Substrate processing apparatus management system, support apparatus, substrate processing apparatus, method for comparing performance between chambers, and program for comparing performance between chambers

Abstract

To provide information regarding performance differences between a plurality of chambers included in a substrate processing device.SOLUTION: A substrate processing device management system 100 includes: an information analysis device including a model generation unit 35 that generates a representative model indicating a correlation between a plurality of pieces of processing information PIrp indicating operations or states related to substrate processing in a representative chamber Crp among the plurality of chambers C1 to Cn; and a support device that obtains the plurality of pieces of processing information for the respective chambers, and generates support information regarding maintenance work for other chambers on the basis of comparison information obtained by comparing a correlation between the plurality of pieces of processing information in the other chambers, with the representative model.SELECTED DRAWING: Figure 2

Description

【Technical Field】 【0001】 The present invention relates to a substrate processing apparatus management system, a support apparatus, a substrate processing apparatus, a method for comparing performance between chambers, and a program for a method for comparing performance between chambers. In particular, the present invention relates to a substrate processing apparatus management system including a support apparatus and an information analysis apparatus for assisting maintenance work of a substrate processing apparatus having a plurality of chambers for processing a substrate, the support apparatus, a substrate processing apparatus including the support apparatus, a method for comparing performance between chambers executed by the support apparatus, and a program for a method for comparing performance between chambers that causes a computer to execute the method for comparing performance between chambers. 【Background Art】 【0002】 A substrate processing apparatus includes a plurality of chambers partitioned as spaces for processing a substrate such as a semiconductor substrate (semiconductor wafer). The plurality of chambers are composed of the same type of components so as to perform the same type of processing on the substrate. The plurality of chambers are adjusted so that the same performance is exhibited in the plurality of chambers. 【0003】 For example, Japanese Unexamined Patent Application Publication No. 2019-140196 discloses a technique for evaluating a unit process by comparing time-series data obtained in a unit process with predetermined reference data. According to the technique described in Japanese Unexamined Patent Application Publication No. 2019-140196, in each of the plurality of chambers, the operations of the plurality of components can be adjusted to match a reference. 【Prior Art Documents】 【Patent Documents】 【0004】 【Patent Document 1】 Japanese Unexamined Patent Application Publication No. 2019-140196 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0005】 Although multiple chambers are manufactured to exhibit identical performance, processing results may differ even when the same processing is performed on a substrate due to factors such as the position of the chamber in the substrate processing apparatus, slight variations in construction, and individual differences in the various components that make up the chamber. However, because the differences in the operation and state of substrate processing between multiple chambers are slight, they may not be detectable even when time-series data is compared with reference data. 【0006】 The object of the present invention is to provide a substrate processing apparatus management system, support device, substrate processing apparatus, inter-chamber performance comparison method, and inter-chamber performance comparison program that can provide information on the performance differences between multiple chambers of a substrate processing apparatus. [Means for solving the problem] 【0013】 This invention be According to the description, the support device is a support device for maintenance work on a substrate processing apparatus having multiple chambers for processing substrates, and comprises a processing information collection unit that acquires multiple processing information indicating operations or states related to substrate processing in each of the multiple chambers, and a support information generation unit that generates support information regarding maintenance work on other chambers based on comparison information obtained by comparing the correlation between multiple processing information in a predetermined representative chamber among the multiple chambers with the correlation between multiple processing information in other chambers, wherein the support information generation unit performs a doubling process on at least one target processing information among the multiple processing information by combining predetermined reference information as new processing information with the multiple processing information, and then calculates the correlation between the multiple processing information after the doubling process. 【0014】 In this scenario, it is possible to combine reference information that is correlated with target processing information that is uncorrelated or only slightly correlated with other processing information. Therefore, it is possible to provide information regarding the performance difference between a representative chamber and other chambers with respect to processing information that is uncorrelated or only slightly correlated with other processing information. 【0015】 Preferably, the comparison information is information that has been weighted and compared according to the correlation between multiple processing information. According to another aspect of this invention, the support device is a support device for assisting maintenance work on a substrate processing apparatus having a plurality of chambers for processing substrates, comprising: a processing information collection unit that acquires a plurality of processing information indicating operations or states related to the processing of substrates in each of the plurality of chambers; and a support information generation unit that generates support information for maintenance work on other chambers based on comparison information obtained by comparing the correlation between the plurality of processing information collected while a predetermined representative chamber among the plurality of chambers is processing a substrate with the correlation between the plurality of processing information in other chambers, wherein the support information generation unit calculates the correlation between the plurality of processing information after performing logarithmic arithmetic on at least one target processing information among the plurality of processing information, and the support information generation unit calculates the correlation between the plurality of processing information after performing a doubling process on at least one target processing information among the plurality of processing information by combining predetermined reference information as new processing information with the plurality of processing information. Preferably, the comparison information is information that has been weighted and compared according to the correlation between multiple processing information. 