality of process information items corresponding to the plurality of substrate processing units and indicating operations or states related to air supply and exhaust. The management device is caused to execute a process of acquiring a plurality of process information items corresponding to the plurality of substrate processing units and a process of detecting a state before the plurality of substrate processing units become abnormal based on comparison information obtained by comparing the correlation relationship between a plurality of process information items related to exhaust among the plurality of process information items acquired corresponding to the plurality of substrate processing units with the group correlation relationship.

[0025] According to still another aspect of the present invention, a substrate processing apparatus management program is a substrate processing apparatus management program for managing a substrate processing apparatus. The substrate processing apparatus includes an exhaust path to which a predetermined force is applied, and a plurality of substrate processing units configured to share the exhaust path. A group correlation relationship is predetermined as a correlation relationship between a plurality of pieces of processing information corresponding to each of the plurality of substrate processing units and indicating an operation or state related to air supply and exhaust. The computer is caused to execute a process of acquiring a plurality of pieces of processing information corresponding to each of the plurality of substrate processing units, and a process of detecting a state before the plurality of substrate processing units become abnormal based on comparison information obtained by comparing a correlation relationship between a plurality of pieces of processing information related to exhaust among the plurality of pieces of processing information acquired corresponding to each of the plurality of substrate processing units with the group correlation relationship.

Advantages of the Invention

[0026] According to the present invention, it is possible to prevent the operation rate of the substrate processing apparatus from decreasing.

Brief Description of the Drawings

[0027] [Figure 1] It is a diagram for explaining the configuration of a substrate processing apparatus management system according to an embodiment of the present invention. [Figure 2] It is a plan view showing the inside of the substrate processing apparatus of the present embodiment. [Figure 3] It is a cross-sectional view taken along line A-A of the processing block in FIG. 2. [Figure 4] It is a cross-sectional view taken along line B-B of the processing block. [Figure 5] It is a plan view of the substrate processing unit. [Figure 6] It is a system diagram of the first exhaust path and the second exhaust path. [Figure 7] It is a system diagram of the third exhaust path and the fourth exhaust path. [Figure 8] It is a conceptual diagram for explaining the operation of the substrate processing apparatus management system. [Figure 9] It is a diagram for explaining a specific calculation example of the divergence degree. [Figure 10] This figure shows an example of a deviation degree table. [Figure 11] This figure shows an example of a deviation table for "e. Opening degree". [Figure 12] This is a block diagram illustrating an example of the functional configuration of a substrate processing equipment management system. [Figure 13] This flowchart shows an example of the process flow for transmitting processing information. [Figure 14] This is a flowchart showing an example of the model generation process. [Figure 15] This flowchart shows an example of the process flow for managing substrate processing equipment. [Modes for carrying out the invention]

[0028] Hereinafter, a substrate processing apparatus management system according to one embodiment of the present invention will be described in detail with reference to the drawings. In the following description, "substrate" refers to semiconductor substrates (semiconductor wafers), substrates for FPDs (Flat Panel Displays) such as liquid crystal display devices or organic EL (Electro Luminescence) display devices, substrates for optical discs, substrates for magnetic discs, substrates for magneto-optical discs, substrates for photomasks, ceramic substrates, or substrates for solar cells.

[0029] 1. Overall configuration of the substrate processing equipment management system Figure 1 is a diagram illustrating the configuration of a substrate processing apparatus management system according to one embodiment of the present invention. The substrate processing apparatus management system 100 in Figure 1 includes a substrate processing apparatus 1, an information analysis apparatus 3, and a management apparatus 4. The information analysis apparatus 3 is, for example, a server and includes a CPU (Central Processing Unit) and memory. The management apparatus 4 is, for example, a personal computer and includes a CPU and memory. The management apparatus 4 is also capable of attaching and detaching a CD-ROM (Compact Disc Read Only Memory) 4c, which is a computer-readable recording medium, and can read and execute a program recorded on the CD-ROM 4c.

[0030] The information analysis device 3 and the management device 4 are used to manage the substrate processing device 1. Note that the substrate processing device 1 managed by the information analysis device 3 and the management device 4 is not limited to one unit; multiple substrate processing devices 1 may be managed.

[0031] The management device 4 according to this embodiment is connected to the substrate processing device 1 and the information analysis device 3 by wired or wireless communication lines or a communication network. For example, the management device 4 is connected to the substrate processing device 1 and the information analysis device 3 via a communication network such as the Internet or a local area network. In this embodiment, the management device 4 is connected to the substrate processing device 1 and the information analysis device 3 by wired or wireless means.

[0032] The substrate processing apparatus 1 is equipped with a display device, an audio output device, and an operating unit (not shown). The substrate processing apparatus 1 is operated according to a predetermined processing procedure (processing recipe) for the substrate processing apparatus 1.

[0033] 2. Overview of substrate processing equipment Figure 2 is a plan view showing the interior of the substrate processing apparatus of this embodiment. The substrate processing apparatus 1 comprises an indexer unit 6 and a processing block 7. The processing block 7 is connected to the indexer unit 6. The indexer unit 6 and the processing block 7 are arranged horizontally. The indexer unit 6 includes a transport mechanism 6a. The transport mechanism 6a transports the substrate W placed on the carrier mounting unit 6b to the processing block 7.

[0034] The horizontal direction in which the indexer unit 6 and the processing block 7 are aligned is called the "front-to-back direction X". Within the front-to-back direction X, the direction from the processing block 7 toward the indexer unit 6 is called the "forward direction". The direction opposite to the forward direction is called the "backward direction". The horizontal direction perpendicular to the front-to-back direction X is called the "width direction Y". One direction of the "width direction Y" is appropriately called the "rightward direction". The direction opposite to the rightward direction is called the "leftward direction". The direction perpendicular to the horizontal direction is called the "vertical direction Z". In each diagram, forward, backward, right, left, up, and down are shown as appropriate for reference.

[0035] Figure 3 is a cross-sectional view of the processing block in Figure 2 along line AA. Figure 4 is a cross-sectional view of the processing block along line BB. Referring to Figures 2 to 4, the processing block 7 comprises 24 substrate processing units WU11 to WU16, WU21 to WU26, WU31 to WU36, and WU41 to WU46. Each of the substrate processing units WU11 to WU16, WU21 to WU26, WU31 to WU36, and WU41 to WU46 comprises chambers CH11 to CH16, CH21 to CH26, CH31 to CH36, and CH41 to CH46, respectively. Hereinafter, the substrate processing units WU11~WU16, WU21~WU26, WU31~WU36, and WU41~WU46 will be collectively referred to as the substrate processing unit WU, and the chambers CH11~CH16, CH21~CH26, CH31~CH36, and CH41~CH46 will be collectively referred to as the chamber CH.

[0036] The processing block 7 has a hierarchical structure in which six layers, each containing four substrate processing units WU, are stacked vertically. The bottommost first layer contains substrate processing units WU11, WU21, WU31, and WU41; the second layer above it contains substrate processing units WU12, WU22, WU32, and WU42; the third layer above it contains substrate processing units WU13, WU23, WU33, and WU43; the fourth layer above it contains substrate processing units WU14, WU24, WU34, and WU44; the fifth layer above it contains substrate processing units WU15, WU25, WU35, and WU45; and the sixth layer above it contains substrate processing units WU16, WU26, WU36, and WU46.

[0037] In the first to sixth layers, six substrate processing units WU are arranged in overlapping positions in a plan view to form a group. Specifically, substrate processing units WU11 to WU16 constitute the first group Gr1, substrate processing units WU21 to WU26 constitute the second group Gr2, substrate processing units WU31 to WU36 constitute the third group Gr3, and substrate processing units WU41 to WU46 constitute the fourth group Gr4.

[0038] The substrate processing units WU11 to WU16 belonging to the first group Gr1 are configured to share the first air supply path IN1 and the first exhaust path EX1. In other words, the chambers CH11 to CH16 of each substrate processing unit WU11 to WU16 are configured to communicate with the first air supply path IN1 and with the first exhaust path EX1. Similarly, the substrate processing units WU21 to WU26 belonging to the second group Gr2 are configured to share the second air supply path IN2 and the second exhaust path EX2, the substrate processing units WU31 to WU36 belonging to the third group Gr3 are configured to share the third air supply path IN3 and the third exhaust path EX3, and the substrate processing units WU41 to WU46 belonging to the fourth group Gr4 are configured to share the fourth air supply path IN4 and the fourth exhaust path EX4.

[0039] Figure 2 shows a cross-section of the first layer at the bottom of the processing block 7. As shown in Figure 2, the first layer of the processing block 7 contains substrate processing units WU11, WU21, WU31, and WU41. The processing block 7 includes a transport space 73. Substrate processing units WU11 and WU21 are located to the right of the transport space 73, with substrate processing unit WU11 located behind substrate processing unit WU21. Substrate processing units WU31 and WU41 are located to the left of the transport space 73, with substrate processing unit WU31 located behind substrate processing unit WU41.

[0040] The processing block 7 is equipped with a transport mechanism 71 in the transport space 73. The transport mechanism 71 transports the substrate W received from the indexer unit 6 to the chambers CH11, CH21, CH31, and CH41 provided in the substrate processing units WU11, WU21, WU31, and WU41, respectively.

[0041] Figure 5 is a plan view of the substrate processing unit. Substrate processing units WU11-WU16, WU21-WU26, WU31-WU36, and WU41-WU46 all have the same configuration. Here, we will explain using substrate processing unit WU11 as an example.

[0042] The substrate processing unit WU11 includes a chamber CH11, a spin chuck SC for holding and rotating the substrate W, an exhaust unit ED, an air supply unit FFU, and nozzles 61, 62, and 63. The chamber CH11 is a partitioned space for processing the substrate, has an opening through which the substrate W can pass, and has a shutter (not shown) for opening and closing the opening.

[0043] The spin chuck SC and the three nozzles 61, 62, and 63 are located inside the chamber CH11. Nozzle 61 discharges an acidic cleaning solution, nozzle 62 discharges an alkaline cleaning solution, and nozzle 63 discharges an organic solvent cleaning solution. Inside the chamber CH11, the substrate W held by the spin chuck SC is cleaned by supplying cleaning solution from one of the three nozzles 61, 62, or 63.

[0044] An air supply unit (FFU) is located on the top plate of chamber CH11. The air supply unit (FFU) is connected to a first air supply path (IN1) to supply clean gas (e.g., air or inert gas) into chamber CH11. The air supply unit (FFU) is a fan filter unit, for example, including a fan motor. The rotational speed of the fan motor is adjusted in the air supply unit (FFU). This adjusts the flow rate of the gas supplied into chamber CH. A first pressure sensor (PG1) is located inside the air supply unit (FFU). The first pressure sensor (PG1) measures the pressure of the gas supplied to chamber CH11 (hereinafter referred to as the first control pressure).

[0045] The exhaust section ED connects the internal space of the chamber CH11 to the first exhaust path EX1. The chamber CH11 has an opening 43 that opens the internal space of the chamber CH11 to the exhaust section ED. The internal space of the chamber CH11 communicates with the exhaust section ED through the opening 43.

[0046] The exhaust section ED includes an exhaust damper 41 and a switching device 51. The opening degree of the exhaust damper 41 is adjusted by a damper motor, and the opening area of ​​the opening 43 is changed. By adjusting the opening degree of the exhaust damper 41 by the damper motor, the flow rate of the gas discharged from the chamber CH11 is adjusted. A second pressure sensor PG2 is positioned between the exhaust damper 41 and the internal space of the chamber CH11. The second pressure sensor PG2 measures the value of the pressure of the gas discharged from the chamber CH11 (hereinafter referred to as the second control pressure).

[0047] The first exhaust path EX1 includes a first divided path 11 corresponding to acidic chemicals, a second divided path 21 corresponding to alkaline chemicals, and a third divided path 31 corresponding to organic solvents. The first divided path 11 has a first connecting port 12 formed as an opening that communicates with the chamber CH11, the second divided path 21 has a second connecting port 22 formed as an opening that communicates with the chamber CH11, and the third divided path 31 has a third connecting port 32 formed as an opening that communicates with the chamber CH11.

[0048] The switching device 51 comprises a first door 13, a second door 23, and a third door 33. The first door 13 has a rotating shaft 14 at one end that is pivotally supported by the main body of the switching device 51, and opens and closes the first connecting port 12 by rotating around the rotating shaft 14. The second door 23 has a rotating shaft 24 at one end that is pivotally supported by the main body of the switching device 51, and opens and closes the second connecting port 22 by rotating around the rotating shaft 24. The third door 33 has a rotating shaft 34 at one end that is pivotally supported by the main body of the switching device 51, and opens and closes the third connecting port 32 by rotating around the rotating shaft 34. When any of the first door 13, second door 23, and third door 33 are opened and the other two are closed, the gas in the chamber CH11 is guided into one of the first dividing path 11, the second dividing path 21, or the third dividing path 31.

[0049] Specifically, while the substrate is being processed in chamber CH11 with nozzle 61 discharging the chemical solution, the first door 13 is open and the second door 23 and the third door 33 are closed. While the substrate is being processed in chamber CH11 with nozzle 62 discharging the chemical solution, the second door 23 is open and the first door 13 and the third door 33 are closed. While the substrate is being processed in chamber CH11 with nozzle 63 discharging the chemical solution, the third door 33 is open and the first door 13 and the second door 23 are closed.

[0050] Figure 6 is a diagram of the first and second exhaust paths. Referring to Figure 6, each of the chambers CH11 to CH16 belonging to the first group Gr1 is connected to the first exhaust path EX1. An exhaust damper 41 and a second pressure sensor PG2 are provided between the chambers CH11 to CH16 and the first exhaust path EX1. A switching device 51 is provided between the exhaust damper 41 and the first exhaust path EX1. Each of the chambers CH11 to CH16 is connected to one of the first divided path 11, the second divided path 21, and the third divided path 31 of the first exhaust path EX1.

[0051] Each of the chambers CH21 to CH26 belonging to the second group Gr2 is connected to the second exhaust path EX2. An exhaust damper 41 and a second pressure sensor PG2 are provided between the chambers CH21 to CH26 and the second exhaust path EX2. A switching device 51 is provided between the exhaust damper 41 and the second exhaust path EX2. Each of the chambers CH21 to CH26 is connected to one of the first divided path 11, the second divided path 21, and the third divided path 31 of the second exhaust path EX2.

[0052] The first exhaust path EX1 is connected to the first common path CEX11. The first divided path 11 of the first exhaust path EX1 is connected to the first divided common path 11b of the first common path CEX11, the second divided path 21 of the first exhaust path EX1 is connected to the second divided common path 21b of the first common path CEX11, and the third divided path 31 of the first exhaust path EX1 is connected to the third divided common path 31b of the first common path CEX11.

[0053] The second exhaust path EX2 is connected to the first common path CEX11. The first divided path 11 of the second exhaust path EX2 is connected to the first divided common path 11b of the first common path CEX11, the second divided path 21 of the second exhaust path EX2 is connected to the second divided common path 21b of the first common path CEX11, and the third divided path 31 of the second exhaust path EX2 is connected to the third divided common path 31b of the first common path CEX11.

[0054] Power is supplied to the first common path CEX11 from a power generation device within the factory. In the first exhaust path EX1, the path CEX1 from the part communicating with chamber CH16, which is located on the highest level of the first group Gr1, to the part connected to the first common path CEX11, can be shared by chambers CH11 to CH16, which belong to the first group Gr1.

[0055] In the first divided path 11 of the first exhaust path EX1, the path from the portion communicating with chamber CH16 to the portion connected to the first divided common path 11b of the first common path CEX11 can be shared by chambers CH11 to CH16. In the second divided path 21 of the first exhaust path EX1, the path from the portion communicating with chamber CH16 to the portion connected to the second divided common path 21b of the first common path CEX11 can be shared by chambers CH11 to CH16. In the third divided path 31 of the first exhaust path EX1, the path from the portion communicating with chamber CH16 to the portion connected to the third divided common path 31b of the first common path CEX11 can be shared by chambers CH11 to CH16.

[0056] Similarly, in the second exhaust path EX2, the path CEX2 from the part communicating with chamber CH26, which is located at the highest level of the second group Gr2, to the part connected to the first common path CEX11, can be shared by chambers CH21 to CH26, which also belong to the second group Gr2.

[0057] Figure 7 is a diagram of the third and fourth exhaust paths. Referring to Figure 7, each of the chambers CH31 to CH36 belonging to the third group Gr3 communicates with the third exhaust path EX3. An exhaust damper 41 and a second pressure sensor PG2 are provided between the chambers CH31 to CH36 and the third exhaust path EX3. A switching device 51 is provided between the exhaust damper 41 and the first exhaust path EX1. Each of the chambers CH31 to CH36 communicates with one of the first division path 11, the second division path 21, and the third division path 31 of the third exhaust path EX3.

[0058] Chambers CH41 to CH46, belonging to the fourth group Gr4, are connected to the fourth exhaust path EX4. An exhaust damper 41 and a second pressure sensor PG2 are provided between chambers CH41 to CH46 and the fourth exhaust path EX4. A switching device 51 is provided between the exhaust damper 41 and the second exhaust path EX2. Each of chambers CH41 to CH46 is connected to one of the first division path 11, second division path 21, and third division path 31 of the fourth exhaust path EX4.

[0059] The third exhaust path EX3 is connected to the second common path CEX12. The first divided path 11 of the third exhaust path EX3 is connected to the first divided common path 11b of the second common path CEX12, the second divided path 21 of the third exhaust path EX3 is connected to the second divided common path 21b of the second common path CEX12, and the third divided path 31 of the third exhaust path EX3 is connected to the third divided common path 31b of the second common path CEX12.