【0017】 Preferably, the system further includes a representative chamber determination unit that selects one of the multiple chambers as a representative chamber, which has the correlation closest to the correlation between the averages of multiple processing information from each of the multiple chambers. 【0018】 Following this procedure, a representative chamber is determined from among multiple chambers whose operation or state related to substrate processing is close to the average, thus enabling the automatic determination of the representative chamber. 【0019】 Preferably, each of the plurality of chambers includes a discharge valve for discharging liquid and a flow rate measuring unit for measuring the flow rate of the liquid discharged from the discharge valve, and each of the plurality of processing information includes an open signal indicating the amount of opening input to the discharge valve to control the amount of opening of the discharge valve, and the flow rate measured by the flow rate measuring unit. 【0020】 In this scenario, the difference in the correlation between the open signal and flow rate between the representative chamber and other chambers can be used to generate support information regarding maintenance work on the discharge valves of other chambers. 【0021】 Preferably, each of the plurality of chambers includes a discharge valve for discharging liquid and a concentration measuring unit for measuring the concentration of the liquid, and each of the plurality of processing information includes an open signal indicating the amount of opening input to the discharge valve to control the amount of opening of the discharge valve, and the concentration measured by the concentration measuring unit. 【0022】 In this scenario, the difference in the correlation between the open signal and concentration between the representative chamber and other chambers can be used to generate support information regarding maintenance work on other chambers. 【0023】 Preferably, each of the multiple chambers includes a concentration measuring unit for measuring the concentration of a liquid and an indoor temperature measuring unit for measuring the temperature of the chamber, and each of the multiple processing information includes the concentration measured by the concentration measuring unit and the temperature measured by the indoor temperature measuring unit. 【0024】 In this scenario, support information can be generated to assist with maintenance work on other chambers based on the differences in the correlation between liquid concentration and chamber temperature between the representative chamber and other chambers. 【0025】 Preferably, each of the multiple chambers includes a concentration measuring unit for measuring the concentration of a liquid and a liquid temperature measuring unit for measuring the temperature of a liquid, and each of the multiple processing information includes the concentration measured by the concentration measuring unit and the temperature measured by the liquid temperature measuring unit. 【0026】 In this scenario, it is possible to generate support information regarding maintenance work on other chambers based on the differences in the correlation between liquid concentration and liquid temperature between the representative chamber and other chambers. 【0027】 Preferably, the system further includes a deviation degree acquisition unit that uses the correlation between multiple processing information in a representative chamber to acquire deviation degree information as comparison information, which indicates the degree of deviation between multiple predicted values ​​predicted from multiple processing information in other chambers and multiple processing information acquired for other chambers. 【0028】 In this scenario, multiple predicted values ​​are predicted from multiple processing information in other chambers using the correlation between multiple processing information in the representative chamber, allowing for a comparison of the correlation between multiple processing information between the representative chamber and the other chambers. Since deviation information indicating the degree of discrepancy between multiple predicted values ​​and multiple processing information in other chambers is obtained as comparison information, it is possible to show the difference in the correlation between multiple processing information between the representative chamber and the other chambers. 【0029】 According to another aspect of the present invention, the substrate processing apparatus includes the above-described support apparatus. 【0030】 According to this aspect, it is possible to provide a substrate processing apparatus capable of detecting a performance difference between a plurality of chambers. 【0032】 child According to still another aspect of the present invention, a method for comparing chamber performance is a method for comparing chamber performance executed by a support apparatus that supports maintenance work of a substrate processing apparatus including a plurality of chambers for processing a substrate, the method including: a process of acquiring a plurality of process information indicating operations or states related to processing of a substrate in each of the plurality of chambers; and a process of generating support information regarding maintenance work of other chambers based on comparison information obtained by comparing a correlation relationship between a plurality of process information in a predetermined representative chamber among the plurality of chambers and a correlation relationship between a plurality of process information in other chambers. The process of generating support information includes calculating a correlation relationship between a plurality of process information after a duplication process in which predetermined reference information is combined as new process information with at least one target process information among the plurality of process information. According to still another aspect of the present invention, a program for comparing chamber performance is a program for comparing chamber performance executed by a computer included in a support apparatus that supports maintenance work of a substrate processing apparatus including a plurality of chambers for processing a substrate, the program causing the computer to execute: a process of acquiring a plurality of process information indicating operations or states related to processing of a substrate in each of the plurality of chambers; and a process of generating support information regarding maintenance work of other chambers based on comparison information obtained by comparing a correlation relationship between a plurality of process information in a predetermined representative chamber among the plurality of chambers and a correlation relationship between a plurality of process information in other chambers. The process of generating support information includes calculating a correlation relationship between a plurality of process information after a duplication process in which predetermined reference information is combined as new process information with at least one target process information among the plurality of process information. 【Advantages of the Invention】 【0033】 According to the present invention, it becomes possible to provide information regarding the performance differences between multiple chambers provided in a substrate processing apparatus. [Brief explanation of the drawing] 【0034】 [Figure 1] This figure illustrates the overall configuration of a substrate processing apparatus management system according to one embodiment of the present invention. [Figure 2] This is a conceptual diagram illustrating the operation of a substrate processing equipment management system. [Figure 3] This diagram illustrates a specific example of how to calculate the degree of deviation. [Figure 4] This figure shows an example of a deviation degree table. [Figure 5] This is a block diagram illustrating an example of the functional configuration of a substrate processing equipment management system. [Figure 6] This flowchart shows an example of the process flow for transmitting processing information. [Figure 7] This is a flowchart showing an example of the model generation process. [Figure 8] This is a flowchart illustrating an example of the process for comparing performance between chambers. [Figure 9] This figure shows an example of a weighting table. [Figure 10] This is a block diagram illustrating an example of the functional configuration of a substrate processing apparatus management system in a second modified example. [Figure 11] This is a flowchart illustrating an example of the process for determining the representative chamber. [Figure 12] This is a block diagram illustrating an example of the functional configuration of a substrate processing apparatus management system in a third modified example. [Figure 13] This figure shows an example of a deviation degree table in the third modified example. [Figure 14] This flowchart shows an example of the model generation process flow in the third modified example. [Figure 15]This flowchart shows an example of the flow of the inter-chamber performance comparison process in the third modified example. [Modes for carrying out the invention] 【0035】 The following describes, with reference to drawings, a support device, a chamber-to-chamber performance comparison method, and a chamber-to-chamber performance comparison program according to one embodiment of the present invention. In the following description, "substrate" refers to semiconductor substrates, FPD (Flat Panel Display) substrates such as liquid crystal display devices or organic EL (Electro Luminescence) display devices, optical disk substrates, magnetic disk substrates, magneto-optical disk substrates, photomask substrates, ceramic substrates, or solar cell substrates. In the following description, the person who manages the substrate processing device is referred to as the "manager," and the person who performs maintenance work on the substrate processing device is referred to as the "maintenance worker." 【0036】 1. Overall configuration of the substrate processing equipment management system First, the overall configuration of the substrate processing apparatus management system, including the support device according to this embodiment, will be described. Figure 1 is a diagram illustrating the overall configuration of a substrate processing apparatus management system according to one embodiment of the present invention. As shown in Figure 1, the substrate processing apparatus management system 100 consists of a substrate processing apparatus 1, an information analysis device 3, and a support device 4. The information analysis device 3 is, for example, a server and includes a CPU (Central Processing Unit) and memory. The support device 4 is, for example, a personal computer and includes a CPU and memory. The support device 4 also has a removable CD-ROM (Compact Disc Read Only Memory) 4c, which is a computer-readable recording medium, and the support device 4 can read and execute a program recorded on the CD-ROM 4c. 【0037】 The information analysis device 3 and the support 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 support device 4 is not limited to one unit; multiple substrate processing devices 1 may be managed. 【0038】 The support device 4 according to this embodiment is used to assist maintenance workers in performing maintenance work on the substrate processing device 1. The support device 4 is connected to each of the substrate processing device 1 and the information analysis device 3 by wired or wireless communication lines or a communication network. For example, the support device 4 is connected to each of 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 support device 4 is connected to the substrate processing device 1 and the information analysis device 3 by wired or wireless means. 【0039】 2. An example configuration of the substrate processing apparatus 1 As shown in Figure 1, the substrate processing apparatus 1 is a batch-type substrate cleaning apparatus and includes a control device 10 and a plurality of chambers C1 to Cn, where n is a positive integer. A chamber is a partitioned space for processing substrates. The number of chambers C1 to Cn is not limited, but here we will explain using the case where n=12 as an example. In addition to the plurality of chambers C1 to Cn, the substrate processing apparatus 1 is also provided with a display device, an audio output device, and an operating unit (not shown). 【0040】 Each of the multiple chambers C1 to Cn is composed of multiple identical components. Here, we will explain using chamber C1 as an example. Chamber C1 includes a processing tank 111, a substrate holding section 112, a lifting device 112a, a bubbler tube 115, a heater 117, and a heater drive unit 118. 【0041】 The substrate holding section 112 is configured to hold multiple substrates W. The processing tank 111 is configured to accommodate the multiple substrates W held by the substrate holding section 112. A cleaning solution for cleaning the substrates W is stored in the processing tank 111. 【0042】 The lifting device 112a supports the substrate holder 112 so that it can move vertically, and moves the substrate holder 112 vertically under the control of the control device 10. This allows the lifting device 112a to immerse the multiple substrates W held by the substrate holder 112 in the cleaning solution stored in the processing tank 111, and to lift the multiple substrates W immersed in the cleaning solution from the processing tank 111. The multiple substrates W are cleaned by being immersed in the cleaning solution. The lifting device 112a is also equipped with a motor (not shown) as a power source for moving the substrate holder 112 vertically. 【0043】 The treatment tank 111 is connected to a chemical solution piping 113 and a pure water piping 114. The chemical solution piping 113 is equipped with a chemical solution valve 113a and a flow meter 124. The chemical solution valve 113a is a control valve whose opening degree can be adjusted by the control device 10. 【0044】 The chemical piping 113 guides the chemical supplied from a chemical supply system (not shown) to the treatment tank 111 when the chemical valve 113a is opened. The flow rate (amount supplied per unit time) of the chemical supplied to the treatment tank 111 through the chemical piping 113 changes according to the degree of opening of the chemical valve 113a. Examples of chemicals used include BHF (buffered hydrofluoric acid), DHF (dilute hydrofluoric acid), hydrofluoric acid, hydrochloric acid, sulfuric acid, nitric acid, phosphoric acid, acetic acid, oxalic acid, or ammonia. The flow meter 124 detects the flow rate of the chemical in the portion of the chemical piping 113 downstream of the chemical valve 113a and provides the detection result to the control device 10. 【0045】 The pure water piping 114 is equipped with a pure water valve 114a and a flow meter 125. The pure water valve 114a is a control valve whose opening degree can be adjusted by the control device 10. When the pure water valve 114a is opened, the pure water piping 114 guides pure water supplied from a pure water supply system (not shown) to the treatment tank 111. The flow rate (amount supplied per unit time) of pure water supplied to the treatment tank 111 through the pure water piping 114 changes according to the opening degree of the pure water valve 114a. The flow meter 125 detects the flow rate of pure water in the part of the pure water piping 114 downstream of the pure water valve 114a and provides the detection result to the control device 10. 