[0060] The fourth exhaust path EX4 is connected to the second common path CEX12. The first divided path 11 of the fourth exhaust path EX4 is connected to the first divided common path 11b of the second common path CEX12, the second divided path 21 of the fourth exhaust path EX4 is connected to the second divided common path 21b of the second common path CEX12, and the third divided path 31 of the fourth exhaust path EX4 is connected to the third divided common path 31b of the second common path CEX12.

[0061] Power is supplied to the second common path CEX12 from a power generation device within the factory. In the third exhaust path EX3, the path CEX3 from the part communicating with chamber CH36, which is located on the highest level of the third group Gr3, to the part connected to the second common path CEX12, can be shared by chambers CH31 to CH36, which also belong to the third group Gr3.

[0062] In the first divided path 11 of the third exhaust path EX3, the path from the portion communicating with chamber CH36 to the portion connected to the first divided common path 11b of the second common path CEX12 can be shared by chambers CH31 to CH36. In the second divided path 21 of the third exhaust path EX3, the path from the portion communicating with chamber CH16 to the portion connected to the second divided common path 21b of the second common path CEX12 can be shared by chambers CH31 to CH36. In the third divided path 31 of the third exhaust path EX3, the path from the portion communicating with chamber CH36 to the portion connected to the third divided common path 31b of the second common path CEX12 can be shared by chambers CH31 to CH36.

[0063] Similarly, in the fourth exhaust path EX4, the path CEX4 from the part communicating with chamber CH46, which is located at the highest level of the fourth group Gr4, to the part connected to the second common path CEX12, can be shared by chambers CH41 to CH46, which also belong to the fourth group Gr4.

[0064] In this embodiment, a substrate cleaning unit was described as an example of the multiple substrate processing units WU provided by the substrate processing apparatus 1. However, the multiple substrate processing units WU may be a photosensitive film forming unit, a peripheral exposure unit, a developing unit, etc., or a mixture of these may be present.

[0065] 3. Overview of the PCB Processing Equipment Management System Returning to Figure 1, the substrate processing apparatus 1 includes an air supply and exhaust system AES for maintaining a clean atmosphere in the chamber CH11 of the substrate processing unit WU11. The air supply and exhaust system AES includes an air supply section FFU and an exhaust section ED provided in each substrate processing unit WU. The substrate processing apparatus 1 includes a control device 10. The control device 10 controls the air supply section FFU and exhaust section ED for each of the multiple substrate processing units WU. The control device 10 includes a first control unit 10a for controlling the air supply section FFU of each of the multiple substrate processing units WU and a second control unit 10b for controlling the exhaust section ED of each of the multiple substrate processing units WU.

[0066] The substrate processing unit WU needs to be maintained at a constant pressure to prevent particle scattering due to turbulence in the chamber CH. Therefore, the first control unit 10a of the control device 10 controls the air supply unit FFU to set the value of the first control pressure to a predetermined pressure value (hereinafter referred to as the first target pressure value). The second control unit 10b of the control device 10 controls the exhaust unit ED to set the value of the second control pressure to a predetermined pressure value (hereinafter referred to as the second target pressure value).

[0067] In this embodiment, the first control unit 10a controls the power supplied to the fan motor of the air supply unit FFU using PID (proportional-integral-derivative) control based on the difference between a first control pressure and a first target pressure value, in order to maintain a constant pressure of the gas supplied into the chamber CH. As a result, the pressure of the gas supplied into the chamber CH11 is kept constant.

[0068] Furthermore, the second control unit 10b uses PID control to adjust the power supplied to the damper motor of the exhaust section ED based on the difference between the second control pressure and the second target pressure value, in order to maintain a constant pressure of the gas discharged from the chamber CH. This ensures that the pressure of the gas discharged from the chamber CH 11 remains constant. Through the control of the first control unit 10a and the second control unit 10b described above, the pressure inside the chamber CH is maintained at a constant level.

[0069] The six substrate processing units WU11 to WU16 belonging to the first group Gr1 are configured to share the first exhaust path EX1, the six substrate processing units WU21 to WU26 belonging to the second group Gr2 are configured to share the second exhaust path EX2, the six substrate processing units WU31 to WU36 belonging to the third group Gr3 are configured to share the third exhaust path EX3, and the six substrate processing units WU41 to WU46 belonging to the fourth group Gr4 are configured to share the fourth exhaust path EX4. Force is applied to each of the first to fourth exhaust paths EX1 to EX4. The force applied to each of the first to fourth exhaust paths EX1 to EX4 may differ.

[0070] For example, the first exhaust path EX1 is designed to apply sufficient force to exhaust air from all six substrate processing units WU11 to WU16 belonging to the first group Gr1, but the force may fluctuate. Furthermore, the required force is not necessarily the same for each of the six substrate processing units WU11 to WU16 belonging to the first group Gr1.

[0071] In the air supply unit FFU, the responsiveness of the fan in the air supply unit FFU and the actual rotational speed of the fan in relation to the amount of operation of the air supply unit FFU (the rotational speed commanded by the control signal) may differ due to feedback control by the first control unit 10a and differences in the installation positions of the substrate processing units WU11 to WU16. Similarly, in the exhaust unit ED, the opening degree of the exhaust damper 41 may differ due to feedback control by the second control unit 10b and differences in the installation positions of the substrate processing units WU11 to WU16. As a result, the characteristics of each of the first group Gr1 to fourth group Gr4 of the substrate processing device 1 differ.

[0072] 4. Processing Information The substrate processing apparatus 1 is configured with a plurality of processing information items that indicate operations or states related to air supply and exhaust in the substrate processing apparatus 1, as information for managing abnormalities in the substrate processing apparatus 1. In this embodiment, these processing information items are transmitted from the control device 10 of the substrate processing apparatus 1 to the information analysis device 3 via the management device 4 at predetermined intervals, as shown by the thick solid arrows in Figure 1. The processing information items may also be transmitted from the control device 10 to the management device 4 in real time. Alternatively, the processing information items may be transmitted from the control device 10 to a computer separate from the management device 4, and then transmitted from that computer to the management device 4.

[0073] The processing information transmitted from the substrate processing device 1 to the information analysis device 3 via the management device 4 includes: a. First control pressure, b. Operating amount, c. Manipulated amount, d. Second control pressure, e. Opening degree, f. Type of exhaust path, and g. Type of processing recipe.

[0074] "a. First control pressure" is the internal pressure value of the air supply unit FFU, measured by the second pressure sensor PG2. "b. Operation amount" indicates the rotation speed of the fan of the air supply unit FFU. "c. Manipulated amount" is the power value applied to the fan motor of the air supply unit FFU. "d. Second control pressure" is the exhaust pressure value of the chamber CH11, measured by the second pressure sensor PG2. "e. Opening degree" is a value indicating the opening degree of the exhaust damper 41 of the exhaust unit ED. "f. Exhaust path type" indicates one of the first divided path 11, second divided path 21, and third divided path 31 included in the exhaust path. "g. Processing recipe type" indicates the type of processing recipe which is the condition for the substrate processing unit WU to process the substrate. Thus, in this embodiment, processing information related to the air supply unit FFU and the exhaust unit ED is shown.

[0075] 5. Status detection operation of the substrate processing device Figure 8 is a conceptual diagram illustrating the operation of the substrate processing apparatus management system. Referring to Figure 8, in the substrate processing apparatus management system 100 of this embodiment, the multiple substrate processing units WU provided by the substrate processing apparatus 1 are classified into one of several groups. In this embodiment, multiple substrate processing units WU that share an exhaust path are classified into the same group.

[0076] In the substrate processing apparatus management system 100, the correlation of multiple processing information corresponding to multiple substrate processing units WU belonging to the same group is predetermined when multiple substrate processing units WU are operating normally. The correlation of multiple processing information collected from multiple substrate processing units WU belonging to the same group is compared with the predetermined correlation for that group, and based on the comparison result, the state of the supply and exhaust system for the multiple substrate processing units WU belonging to that group as a whole is determined. In Figure 2, PI1 to PI4 correspond to the first group Gr1 to the fourth group Gr4, respectively, and show the processing information corresponding to each of the multiple substrate processing units WU belonging to each group.

[0077] The information analysis device 3 generates a learning model (group correlation) corresponding to a group by performing machine learning using the processing information corresponding to that group. Specifically, the information analysis device 3 defines multiple combinations of two different processing information sets and derives the correlation between the two processing information sets constituting each combination as an invariant relationship (hereinafter referred to as the invariant relationship). For example, the correlation between two processing information sets can be expressed as a function in which one set of processing information sets is a variable, and the other set of processing information sets is a function in which the other set of processing information sets is a variable. The invariant relationship derived by the information analysis device 3 through machine learning of multiple processing information sets is the model.