【0046】 A bubbler pipe 115 is fixed inside the processing tank 111, near its lower end, by a bubbler fixing device (not shown). A gas pipe 116 is connected to the bubbler pipe 115. A gas valve 116a and a flow meter 127 are provided on the gas pipe 116. The gas valve 116a is a control valve whose opening degree can be adjusted by the control device 10. 【0047】 The gas piping 116 guides nitrogen gas supplied from a gas supply system (not shown) to a bubbler pipe 115 in the treatment tank 111 when the gas valve 116a is opened. The flow rate (amount supplied per unit time) of nitrogen gas supplied to the bubbler pipe 115 through the gas piping 116 changes according to the opening degree of the gas valve 116a. The flow meter 127 detects the flow rate of nitrogen gas in the portion of the gas piping 116 downstream of the gas valve 116a and provides the detection result to the control device 10. In this case, the control device 10 adjusts the opening degree of the gas valve 116a based on the flow rate provided by the flow meter 127 so that nitrogen gas is supplied to the bubbler pipe 115 at a predetermined flow rate. 【0048】 By adjusting the opening of the chemical valve 113a and the pure water valve 114a, the chemical solution and pure water are supplied to the treatment tank 111 in predetermined proportions. In this state, the gas valve 116a is further opened. This supplies nitrogen gas to the bubbler tube 115, generating a large amount of bubbles from the bubbler tube 115. As the generated bubbles rise in the treatment tank 111, the chemical solution and pure water supplied to the treatment tank 111 are mixed, and a cleaning solution is produced. Note that other inert gases such as argon gas may be supplied to the bubbler tube 115 instead of nitrogen gas. 【0049】 A heater 117 is further provided inside the processing tank 111, near its lower end. The heater 117 has a path through which the cleaning liquid passes, and the cleaning liquid is heated by the heater 117 as it passes through the path. The heater drive unit 118 drives the heater 117 by supplying power to it. The heater thermometer 118a detects the surface temperature of the heater 117 and provides the detection result to the control device 10. The outlet thermometer 118b is provided near the outlet of the path of the heater 117 and detects the temperature of the cleaning liquid immediately after it has been heated by the heater 117 and provides the detection result to the control device 10. Based on the surface temperature provided by the heater thermometer 118a and the temperature of the cleaning liquid provided by the outlet thermometer 118b, the control device 10 determines the amount of power to supply to the heater 117 and controls the heater drive unit 118. 【0050】 Furthermore, a drain pipe is connected to the bottom of the treatment tank 111. A valve (not shown) is provided in the drain pipe. When the valve in the drain pipe is opened, the cleaning liquid inside the treatment tank 111 is discharged to the outside of chamber C1. 【0051】 The processing tank 111 is further equipped with a liquid level gauge 121, a concentration meter 122, a first thermometer 123, and a second thermometer 123a as multiple detectors for detecting the state of the washing liquid stored in the processing tank 111. 【0052】 The liquid level gauge 121 detects the liquid level of the cleaning solution stored in the treatment tank 111 and provides the detection result to the control device 10. The concentration meter 122 detects the concentration of the cleaning solution stored in the treatment tank 111 and provides the detection result to the control device 10. The first thermometer 123 is located inside the chamber C1 but outside the treatment tank 111 and detects the temperature inside the chamber C1 and provides the detection result to the control device 10. The second thermometer 123a is installed inside the treatment tank 111 and detects the temperature of the cleaning solution accumulated in the treatment tank 111 and provides the detection result to the control device 10. 【0053】 The control device 10 consists of, for example, a CPU (Central Processing Unit) and memory, and controls the operation of the lifting device 112a, the operation of the heater drive unit 118, and the opening degrees of the chemical valve 113a, the pure water valve 114a, and the gas valve 116a. The memory of the control device 10 stores a program for executing the cleaning process of the substrate W by the substrate processing device 1. 【0054】 Furthermore, the control device 10 adjusts the opening of the chemical valve 113a, the pure water valve 114a, the gas valve 116a, and the valve of the drain pipe (not shown) based on the detection results of the flow meters 124 and 125, the liquid level meter 121, and the concentration meter 122. As a result, the liquid level of the cleaning solution, which has a predetermined chemical concentration, is maintained at a predetermined height while the cleaning solution is stored in the treatment tank 111. Based on the detection result of the chemical concentration provided by the concentration meter 122, the control device 10 provides feedback control to the amount of chemical solution supplied to the treatment tank 111 so that the chemical concentration of the cleaning solution stored in the treatment tank 111 is within a predetermined range. For example, if the chemical concentration of the cleaning solution detected by the concentration meter 122 is lower than a predetermined range, the control device 10 controls the chemical valve 113a to supply additional chemical solution to the treatment tank 111 from a chemical supply system (not shown) through the chemical piping 113. 【0055】 Furthermore, the control device 10 controls the heater drive unit 118 based on the temperatures supplied by the first thermometer 123, the second thermometer 123a, the heater thermometer 118a, and the outlet thermometer 118b, respectively, so that the temperature of the cleaning liquid stored in the processing tank 111 is maintained at a predetermined temperature. 【0056】 3. Processing Information Each of the multiple chambers C1 to Cn provided by the substrate processing apparatus 1 is assigned processing information indicating the operation or state related to the processing of the substrate W. This processing information is transmitted from the control device 10 of the substrate processing apparatus 1 to the support device 4 at predetermined intervals, as shown by the thick solid arrows in Figure 1. The processing information may also be transmitted from the control device 10 to the support device 4 in real time. Alternatively, the processing information may be transmitted from the control device 10 to a computer separate from the support device 4, and then transmitted from that computer to the support device 4. 【0057】 The processing information transmitted from the control device 10 of the substrate processing apparatus 1 to the support device 4 includes: a. chemical solution concentration, b. chemical solution flow rate, c. room temperature, d. chemical solution temperature, e. heater outlet temperature, f. heater surface temperature, g. concentration FB input amount, h. input valve opening degree, and i. heater ON signal. 【0058】 "a. Chemical concentration" indicates the chemical concentration of the cleaning solution in the treatment tank 111 as detected by the concentration meter 122. "b. Chemical flow rate" indicates the flow rate of the chemical solution flowing through the chemical solution piping 113 as detected by the flow meter 124. "c. Chemical temperature" indicates the temperature inside the chamber as detected by the first thermometer 123. "d. Chemical temperature" indicates the temperature of the chemical solution accumulated in the treatment tank 111 as detected by the second thermometer 123a. "e. Heater outlet temperature" indicates the temperature of the chemical solution near the cover outlet of the heater 117 as detected by the outlet thermometer 118b. "f. Heater surface temperature" indicates the surface temperature of the heater 117 as detected by the heater thermometer 118a. "g. Concentration FB input amount" is a value calculated by the control device 10 from the flow rate detected by the flow meter 124, and indicates the amount of chemical solution introduced into the treatment tank 111 from the chemical solution piping 113 so that the chemical solution concentration of the cleaning solution stored in the treatment tank 111 is within a predetermined range. "h. Input valve opening" indicates the opening degree of the chemical solution valve 113a provided in the chemical solution piping 113. "i. Heater ON signal" is a signal output by the heater drive unit 118 to control the heater 117. 【0059】 4. Inter-chamber difference detection operation Figure 2 is a conceptual diagram illustrating the operation of the substrate processing apparatus management system. Referring to Figure 2, in the substrate processing apparatus management system 100 of this embodiment, a representative chamber Crp is predetermined from among the multiple chambers C1 to Cn provided by the substrate processing apparatus 1. Then, in the substrate processing apparatus management system 100, the correlation of multiple processing information for each of the multiple chambers C1 to Cn provided by the substrate processing apparatus 1 is compared with the correlation of multiple processing information for the representative chamber Crp, and the operation and state of substrate processing for each of the multiple chambers C1 to Cn are determined based on the comparison result. In Figure 2, PI1 to PIn indicate the processing information corresponding to each of the multiple chambers C1 to Cn. 【0060】 The information analysis device 3 generates a model corresponding to the representative chamber Crp by performing machine learning using the processing information PIrp corresponding to the representative chamber Crp. Here, the model corresponding to the representative chamber Crp generated by the information analysis device 3 is called the representative model. Specifically, the information analysis device 3 uses the processing information PIrp to define multiple combinations of two different processing information, and derives the correlation between the two processing information constituting each combination as an invariant relationship (hereinafter referred to as the invariant relationship). For example, the correlation between two processing information can be expressed as a function in which one variable is the other, and the other variable is the other. The invariant relationship derived by the information analysis device 3 through machine learning of multiple processing information PIrp is the representative model. 【0061】 The support device 4 uses processing information PI1 to PIn, corresponding to each of the multiple chambers C1 to Cn, to compare the correlation between the two processing pieces constituting each combination of two different processing pieces for each of the multiple combinations of two different processing pieces for each of the multiple chambers C1 to Cn with the invariant relationship defined in the representative model. The support device 4 compares each of the multiple chambers C1 to Cn with the representative model. Here, we will explain using the comparison between chamber C1 and the representative model as an example. The support device 4 uses processing information PI1 corresponding to chamber C1 to compare the correlation between multiple combinations of two different processing pieces with the invariant relationship defined in the representative model. Specifically, the support device 4 uses processing information PI1 corresponding to chamber C1 to calculate the degree of deviation from the representative model as the deviation degree. The support device 4 also calculates... Ta Based on the degree of deviation, the degree of abnormality in chamber C1 is calculated as an abnormality score. Specifically, a low abnormality score indicates that chamber C1's operation is close to that of the representative chamber Crp and is functioning normally. A high abnormality score indicates that chamber C1's operation differs from that of the representative chamber Crp. If the operation differs from that of the representative chamber Crp, there is a high probability that the operation is abnormal. 【0062】 <Example of calculating abnormal score> Next, we will explain a specific example of how to calculate the anomaly score. The representative model defines the correlation between multiple combinations of two different processing information. In order to calculate the anomaly score for chamber C1, the degree of deviation of the processing information PI1 corresponding to chamber C1 is calculated. Figure 3 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 "h. Inlet valve opening" and "b. Chemical flow rate" of the processing information shown in Figure 1. In the following explanation, the data for "h. Inlet valve opening" will be referred to as "h" data as appropriate, and the data for "b. Chemical flow rate" will be referred to as "b" data as appropriate. 【0063】 To calculate the degree of deviation, reference data based on the invariant relationship between "h. valve opening" and "b. chemical flow rate" is required. Therefore, the support device 4 stores a representative model generated by the information analysis device 3 before the actual substrate W is processed in the substrate processing device 1. The representative model defines the correlation between the "h" data and the "b" data when the representative chamber Crp is operating normally according to a predetermined processing recipe. 【0064】 A representative model is generated by the information analysis device 3 based on processing information PIrp corresponding to a representative chamber Crp selected from among multiple chambers C1 to Cn when the substrate processing device 1 is actually operating normally. 【0065】 At the top of Figure 3, a graph shows an example of the temporal changes in the "h" and "b" data of a representative model. In the "h" data graph, the horizontal axis represents time, and the vertical axis shows the opening degree of the chemical valve 113a installed in the chemical pipe 113. In the "b" data graph, the horizontal axis represents time, and the vertical axis shows the flow rate of the chemical liquid flowing through the chemical pipe 113 as detected by the flow meter 124. The horizontal axis (time axis) is common to both the "h" data graph and the "b" data graph. 【0066】 As can be seen from the two graphs at the top of Figure 3, as the opening degree of the chemical valve 113a increases, the flow rate of the chemical solution flowing through the chemical pipe 113, as detected by the flow meter 124, also increases at a nearly constant rate. In other words, there is a correlation between the opening degree of the chemical valve 113a and the flow rate of the chemical solution flowing through the chemical pipe 113. When the representative chamber Crp is operating normally, the correlation in each combination of multiple processing information is equal to the correlation defined in the representative model. 