[0078] The information analysis device 3 generates a first model GM1 corresponding to the first group Gr1 by performing machine learning using processing information PI1 corresponding to the first group Gr1. Similarly, the information analysis device 3 generates a second model GM2 using processing information PI2 corresponding to the second group Gr2, generates a third model GM3 using processing information PI3 corresponding to the third group Gr3, and generates a fourth model GM4 using processing information PI4 corresponding to the fourth group Gr4.

[0079] The control device 4 uses the processing information PI1 corresponding to the first group Gr1 to compare the correlation between the two processing information components of each combination of two different processing information components with the invariant relationship defined in the first model GM1. Specifically, the control device 4 uses the processing information PI1 corresponding to the first group Gr1 to calculate the degree of deviation from the first model GM1 as the deviation score. The control device 4 also calculates the degree of abnormality in the first group Gr1 as the abnormality score based on the calculated deviation score. That is, a low abnormality score indicates that the operation of the substrate processing units WU11 to WU16 belonging to the first group Gr1 is close to the operation of the first model GM1 and that the supply and exhaust system AES is operating normally. A high abnormality score indicates that the operation of the substrate processing units WU11 to WU16 belonging to the first group Gr1 is different from the operation of the first model GM1. Model 1 GM1 indicates that the substrate processing units WU11 to WU16 belonging to Group 1 Gr1 are operating normally. Therefore, if the operation of the substrate processing units WU11 to WU16 belonging to Group 1 Gr1 differs from the operation of Model 1 GM1, there is a high probability that the AES supply and exhaust system is malfunctioning.

[0080] <Example of calculating abnormal score> Next, we will explain a specific example of how to calculate the anomaly score. Models GM1 to GM4 define the correlation between multiple combinations of two different processing information. In order to calculate the anomaly score for a group, the degree of deviation of the processing information corresponding to the group is calculated. Figure 9 is a diagram illustrating a specific example of calculating the degree of deviation. Here, we will explain an example of calculating the degree of deviation corresponding to the combination of "e.opening" of substrate processing unit WU11 and "e.opening" of substrate processing unit WU16, which belong to the first group Gr1. In the following explanation, the "e.opening" data will be referred to as "e" data as appropriate.

[0081] To calculate the degree of deviation, reference data based on the invariant relationship between the "e.opening" of substrate processing unit WU11 and the "e.opening" of substrate processing unit WU16 is required. Therefore, the control device 4 stores the first model GM1 generated by the information analysis device 3 before the actual processing of the substrate W in the substrate processing device 1. The first model GM1 defines the correlation between the "e" data of substrate processing unit WU11 and the "e" data of substrate processing unit WU16 when the substrate processing units WU11 to WU16 belonging to the first group Gr1 are operating normally.

[0082] The first model GM1 is generated by the information analysis device 3 based on the processing information PI1 corresponding to each of the multiple substrate processing units WU11 to WU16 belonging to the first group Gr1, for example, when the substrate processing device 1 is actually operating normally.

[0083] At the top of Figure 9, a graph shows an example of the temporal change in the "e" data of the substrate processing unit WU11 and the substrate processing unit WU16 of the first model GM1. In the "e" data graph, the horizontal axis represents time, and the vertical axis represents the opening degree of the exhaust damper 41.

[0084] As can be seen from the two graphs at the top of Figure 9, as the opening degree of the exhaust damper 41 of substrate processing unit WU11 increases, the opening degree of the exhaust damper 41 of substrate processing unit WU16 also increases at a roughly constant rate. In other words, there is a correlation between the opening degree of the exhaust damper 41 of substrate processing unit WU11 and the opening degree of the exhaust damper 41 of substrate processing unit WU16. When multiple substrate processing units WU11 to WU16 belonging to the first group Gr1 are operating normally, the correlation between each combination of multiple processing information corresponding to each substrate processing unit WU11 to WU16 is equal to the correlation defined in the first model GM1.

[0085] In this state, substrate W is processed in substrate processing unit WU11 and substrate processing unit WU16, and the actual "e" data is collected by the management device 4. In the center of Figure 9, an example of the temporal change in the "e" data of substrate processing unit WU11 and substrate processing unit WU16 is shown in a graph.

[0086] The correlation between the "e" data of board processing unit WU11 and the "e" data of board processing unit WU16 is compared with the correlation between the "e" data of board processing unit WU11 and the "e" data of board processing unit WU16 in the first model GM1. Specifically, the "e" data of board processing unit WU11 is predicted from the "e" data of board processing unit WU16 based on the first model GM1. In other words, the "e" data of board processing unit WU11 is predicted from the "e" data of board processing unit WU16 based on the correlation between the "e" data of board processing unit WU16 and the "e" data of board processing unit WU11 as defined in the first model GM1. Furthermore, the "e" data of board processing unit WU16 is predicted from the "e" data of board processing unit WU11 based on the first model GM1. In other words, based on the correlation between the "e" data of the substrate processing unit WU11 and the "e" data of the substrate processing unit WU16 as defined in the first model GM1, the "e" data of the substrate processing unit WU16 can be predicted from the "e" data of the substrate processing unit WU11.

[0087] At the bottom of Figure 9, a graph shows an example of the temporal change in the "e" data for substrate processing unit WU11 and substrate processing unit WU16, predicted based on the first model GM1. In the graph at the bottom of Figure 9, the predicted "e" data for substrate processing unit WU11 and substrate processing unit WU16 are shown as solid lines, and the predicted "e" data for substrate processing unit WU11 and substrate processing unit WU16 are shown as dotted lines.

[0088] If the board processing units WU11 to WU16 are operating in the same way as the first model GM1, the "e" data of board processing unit WU11 will match or nearly match the predicted "e" data of board processing unit WU11. Similarly, the "e" data of board processing unit WU16 will match or nearly match the predicted "e" data of board processing unit WU16.

[0089] However, if at least one of the board processing units WU11 to WU16 operates differently from the first model GM1, there is a high probability that the "e" data of board processing unit WU11 will deviate from the predicted "e" data of board processing unit WU11. Similarly, there is a high probability that the "e" data of board processing unit WU16 will deviate from the predicted "e" data of board processing unit WU16.

[0090] The degree of this discrepancy is considered to be greater the greater the difference in the operation of the substrate processing units WU11~WU16 compared to the operation of the substrate processing units WU11~WU16 in the first model GM1, and smaller the greater the difference in the operation of the substrate processing units WU11~WU16 compared to the operation of the substrate processing units WU11~WU16 in the first model GM1.

[0091] Therefore, in this embodiment, the difference between the processing information PI1 corresponding to the first group Gr1 and the predicted value predicted from the first model GM1 is calculated as the degree of deviation. In the example in Figure 9, when calculating the degree of deviation at a certain point in time, the management device 4 calculates the difference between the "e" data of the substrate processing unit WU11 of the first group Gr1 and the predicted "e" data of the substrate processing unit WU11 as the degree of deviation. The management device 4 also calculates the difference between the "e" data of the substrate processing unit WU16 of the first group Gr1 and the predicted "e" data of the substrate processing unit WU16 as the degree of deviation.

[0092] Figure 10 shows an example of a deviation degree table. The deviation degree table is a table that shows the degree of deviation for all combinations of processing information. The management device 4 calculates the above degree of deviation for all combinations of processing information. Referring to Figure 10, the 1st to 6th units in the left vertical column of the deviation degree table represent substrate processing units WU11 to WU16. Processing information "a" to "g" is defined for each of the 1st to 6th units in the vertical column. The 1st to 6th units in the upper horizontal column of the deviation degree table also represent substrate processing units WU11 to WU16. Processing information "a" to "g" is defined for each of the 1st to 6th units in the horizontal column.

[0093] For example, the multiple values ​​listed in the row to the right of the processing information "a" for the first unit in the left vertical column of the deviation table represent the degree of deviation between the processing information predicted from each of the processing information "a" to "g" for the first to sixth units in the upper horizontal column and the processing information actually obtained.

[0094] The multiple values ​​listed in the column below the processing information "a" in the horizontal column above the deviation table represent the degree of deviation between the processing information predicted from "a" to "g" for each of the 1st to 6th units in the vertical column on the left and the processing information actually obtained.

[0095] The deviation degree table shown in Figure 10 shows multiple deviation degrees calculated for all combinations of processing information related to substrate processing units WU1 to WU6 belonging to the first group Gr1.

[0096] Anomaly scores are calculated based on a deviation table. In this embodiment, anomaly scores are calculated for each of the six groups, from the first group Gr1 to the sixth group Gr6. For example, let's describe the first group Gr1. For the first group Gr1, the deviation table shown in Figure 10 is generated. In this embodiment, a state is detected in which the exhaust operation or condition of the substrate processing units WU11 to WU16 belonging to the first group Gr1 is not abnormal, but is expected to become abnormal if it continues to operate (hereinafter referred to as the "pre-abnormal state"). Specifically, the pre-abnormal state is detected based on the correlation of the opening degrees of the exhaust dampers 41 of each of the substrate processing units WU11 to WU16.