【0067】 In this state, the substrate W is processed in chamber C1, and the actual "h" data and "b" data are collected by the support device 4. In the center of Figure 3, an example of the temporal changes in the "h" data and "b" data of chamber C1 is shown in a graph. 【0068】 The correlation between the "h" data and "b" data in Chamber C1 is compared with the correlation between the "h" data and "b" data in the representative Chamber Crp. Specifically, the "b" data is predicted from the "h" data in Chamber C1 based on the representative model. In other words, the "b" data is predicted from the "h" data in Chamber C1 based on the correlation between the "h" data and the "b" data as defined by the representative model. Furthermore, the "h" data is predicted from the "b" data in Chamber C1 based on the representative model. In other words, the "h" data is predicted from the "b" data in Chamber C1 based on the correlation between the "b" data and the "h" data as defined by the representative model. 【0069】 The lower part of Figure 3 shows a graph illustrating an example of the temporal changes in the "h" and "b" data predicted based on the representative model. In the graph at the bottom of Figure 3, the predicted "h" and "b" data are shown as solid lines, while the "h" and "b" data for chamber C1 are shown as dotted lines. 【0070】 If chamber C1 operates in the same manner as the representative chamber Crp, the "h" data from chamber C1 will match or nearly match the predicted "b" data. Similarly, the "b" data from chamber C1 will match or nearly match the predicted "h" data. However, if chamber C1 operates differently from the representative chamber Crp, there is a high probability that the "h" data from chamber C1 will diverge from the predicted "h" data. Similarly, there is a high probability that the "b" data from chamber C1 will diverge from the predicted "b" data. The degree of this divergence is considered to be greater the greater the difference between the operation of chamber C1 and the operation of the representative chamber Crp, and smaller the less the difference between the operation of chamber C1 and the operation of the representative chamber Crp. 【0071】 Therefore, in this embodiment, the difference between the processing information PI1 corresponding to chamber C1 and the predicted value predicted from the representative model is calculated as the degree of deviation. In the example in Figure 3, when calculating the degree of deviation at a certain point in time, the support device 4 calculates the difference between the "h" data of chamber C1 and the predicted "h" data as the degree of deviation. The support device 4 also calculates the difference between the "b" data of chamber C1 and the predicted "b" data as the degree of deviation. 【0072】 Figure 4 shows an example of a deviation table. The deviation table is a table that shows the degree of deviation for every combination of processing information. The support device 4 calculates the above degree of deviation for every combination of processing information. Referring to Figure 4, the multiple values ​​listed in the rows to the right of each of the processing information "a" to "i" 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 "i" in the upper horizontal column and the processing information actually obtained. The multiple values ​​listed in the columns below each of the processing information "a" to "i" in the upper horizontal column of the deviation table represent the degree of deviation between the processing information predicted from each of the "a" to "i" in the left vertical column and the processing information actually obtained. 【0073】 For example, the value "10" in the column at the intersection of the rightmost row of processing information "a" in the left vertical column and the lowermost column of processing information "b" in the upper horizontal column represents the degree of deviation between the predicted value of processing information "a" predicted from processing information "b" and the processing information "a" corresponding to chamber C1. Similarly, the value "11" in the column at the intersection of the lowermost column of processing information "a" in the upper horizontal column and the rightmost row of processing information "b" in the left vertical column represents the degree of deviation between the predicted value of processing information "b" predicted from processing information "a" and the processing information "b" corresponding to chamber C1. 【0074】 The deviation degree table shown in Figure 4 shows multiple deviation degrees calculated for all combinations of processing information related to chamber C1. Once all deviation degrees have been calculated, the support device 4 calculates the sum of the multiple calculated deviation degrees as the anomaly score corresponding to chamber C1. In the deviation degree table shown in Figure 4, the anomaly score is, 264 This is the result. 【0075】 5. An example of the functional configuration of support device 4 Figure 5 is a block diagram illustrating an example of the functional configuration of a substrate processing apparatus management system. Referring to Figure 5, the control device 10 provided in the substrate processing apparatus 1 includes a processing information acquisition unit 11 and a processing information transmission unit 13. The functions of the control device 10 are realized by the CPU provided in the control device, which executes a control program stored in memory. 【0076】 The processing information acquisition unit 11 acquires processing information PI1 to PIn from each of the multiple chambers C1 to Cn. Here, we will explain using the case where processing information PI1 is acquired from chamber C1 as an example. The processing information acquisition unit 11 acquires processing information PI1 from chamber C1 at predetermined time intervals. Therefore, the multiple processing information PI1s are time-series data with values ​​defined for each time determined at predetermined time intervals. The processing information transmission unit 13 transmits the processing information PI1 to PIn acquired from each of the chambers C1 to Cn to the support device 4. Since chambers C1 to Cn process the substrate according to the processing recipe, the processing information PI1 to PIn may be time-series data collected during the period while chambers C1 to Cn process the substrate according to the processing recipe. 【0077】 The support device 4 includes an information collection unit 41, a comparison unit 43, a target determination unit 45, a model receiving unit 47, and a support unit 49. The functions of the support device 4 are realized by the CPU of the support device 4, which executes a support information generation program stored in memory. 【0078】 The information collection unit 41 collects processing information PI1 to PIn, which indicates the operation and status of each of the multiple chambers C1 to Cn processing the substrate. The information collection unit 41 receives the multiple processing information PI1 to PIn from the control device 10, outputs the received multiple processing information PI1 to PIn to the comparison unit 43, and also transmits it to the information analysis device 3. 【0079】 The model receiving unit 47 receives a representative model from the information analysis device 3 and outputs the received representative model to the comparison unit 43. The representative model determines the correlation of the processing information PIrp corresponding to the representative chamber Crp. 