[0097] Figure 11 shows an example of a deviation table for "e. Opening." The deviation table for "e. Opening" shown in Figure 11 is obtained by extracting the rows and columns for "e. Opening" from the deviation table shown in Figure 10.

[0098] The deviation degree shown in the "e. Opening Degree" deviation degree table represents the difference between the correlation of the opening degrees of each substrate processing unit WU11 to WU16 and the correlation of the opening degrees under normal conditions.

[0099] The sum of the multiple deviations shown in the deviation table for "e. Opening" corresponding to the first group Gr1 shown in Figure 11 is calculated as the anomaly score corresponding to the first group Gr1.

[0100] 6. An example of a functional configuration of a substrate processing equipment management system. Figure 12 is a block diagram illustrating an example of the functional configuration of a substrate processing apparatus management system. Referring to Figure 12, the control device 10 of the substrate processing apparatus 1 includes a processing information acquisition unit 111 and a processing information transmission unit 113. The functions of the control device 10 are realized by the CPU of the control device 10, which executes a control program stored in memory.

[0101] The processing information acquisition unit 111 acquires processing information from each of the multiple substrate processing units WU. Here, the processing information includes "a. First control pressure", "e. Opening degree", "d. Second control pressure", "b. Operating amount", "c. Manipulated amount", "f. Exhaust path type", and "g. Processing recipe type". The processing information acquisition unit 111 acquires processing information from each of the 24 substrate processing units WU at predetermined time intervals. Therefore, the processing information acquired from the substrate processing units WU is time-series data with values ​​defined for each time determined at predetermined time intervals. The processing information transmission unit 113 transmits the multiple processing information acquired from each of the 24 substrate processing units WU to the management device 4.

[0102] The management device 4 includes an information acquisition unit 141, a comparison unit 143, a detection unit 145, a model receiving unit 47, and a management unit 149. The functions of the management device 4 are realized by the CPU of the management device 4, which executes a board processing device management program stored in memory.

[0103] The information collection unit 141 collects processing information indicating the operation and status of the air supply and exhaust while each of the multiple substrate processing units WU processes a substrate. The information collection unit 141 receives multiple processing information corresponding to each of the multiple substrate processing units WU from the control device 10, outputs the received multiple processing information to the comparison unit 143, and also transmits it to the information analysis device 3.

[0104] The model receiving unit 47 receives the first model GM1 to the fourth model GM4 corresponding to the first group Gr1 to the fourth group Gr4 from the information analysis device 3, and outputs the received first model GM1 to the fourth model GM4 to the comparison unit 143. The first model GM1 to the fourth model GM4 determine the correlation between the processing information PI1 to PI4 corresponding to the first group Gr1 to the fourth group Gr4.

[0105] The comparison unit 143 compares the correlation of processing information corresponding to each group from the first group Gr1 to the fourth group Gr4 with the correlation of the model corresponding to that group, and outputs the comparison result to the detection unit 145. For example, let's describe the comparison in the first group Gr1. The comparison unit 143 receives processing information PI1 corresponding to the first group Gr1 from the information acquisition unit 141, and the first model GM1 corresponding to the first group Gr1 from the model receiving unit 47. The processing information PI1 includes the "a" data to "h" data for each of the substrate processing units WU11 to WU16.

[0106] The comparison unit 143 uses the first model GM1 to calculate predicted values ​​corresponding to the "a" data to "h" data for each of the substrate processing units WU11 to WU16 included in the processing information PI1. The comparison unit 143 then calculates the difference between the predicted values ​​and the "a" data to "h" data for each of the substrate processing units WU11 to WU16 as the deviation degree. This generates the deviation degree table shown in Figure 10. The comparison unit 143 outputs the deviation degree table to the detection unit 145.

[0107] The detection unit 145 compares the correlation of exhaust processing information corresponding to each group from the first group Gr1 to the fourth group Gr4 with the correlation of the model corresponding to that group and calculates an anomaly score. Here, we will explain using the first group Gr1 as an example. The deviation degree of "e. Opening degree" is extracted from the deviation degree table input from the comparison unit 143, and the sum of the extracted deviation degrees is calculated as the anomaly score for the first group Gr1.

[0108] The detection unit 145 detects groups in which abnormalities are predicted based on the abnormality scores calculated for each of the first group Gr1 to the fourth group Gr4. For example, the detection unit 145 compares the abnormality score for each of the first group Gr1 to the fourth group Gr4 with a predetermined threshold. The detection unit 145 determines that abnormalities are predicted for groups among the first group Gr1 to the fourth group Gr4 in which the abnormality score is equal to or greater than the threshold. Groups in which the abnormality score is equal to or greater than the threshold are in a state where the operation or state of the exhaust of each of the multiple substrate processing units belonging to that group is not abnormal, but it is predicted that an abnormality will occur if it continues to operate (hereinafter referred to as the "state before abnormality").

[0109] The management unit 149 generates management information regarding countermeasures for the group whose previous state was detected by the detection unit 145. Here, we will explain using the case where the first group Gr1 is determined to be the target as an example. For example, the management unit 149 generates management information that includes a command to close the exhaust damper 41 of the substrate processing units WU11 to WU16 belonging to the first group Gr1 that are not processing substrate W. The management unit 149 may transmit the management information to the substrate processing device 1 and control the substrate processing device 1. In this case, the substrate processing device 1 closes the exhaust damper 41 of the substrate processing units WU11 to WU16 belonging to the first group Gr1 that are not processing substrate W.

[0110] Furthermore, the control unit 149 generates management information that includes commands to reset the damper position of the exhaust section ED and the rotation speed of the fan motor of the air supply section FFU of the substrate processing units WU11 to WU16 belonging to the first group Gr1 to their initial values. The control unit 149 may transmit the management information to the substrate processing device 1 and control the substrate processing device 1. In this case, the substrate processing device 1 resets the damper position of the exhaust section ED and the rotation speed of the fan motor of the air supply section FFU of the substrate processing units WU11 to WU16 belonging to the first group Gr1 to their initial values.

[0111] Furthermore, the management unit 149 generates management information including group identification information to identify the first group Gr1 which has been determined to be the target, and notifies the administrator. This notifies the administrator that it is necessary to change the air supply or exhaust control of the substrate processing units WU11~16 belonging to the first group Gr1.

[0112] Furthermore, the management unit 149 generates management information for the first group Gr1, including multiple deviation degrees calculated by the comparison unit 143, and notifies the administrator. For example, the management unit 149 notifies the administrator of the deviation degree table shown in Figure 10 as management information. This allows the administrator to be notified of information for adjusting the air supply or exhaust control of the substrate processing units WU11 to WU16 belonging to the first group Gr1.

[0113] The management unit 149 may display management information on the display unit provided by the management device 4, or it may send an email containing management information to the administrator, in order to notify the administrator of the management information.

[0114] The information analysis device 3 includes a model generation unit 131 and a model transmission unit 133. The functions of the information analysis device 3 are realized by the CPU of the information analysis device 3, which executes a model generation program stored in memory.

[0115] The model generation unit 131 generates the first model GM1 to the fourth model GM4, corresponding to the first group Gr1 to the fourth group Gr4, using the processing information PI1 to PI4 collected by the information collection unit 141. The processing information PI1 to PI4 used by the model generation unit 44 when generating the first model GM1 to the fourth model GM4 is information collected by the information collection unit 141 while the substrate processing device 1 is operating normally.

[0116] The model generation unit 44 generates the first model GM1 by machine learning the processing information PI1 of the substrate processing units WU11 to WU16 belonging to the first group Gr1. Specifically, the processing information PI1 includes multiple "a" data to "h" data for each of the substrate processing units WU11 to WU16. The model generation unit 131 learns the correlation between each pair of different sets of data included in the processing information PI1.

[0117] Similarly, the model generation unit 44 generates the second model GM2 by machine learning the processing information PI2 of the substrate processing units WU21 to WU26 belonging to the second group Gr2. The processing information PI1 includes multiple "a" data to "h" data for each of the substrate processing units WU21 to WU26. The model generation unit 44 generates the third model GM3 by machine learning the processing information PI3 of the substrate processing units WU31 to WU36 belonging to the third group Gr3. The processing information PI3 includes multiple "a" data to "h" data for each of the substrate processing units WU31 to WU36. The model generation unit 44 generates the fourth model GM4 by machine learning the processing information PI4 of the substrate processing units WU41 to WU46 belonging to the fourth group Gr4. The processing information PI4 includes multiple "a" data to "h" data for each of the substrate processing units WU41 to WU46.

[0118] The model generation unit 131 outputs the generated first model GM1 to fourth model GM4 to the model transmission unit 133. The model transmission unit 133 transmits the first model GM1 to fourth model GM4 to the management device 4.