【0080】 The comparison unit 43 compares the correlation between each of the processing information PI1 to PIn corresponding to each of the multiple chambers C1 to Cn with the correlation between the representative model and outputs the comparison result to the target determination unit 45. For example, let's describe the processing information PI1 corresponding to chamber C1. The processing information PI includes multiple "a" data to "i" data. The comparison unit 43 uses the representative model to compare each of the "a" data to "i" data of the processing information PI1. prediction The unit calculates a value and determines the deviation between the "a" data to "i" data and the predicted value. Furthermore, the comparison unit 43 calculates the sum of the multiple deviations calculated in accordance with the processing information PI1 as an anomaly score. 【0081】 The target determination unit 45 determines the chambers to be maintained based on the abnormality score calculated by the comparison unit 43. For example, the target determination unit 45 compares the abnormality score of each of the multiple chambers C1 to Cn with a predetermined threshold. The target determination unit 45 determines the chambers among the multiple chambers C1 to Cn whose abnormality score is equal to or greater than the predetermined threshold as the chambers to be maintained. 【0082】 The support unit 49 generates support information regarding maintenance work on the chamber that has been determined to be the target of maintenance by the target determination unit 45. Here, we will explain using the case where chamber C1 has been determined to be the target of maintenance as an example. The support unit 49 notifies the maintenance worker of chamber identification information to identify chamber C1, which has been determined to be the target of maintenance, as support information. This notifies the maintenance worker that chamber C1 needs to be maintained. 【0083】 Furthermore, the support unit 49 displays multiple deviations calculated by the comparison unit 43 for the processing information PI1 for the chamber C1 as support information. For example, the deviation table shown in Figure 4 is notified to the maintenance worker as support information. This allows the maintenance worker to be notified of information that will help them determine which of the multiple components of the chamber C1 requires adjustment. 【0084】 The support unit 49 may display support information on the display unit provided by the support device 4, or it may send an email containing support information to the maintenance worker, in order to notify the maintenance worker of the support information. 【0085】 From the deviation table shown in Figure 4, for example, the deviation of the row and column for "a. Chemical concentration" allows us to understand the fluctuation of the chemical concentration. Also, the deviation of the row and column for "d. Chemical temperature" allows us to understand the fluctuation of the chemical temperature. Furthermore, the deviation of the row and column for "e. Heater outlet temperature" and "f. Heater surface temperature" allows us to understand whether the temperature rise is stable. Furthermore, the deviation of the row and column for "g. Concentration FB input amount" allows us to understand whether the feedback control of the chemical concentration is stable. Furthermore, the deviation of the row and column for "b. Chemical flow rate" and "h. Input valve opening" allows us to understand whether the rise of the chemical flow rate is stable. Furthermore, the deviation of the row and column for "i. Heater ON signal" allows us to understand the timing delay of the heater control. 【0086】 The information analysis device 3 includes a representative information receiving unit 31, a model generation unit 33, and a model transmission unit 35. 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. 【0087】 The representative information receiving unit 31 receives multiple processing information PIrp corresponding to the representative chamber Crp. The representative chamber Crp is a chamber selected from among multiple chambers C1 to Cn. The representative chamber Crp is selected by the administrator or maintenance worker who manages the substrate processing apparatus 1. For example, the substrate processing apparatus 1 is driven to perform processing on the substrate W, and for each of the multiple chambers C1 to Cn, the results of inspecting the substrate W after processing in that chamber are stored as processing results associated with that chamber. The administrator or maintenance worker can refer to the processing results and select the chamber with the best processing results from among the multiple chambers C1 to Cn as the representative chamber. The processing information PIrp received by the representative information receiving unit 31 is processing information corresponding to the processing results that the maintenance worker referred to in order to select the representative chamber Crp. 【0088】 The model generation unit 33 generates a representative model by machine learning the processing information PIrp corresponding to the representative chamber Crp. Specifically, the processing information PIrp includes multiple data from "a" to "i". The model generation unit 33 learns the correlation between each pair of different data contained in the processing information PIrp. The model generation unit 33 outputs the generated representative model to the model transmission unit 35. The model transmission unit 35 transmits the representative model to the support device 4. 【0089】 Figure 6 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 6, the control device 10 acquires processing information for each of the multiple chambers C1 to Cn (step S11). For example, the control device 10 acquires processing information PI1 from chamber C1 at a predetermined interval. 【0090】 In the next step, S12, it is determined whether a predetermined time has elapsed. The predetermined time may be, for example, the time it takes for chamber C1 to process one substrate according to the processing recipe. If the predetermined time has elapsed, the process proceeds to step S13; otherwise, the process returns to step S11. The processes in steps S11 and S12 are performed for each of the multiple chambers C1 to Cn. For this reason, processing information PI1 to PIn corresponding to each of the multiple chambers C1 to Cn is acquired. 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. 【0091】 In step S13, processing information PI1 to PIn corresponding to each of the multiple chambers C1 to Cn is transmitted to the support 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. Therefore, while the substrate processing device 1 is performing processing, processing information PI1 to PIn corresponding to each of the multiple chambers C1 to Cn is transmitted to the support device 4. Note that instead of transmitting the processing information PI1 to PIn at predetermined intervals, the control device 10 may transmit all the processing information PI1 to PIn collected up to that point at a time specified by the administrator of the substrate processing device 1, for example, after a longer period of time has elapsed. 【0092】 Figure 7 is a flowchart illustrating 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 7, the CPU of the information analysis device 3 determines whether or not it has received a representative model generation command (step S21). It remains in a waiting state until a representative model generation command is received (NO in step S21), and if a representative model generation command is received (YES in step S21), the process proceeds to step S22. 【0093】 In the next step S22, processing information PI1 to PIn is received, and processing proceeds to step S23. The support device 4 is requested to transmit processing information PI1 to PIn, and the processing information PI1 to PIn transmitted from the support device 4 is received. Processing information PI1 to PIn is processing information collected by the support device 4 while the substrate processing device 1 is operating normally. 