[0119] Figure 13 is a flowchart showing an example of the processing information transmission process. The processing information transmission process is performed by the CPU of the control device 10 of the substrate processing apparatus 1 when the CPU executes a processing information transmission program stored in memory. Referring to Figure 13, the control device 10 acquires processing information for each of the multiple substrate processing units WU (step S11). For example, the control device 10 acquires processing information PI1, including "a" data to "h" data, for each of the substrate processing units WU11 to WU16 belonging to the first group Gr1, at predetermined intervals.

[0120] In the next step, S12, it is determined whether a predetermined time has elapsed. The predetermined time can be arbitrarily set as a period longer than or equal to a predetermined cycle. If the predetermined time has elapsed, the process proceeds to step S13; otherwise, the process returns to step S11. If the predetermined cycle is repeated multiple times within the predetermined time, the processing information PI1 is a set of processing information equal to the number of times the predetermined cycle was repeated.

[0121] The processes in steps S11 and S12 are performed on all of the multiple substrate processing units WU. Therefore, processing information PI1 to PI4 corresponding to the first group Gr1 to the fourth group Gr4 is obtained.

[0122] In step S13, processing information is transmitted. Processing information corresponding to all of the multiple substrate processing units WU is transmitted to the management device 4, and processing proceeds to step S14. In step S14, it is determined whether or not processing by the substrate processing device 1 has stopped. If processing has stopped, processing ends; otherwise, processing returns to step S11.

[0123] Therefore, while the substrate processing apparatus 1 is processing the substrate, processing information PI1 to PI4 corresponding to the first group Gr1 to the fourth group Gr4 is transmitted to the management device 4. Alternatively, instead of transmitting processing information PI1 to PI4 at predetermined intervals, the control device 10 may transmit all the collected processing information PI1 to PI4 at once, for example, at a time specified by the administrator of the substrate processing apparatus 1, after a longer period than the predetermined time has elapsed.

[0124] Figure 14 is a flowchart showing an example of the model generation process flow. The model generation process is performed by the CPU of the information analysis device 3, which executes a model generation program stored in memory. Referring to Figure 14, the CPU of the information analysis device 3 determines whether or not it has received a model generation command (step S21). It remains in a waiting state until a model generation command is received (NO in step S21), and if a model generation command is received (YES in step S21), the process proceeds to step S22.

[0125] In the next step, S22, processing information for all multiple substrate processing units (WUs) is received, and processing proceeds to step S23. The control device 4 is requested to transmit processing information for all multiple substrate processing units (WUs), and the processing information transmitted from the control device 4 is received. The processing information is the processing information collected by the control device 4 while the substrate processing device 1 is operating normally.

[0126] In step S23, a group Grk (where k is a positive integer) to be processed is selected from among the first group Gr1 to the fourth group Gr4, and the process proceeds to step S24.

[0127] In step S24, a model GMk corresponding to the group Grk to be processed is generated, and the process proceeds to step S25. From the processing information of each of the multiple substrate processing units WU received in step S22, processing information PIk for each of the multiple substrate processing units WU belonging to group Grk is extracted. The processing information PIk is subjected to machine learning to generate the model GMk.

[0128] In step S25, it is determined whether there are any groups among the first group Gr1 to the fourth group Gr4 that were not selected for processing in step S23. If there are unselected groups (YES in step S25), the process returns to step S23; otherwise (NO in step S25), the process proceeds to step S26.

[0129] The loop from steps S23 to S25 is repeated, generating the first model GM1 to the fourth model GM4, which correspond to the first group Gr1 to the fourth group Gr4, respectively. In step S26, the first model GM1 to the fourth model GM4 are sent to the management device 4, and the process ends.

[0130] Figure 15 is a flowchart showing an example of the flow of the substrate processing apparatus management process. The substrate processing apparatus management process is a process executed by the CPU of the management device 4, which executes the substrate processing apparatus management program stored in memory.

[0131] Referring to Figure 15, the CPU of the management device 4 determines whether or not it has received an analysis command (step S31). It remains in a waiting state until an analysis command is received (NO in step S31), and if an analysis command is received (YES in step S31), the process proceeds to step S32. The analysis command may be entered each time processing information PI1 to PI4 is collected. Alternatively, the administrator of the substrate processing device 1 may enter the analysis command into the management device 4.

[0132] In step S32, the first model GM1 to the fourth model GM4 are received, and processing proceeds to step S33. A model is requested from the information analysis device 3, and the first model GM1 to the fourth model GM4, corresponding to the first group Gr1 to the fourth group Gr4 transmitted from the information analysis device 3, are received.

[0133] In step S33, the group Grk to be analyzed is selected from the first group Gr1 to the fourth group Gr4, and the process proceeds to step S34. Here, k is an integer from 1 to 4. In step S34, the processing information PIk corresponding to the group Grk selected for analysis in step S33 is selected from the processing information PI1 to PI4.

[0134] In the next step, S35, the degree of deviation between the model GMk corresponding to the group Grk selected for analysis is calculated. Specifically, for each of the "a" to "h" data of multiple substrate processing units WU included in the processing information PIk corresponding to the group Grk selected for analysis, a predicted value is obtained using the model GMk, and the difference between the predicted value and the actual data is calculated as the degree of deviation. This generates the degree of deviation table shown in Figure 10.

[0135] In the next step, S36, an abnormal score is calculated. In step S35, the sum of the deviations related to exhaust processing information from among the multiple deviations calculated is calculated as the abnormal score. In this embodiment, the deviation used to calculate the abnormal score is the deviation of "e. Opening," which indicates the opening of the exhaust damper 41. For example, in the first group Gr1, the sum of the deviations shown in the deviation table of "e. Opening" shown in Figure 11 is calculated as the abnormal score.

[0136] In step S37, it is determined whether there are any groups among the first group Gr1 to the fourth group Gr4 that were not selected for analysis in step S33. If there are any unselected groups, the process returns to step S33; otherwise, the process proceeds to step S38.

[0137] The loop from step S33 to step S37 is repeated, and an anomaly score corresponding to each of the four groups, from Group 1 (Gr1) to Group 4 (Gr4), is calculated.

[0138] In step S38, groups with abnormality scores above a threshold are extracted from the first group Gr1 to the fourth group Gr4, and the process proceeds to step S39. The groups extracted in step S38 are those in which the exhaust operation of each of the multiple substrate processing units WU included in them is not abnormal, but is expected to become abnormal if it continues to operate (a state before it becomes abnormal). In step S39, management information is generated and output, and the process ends. For example, the management information includes a command to close the exhaust damper 41 of the multiple substrate processing units WU belonging to the group with abnormality scores above a threshold from the first group Gr1 to the fourth group Gr4, which are not processing substrate W. In this case, the management device 4 may transmit this management information to the substrate processing device 1.

[0139] Furthermore, the management information may include commands to reset the damper position of the exhaust section ED and the rotation speed of the fan motor of the air supply section FFU for each of the multiple substrate processing units WU belonging to the group whose abnormality score is above a threshold, selected from the first group Gr1 to the fourth group Gr4, to their initial values. In this case, the management device 4 may transmit this management information to the substrate processing device 1.

[0140] Furthermore, the management information may include group identification information for identifying groups from the first group Gr1 to the fourth group Gr4 whose abnormality scores are above a threshold. In this case, the management device 4 may notify the administrator of this management information. The administrator can then know that it is necessary to adjust the air supply or exhaust control of multiple substrate processing units WU belonging to groups whose abnormality scores are above a threshold.

[0141] Furthermore, the management information may include a deviation degree table calculated in step S35 for groups from the first group Gr1 to the fourth group Gr4 whose abnormality scores are above a threshold. In this case, the management device 4 may notify the administrator of the management information. The administrator can then obtain information for adjusting the air supply unit FFU and exhaust unit ED provided by multiple substrate processing units WU belonging to groups whose abnormality scores are above a threshold.

[0142] 7. Effects of the Embodiment The substrate processing apparatus management system 100 in this embodiment comprises a management device 4 for managing the substrate processing apparatus 1 and an information analysis device 3. The substrate processing apparatus 1 comprises a first exhaust path EX1, a second exhaust path EX2, and a third exhaust path EX3 to which a predetermined force is applied, and 24 substrate processing units WU that process substrates W, each classified into one of the first group Gr1, second group Gr2, and third group Gr3, corresponding to the first exhaust path EX1, second exhaust path EX2, and third exhaust path EX3, respectively. In each of the first group Gr1, second group Gr2, and third group Gr3, for example, substrate processing units WU11 to WU16 classified into the first group Gr1 are configured to share the first exhaust path EX1 corresponding to the first group Gr1. The information analysis device 3 includes a model generation unit 131 that generates a first model GM1 that shows group correlations as correlations between multiple processing information indicating operations or states related to supply and exhaust, corresponding to, for example, substrate processing units WU11 to WU16 classified in the first group Gr1, for each of the first group Gr1, second group Gr2, and third group Gr3. The management device 4 includes an information collection unit 141 that acquires multiple processing information corresponding to each of the multiple substrate processing units WU, and a detection unit 147 that detects the state of any of the substrate processing units WU11 to WU16 classified in the first group Gr1 before they become abnormal, based on comparison information obtained by comparing the correlations between multiple processing information related to exhaust among the multiple processing information acquired by the information collection unit 141 corresponding to, for example, substrate processing units WU11 to WU16 classified in the first group Gr1, for each of the first group Gr1, second group Gr2, and third group Gr3, with the first model GM1 corresponding to the first group Gr1.

[0143] Therefore, for example, if the first model GM1 shows the correlation of multiple processing information when each of the substrate processing units WU11 to WU16 is operating normally, the operation or state of the exhaust of the substrate processing units WU11 to WU16 that share the first exhaust path EX1 can be compared with the normal state. Furthermore, it can be detected that the operation or state of the exhaust of the substrate processing units WU11 to WU16 that share the first exhaust path EX1 is different from the normal state. Thus, in the first group Gr1, it is possible to detect a state in which the operation or state of the exhaust of each of the substrate processing units WU11 to WU16 that share the first exhaust path EX1 is not abnormal, but is expected to become abnormal if it continues to operate (a state before it becomes abnormal). The same applies to the second group Gr2 and the third group Gr3. An example of a case in which any of the multiple substrate processing units WU becomes abnormal is when the pressure in the chamber CH of any of the multiple substrate processing units WU falls below the lower threshold. Therefore, since the state before any of the multiple substrate processing units (WU) malfunctions is detected, there is no need to stop the substrate processing device 1, and opportunities for reduced operating time and throughput can be decreased.

[0144] Furthermore, within each of the four groups Gr1 through Gr4, for example, the types of processing information obtained from each of the substrate processing units WU11 through WU16 belonging to Group Gr1 are the same. The types of processing information are: a. First control pressure, b. Operating amount, c. Manipulated amount, d. Second control pressure, e. Opening degree, f. Exhaust path type, and g. Processing recipe type. Therefore, within each of the four groups Gr1 through Gr4, for example, the same types of processing information are obtained between the substrate processing units WU11 through WU16 belonging to Group Gr1, so the correlation between processing information of the same type can be compared with that under normal conditions. Consequently, the correlation between processing information of the same type can be compared between the substrate processing units WU11 through WU16 belonging to Group Gr1.

[0145] Furthermore, for each of the four groups, Group 1 Gr1 to Group 4 Gr4, for example, for each substrate processing unit WU11 to WU16 belonging to Group 1 Gr1, deviation information is obtained that shows the degree of discrepancy between the processing information acquired in relation to substrate processing unit WU11 and the predicted value predicted using the first model GM1 from that processing information. Therefore, the correlation of the operation or state related to exhaust for each of the substrate processing units WU11 to WU16 belonging to Group 1 Gr1 can be compared with the normal state. Since deviation information showing the degree of discrepancy between the predicted value and the processing information is obtained, it is possible to show the difference between the correlation of the operation or state related to supply and exhaust and the normal state.

[0146] Furthermore, each of the multiple substrate processing units WU includes, for example, an air supply unit FFU that supplies gas into the substrate processing unit WU11, a first pressure sensor PG1 that measures the pressure of the gas supplied into the substrate processing unit WU11, a first control unit 10a that controls the air supply unit so that the pressure measured by the first pressure sensor PG1 becomes a first target value, and an exhaust damper connected to a first exhaust path EX1 corresponding to the first group Gr1 to which the substrate processing unit WU11 belongs, which adjusts the size of the opening 43 that connects the internal space of the substrate processing unit WU11 to the first exhaust path EX1. The system includes an exhaust section ED equipped with a 41, a second pressure sensor PG2 that measures the pressure of the gas discharged from the substrate processing unit WU11, and a second control unit 10b that controls the exhaust section ED so that the pressure measured by the second pressure sensor PG2 becomes a second target value. The processing information includes a first pressure value measured by the first pressure sensor PG1, the operating amount of the air supply section FFU, the operation amount output by the first control unit 10a to the air supply section FFU, a second pressure value measured by the second pressure sensor PG2, and an opening degree indicating the size of the opening output by the second control unit 10b to the exhaust section ED. Therefore, the correlation between the first pressure value, the operating amount of the air supply section FFU, the operation amount of the air supply section FFU, the second pressure value, and the size of the opening of the exhaust section ED can be compared with the system in a normal state and with other substrate processing units.

[0147] Furthermore, each of the first exhaust path EX1, second exhaust path EX2, third exhaust path EX3, and fourth exhaust path EX4 is, for example, the first exhaust path EX1 has a first divided path 11, a second divided path 21, and a third divided path 31 that are different from each other, and the exhaust section includes a switching device 51 that switches to one of ED, the first divided path 11, the second divided path 21, and the third divided path 31, and the processing information further includes information that identifies the path that the substrate processing unit WU communicates with among the first divided path 11, the second divided path 21, and the third divided path 31. Therefore, the first divided path 11, the second divided path 21, and the third divided path 31 can be distinguished and their correlation can be compared with that of a normal state.

[0148] Furthermore, each of the multiple substrate processing units WU processes the substrate according to one of several processing recipes, and the processing information further includes information that identifies the processing recipe to which the substrate W is processed by the substrate processing unit WU. Therefore, multiple processing recipes are distinguished and the current state is compared with the normal state.

[0149] 8. Other Embodiments (1) In the above embodiment, the control device 4 calculates the deviation degree table shown in Figure 10, but the present invention is not limited thereto. The deviation degree table of "e. Opening degree" shown in Figure 11 may be calculated without calculating the deviation degree table shown in Figure 10. Furthermore, although an example of using "e. Opening degree" as processing information for exhaust gas has been shown, the present invention is not limited thereto. In addition to "e. Opening degree", or separately, "second control pressure" may be used as processing information for exhaust gas.

[0150] (2) In the above embodiment, the processing information includes "f. type of exhaust path" and "g. type of processing recipe," but the present invention is not limited thereto. The processing information of the substrate processing apparatus 1 does not have to include "f. type of exhaust path" and "g. type of processing recipe." In this case, the first model GM1 to the fourth model GM4 may be generated for each "f. type of exhaust path." Also, the first model GM1 to the fourth model GM4 may be generated for each "g. type of processing recipe." Furthermore, the first model GM1 to the fourth model GM4 may be generated for each combination of "f. type of exhaust path" and "g. type of processing recipe."

[0151] (3) In the above embodiment, the substrate processing apparatus 1 is a single-wafer type substrate cleaning apparatus, but the present invention is not limited thereto. The substrate processing apparatus 1 may be a batch type substrate cleaning apparatus rather than a single-wafer type, as long as it is equipped with a plurality of substrate processing units WU, and may have a configuration that performs processing other than cleaning. The plurality of substrate processing units WU may be a photosensitive film forming unit, a peripheral exposure unit, a developing unit, etc., or they may be mixed together.

[0152] (4) The multiple processing information may include other physical quantities in addition to or instead of the multiple specific examples described in the above embodiment. The physical quantity may include at least one physical quantity such as the rotational speed, rotational velocity, acceleration, gas flow rate, temperature, humidity, and pressure of the motor provided in the chamber. The multiple processing information may also include information regarding the output signal of a detector provided in the motor.

[0153] (5) In the above embodiment, the management device 4 includes a management unit 149, but the present invention is not limited thereto. For example, management information may be transmitted from the management device 4 to the substrate processing device 1, and the management information may be output by the substrate processing device 1, such as being displayed.

[0154] (6) In the substrate processing apparatus management system 100 according to the above embodiment, the series of processes performed by the information analysis device 3 may be performed by the management device 4. Alternatively, the series of processes performed by the information analysis device 3 may be performed by the substrate processing apparatus 1. In this case, the information analysis device 3 becomes unnecessary.

[0155] (7) In the above embodiment, the substrate processing apparatus 1 and the management apparatus 4 are provided separately, but the substrate processing apparatus 1 and the management apparatus 4 may be provided as an integrated unit. Also, in the above embodiment, the information analysis apparatus 3 and the management apparatus 4 are provided separately, but the information analysis apparatus 3 and the management apparatus 4 may be provided as an integrated unit.

[0156] (8) In the management device 4 of the above embodiment, a representative model or an average model is generated based on invariant relationships, but the present invention is not limited thereto. For example, in the management device 4, a representative model or an average model may be generated by using other machine learning methods such as deep learning.

[0157] (9) In the above embodiment, processing information PI1 to PI4 is transmitted from the processing information transmission unit 113 of the substrate processing apparatus 1 to the information collection unit 141 of the management device 4, but the present invention is not limited thereto. For example, processing information PI1 to PI4 may be transmitted via a cloud on the internet or the like. In this case, the substrate processing apparatus 1 and the management device 4 do not need to communicate directly, so the communication load can be reduced.

[0158] (10) In the above embodiment, the management device 4 includes a model receiving unit 47 and a comparison unit 143, but the present invention is not limited thereto. For example, the management device 4 may transmit each of the processing information PI1 to PI4 collected by the information collection unit 141 to the information analysis device 3 and request the information analysis device 3 to calculate the degree of deviation and the anomaly score. In this case, the information analysis device 3 generates a degree of deviation table and transmits the degree of deviation table to the management device 4.

[0159] 9. Correspondence between each component of the claim and each part of the embodiment The following describes an example of the correspondence between each component of the claim and each element of the embodiment. In the above embodiment, paths CEX1 to CEX4 are examples of exhaust paths, substrate processing units WU11 to 16 are examples of multiple substrate processing units, information analysis device 3 is an example of an information analysis device, model generation unit 131 is an example of a model generation unit, management device 4 is an example of a management device, processing information PI1 is an example of multiple processing information, information collection unit 141 is an example of a processing information acquisition unit, and detection unit 145 is an example of a detection unit. Comparison unit 143 is an example of a deviation degree acquisition unit, air supply unit FFU is an example of an air supply unit, first pressure sensor PG1 is an example of a first pressure gauge, first control unit 10a is an example of a first control unit, exhaust unit ED is an example of an exhaust unit, second pressure sensor PG2 is an example of a second pressure gauge, and second control unit 10b is an example of a second control unit. The first division path 11, the second division path 21, and the third division path 31 are examples of multiple division paths, and the switching device 51 is an example of a switching unit. [Explanation of symbols]

[0160] 100...Substrate processing device management system, 1...Substrate processing device, 3...Information analysis device, 4...Management device, 10...Control device, 10a...First control unit, 10b...Second control unit, 11...First division path, 11b...First division common path, 21...Second division path, 21b...Second division common path, 31...Third division path, 31b...Third division common path, 41...Exhaust damper, 43...Opening, 44...Model generation unit, 47...Model receiving unit, 51...Switching device, 111...Processing information acquisition unit, 113...Processing information transmission unit, 131...Model generation unit, 133...Model transmission unit, 141...Information collection unit, 143...Comparison unit, 14 5...Detection unit, 149...Management unit, AES...Air supply and exhaust system, CEX1~CEX4...Path, CEX11...First common path, CEX12...Second common path, CH...Chamber, ED...Exhaust unit, EX1~EX4...First exhaust path~Fourth exhaust path, FFU...Air supply unit, GM1~GM4...First model~Fourth model, Gr1~Gr4...First group~Fourth group, IN1~IN4...First air supply path~Fourth air supply path, PG1...First pressure sensor, PG2...Second pressure sensor, PI1~PI4...Processing information, PIk...Processing information, SC...Spin chuck, W...Substrate, WU...Substrate processing unit.

Claims

1. A substrate processing apparatus management system comprising a control device for managing substrate processing apparatus and an information analysis device, The substrate processing apparatus includes an exhaust path to which a predetermined force is applied, The system comprises a plurality of substrate processing units configured to share the aforementioned exhaust path, The information analysis device includes a model generation unit that generates group correlations as correlations between multiple processing information items that indicate operations or states related to supply and exhaust before detection, each corresponding to one of the multiple substrate processing units. The management device includes a processing information acquisition unit that acquires a plurality of processing information corresponding to each of the plurality of substrate processing units at the time of detection, A substrate processing apparatus management system comprising: a detection unit that detects the state of the plurality of substrate processing units before they become abnormal, based on comparison information obtained by comparing the correlation between a plurality of processing pieces of processing information related to exhaust, among the plurality of processing pieces of processing information obtained by the processing information acquisition unit, with the group correlation.

2. A control device for managing a substrate processing apparatus, The substrate processing apparatus includes an exhaust path to which a predetermined force is applied, The system comprises a plurality of substrate processing units configured to share the aforementioned exhaust path, A group correlation is predetermined as the correlation between multiple processing information items that correspond to each of the aforementioned multiple substrate processing units and indicate the operation or status related to air supply and exhaust. A processing information acquisition unit that acquires a plurality of processing information corresponding to each of the plurality of substrate processing units, A management device comprising: a detection unit that detects the state of the plurality of substrate processing units before they become abnormal, based on comparison information obtained by comparing the correlation between a plurality of processing pieces of processing information related to exhaust, among the plurality of processing pieces of processing information obtained by the processing information acquisition unit, with the group correlation.

3. The management device according to claim 2, wherein the type of processing information obtained from each of the plurality of substrate processing units is the same.

4. The management device according to claim 2 or 3, further comprising a deviation degree acquisition unit that acquires deviation degree information for each of the plurality of substrate processing units, which indicates the degree of deviation between a predicted value predicted using the group correlation relationship from processing information acquired by the processing information acquisition unit corresponding to the substrate processing unit and the processing information acquired by the processing information acquisition unit corresponding to the substrate processing unit.

5. Each of the aforementioned plurality of substrate processing units includes an air supply unit that supplies gas into the substrate processing unit, A first pressure gauge for measuring the pressure of the gas supplied into the substrate processing unit, A first control unit controls the air supply unit so that the pressure measured by the first pressure gauge becomes a first target value, An exhaust section that adjusts the size of an opening that connects the internal space of the substrate processing unit and the exhaust path, A second pressure gauge for measuring the pressure of the gas discharged from the substrate processing unit, The system includes a second control unit that controls the exhaust section so that the pressure measured by the second pressure gauge becomes a second target value, The processing information includes a first pressure value measured by the first pressure gauge, the amount of operation of the air supply unit, the amount of operation output by the first control unit to the air supply unit, a second pressure value measured by the second pressure gauge, and an opening degree indicating the size of the opening output by the second control unit to the exhaust unit. The control device according to any one of claims 2 to 4, wherein the processing information relating to the exhaust is the opening degree.

6. Each of the aforementioned substrate processing units processes the substrate according to one of several processing recipes. The management device according to claim 5, wherein the processing information further includes information that identifies the processing recipe in which the substrate is processed by the substrate processing unit.

7. The exhaust path has a plurality of separate paths that are different from each other. The exhaust section includes a switching section that switches to one of the plurality of divided paths, The management device according to claim 5 or 6, wherein the processing information further includes information that identifies the divided path switched by the switching unit.

8. The substrate processing apparatus has a plurality of exhaust paths, The plurality of substrate processing units are classified into one of the plurality of groups corresponding to each of the plurality of exhaust paths, In each of the aforementioned multiple groups, the multiple substrate processing units classified into the group are configured to share the exhaust path corresponding to the group among the multiple exhaust paths. For each of the aforementioned multiple groups, the group correlation is predetermined as a correlation between multiple processing information items that indicate the operation or state related to supply and exhaust, corresponding to each of the multiple substrate processing units classified into the group. The control device according to any one of claims 2 to 7, wherein the detection unit detects the state of a plurality of substrate processing units classified into a group before they become abnormal, based on comparison information obtained by comparing the correlation between a plurality of processing pieces of processing information related to exhaust, which is acquired by the processing information acquisition unit corresponding to each of the plurality of substrate processing units classified into the group, with the group correlation corresponding to the group.

9. A substrate processing apparatus comprising a control device according to any one of claims 2 to 8.

10. A method for managing a substrate processing apparatus, The substrate processing apparatus includes an exhaust path to which a predetermined force is applied, The system comprises a plurality of substrate processing units configured to share the aforementioned exhaust path, A group correlation is predetermined as the correlation between multiple processing information items that correspond to each of the aforementioned multiple substrate processing units and indicate the operation or status related to air supply and exhaust. The process of acquiring multiple processing information corresponding to each of the aforementioned multiple substrate processing units, A substrate processing apparatus management method that causes a management device to execute a process to detect the state of the plurality of substrate processing units before they become abnormal, based on comparison information obtained by comparing the correlation between a plurality of processing pieces of processing information related to exhaust, which is acquired in relation to each of the plurality of substrate processing units, with the group correlation.

11. A substrate processing apparatus management program for managing substrate processing apparatus, The substrate processing apparatus includes an exhaust path to which a predetermined force is applied, The system comprises a plurality of substrate processing units configured to share the aforementioned exhaust path, A group correlation is predetermined as the correlation between multiple processing information items that correspond to each of the aforementioned multiple substrate processing units and indicate the operation or status related to air supply and exhaust. The process of acquiring multiple processing information corresponding to each of the aforementioned multiple substrate processing units, A substrate processing apparatus management program that causes a computer to perform a process to detect the state of the plurality of substrate processing units before they become abnormal, based on comparison information obtained by comparing the correlation between a plurality of processing pieces of processing information related to exhaust, which is acquired in relation to each of the plurality of substrate processing units, with the group correlation.

Images