【0094】 In step S23, a representative chamber is determined, and the process proceeds to step S24. The representative chamber is determined to be selected by the administrator or maintenance worker from among several chambers C1 to Cn. 【0095】 In step S24, a representative model is generated, and the process proceeds to step S25. From the processing information PI1 to PIn received in step S22, processing information PIrp corresponding to the representative chamber Crp determined in step S23 is extracted. Through machine learning, the correlation between two different "a" data and "i" data contained in processing information PIrp is learned, and a representative model is generated. In step S25, the representative model is transmitted to the support device 4, and the process ends. 【0096】 Figure 8 is a flowchart showing an example of the process for comparing performance between chambers. Inter-chamber performance comparison processing This is a process performed by the CPU of the support device 4 when the CPU executes the inter-chamber performance comparison program stored in memory. 【0097】 Referring to Figure 8, the CPU of the support device 4 determines whether or not it has received a chamber analysis command (step S31). It remains in a waiting state until a chamber analysis command is received (NO in step S31), and if a chamber analysis command is received (YES in step S31), the process proceeds to step S32. 【0098】 In step S32, the representative model is received, and the process proceeds to step S33. The representative model is requested from the information analysis device 3, and the representative model transmitted from the information analysis device 3 is received. 【0099】 In step S33, the chamber Ck to be processed is selected from among multiple chambers C1 to Cn, and the process proceeds to step S34. Here, k is an integer from 1 to n. In step S34, the processing information PIk corresponding to the chamber Ck selected as the processing target in step S33 is selected from among multiple processing information PI1 to PIn. 【0100】 In the next step, S35, the degree of deviation between the chamber Ck selected for processing and the representative model is calculated. Specifically, for each of the "a" data to "i" data included in the processing information PIk corresponding to the chamber Ck selected for processing, a predicted value is obtained using the representative model, and the difference between the predicted value and the "a" data to "i" data is calculated as the degree of deviation. In the next step, S36, the anomaly score is calculated. In step S35, the sum of the multiple degree of deviations calculated for each of the "a" data to "i" data included in the processing information PIk is calculated as the anomaly score. 【0101】 In step S37, it is determined whether there are any chambers that were not selected as targets for processing in step S33. If there are unselected chambers, the process returns to step S33; otherwise, the process proceeds to step S38. 【0102】 In step S38, chambers with an anomaly score above the threshold are extracted, and processing proceeds to step S39. In step S39, support information is generated and output, and processing ends. The support information includes chamber identification information for identifying chambers with an anomaly score above the threshold, or a deviation degree table. The deviation degree table includes deviation degrees corresponding to the "a" data to "i" data included in the processing information calculated in step S35 for chambers with an anomaly score above the threshold. 【0103】 Maintenance workers viewing the support information can identify the chambers requiring maintenance from the chamber identification information included in the maintenance information. The anomaly score indicates the degree to which the chamber identified by the chamber identification information included in the maintenance information differs from the representative chamber in terms of processing operation and state. Anomalies range from those requiring the cessation of processing on the board to those requiring improvement without requiring the cessation of processing. Maintenance workers can identify the components to be maintained by referring to the deviation degree table included in the maintenance information. 【0104】 6. The first variation In the embodiment described above, the sum of multiple deviations shown in the deviation table in Figure 4 is calculated as the anomaly score. Among the multiple combinations of "a" data to "i" data included in the processing information shown in the deviation table in Figure 4, there may be combinations that should be given particular importance when determining the degree of anomaly in the substrate processing apparatus 1, and combinations that do not need to be given importance. In the first modified example, when determining the degree of anomaly in the substrate processing apparatus 1, the anomaly score is calculated by assigning a larger weight to the combinations of "a" data to "i" data included in the processing information, with the greater the importance. Specifically, multiple weighting tables corresponding to multiple combinations of multiple processing information are used. 【0105】 Figure 9 shows an example of a weighting table. In the weighting table, a weight corresponding to the relationship between the "a" data to "i" data included in the processing information is assigned to each combination in order to calculate the anomaly score. 【0106】 For example, the combination of "a. Chemical concentration" and "g. Concentration FB input amount" is a combination with a very high correlation in calculating an anomaly score to detect differences from the representative model. Therefore, it is assigned a weight of "6". The combination of "a. Chemical concentration" and any of "b. Chemical flow rate", "h. Input valve opening", and "i. Heater ON signal" is the next highest correlation in calculating an anomaly score to detect differences from the representative model. Therefore, it is assigned a weight of "5". The combination of "a. Chemical concentration" and any of "c. Room temperature" and "d. Chemical temperature" is a combination with a moderately high correlation in calculating an anomaly score to detect differences from the representative model. Therefore, it is assigned a weight of "2". Furthermore, the combination of "a. Chemical concentration" and any of "e. Heater outlet temperature" and "f. Heater surface temperature" is a combination with a low correlation in calculating an anomaly score to detect differences from the representative model. Therefore, it is assigned a weight of "1". 【0107】 In the first modification, the degree of deviation defined for each combination of "a" to "i" data included in the processing information, as defined in the deviation table shown in Figure 4, is changed based on the weights assigned to each combination of "a" to "i" data included in the processing information in the weighting table shown in Figure 9. Specifically, the degree of deviation defined for each combination of "a" to "i" data included in the processing information, as defined in the deviation table shown in Figure 4, is changed to a value (multiplicative value) obtained by multiplying it by the weights assigned to each combination of "a" to "i" data included in the processing information in the weighting table shown in Figure 9. Then, the sum of the multiple degree of deviations that have been changed considering the weights is calculated as the anomaly score. The anomaly score calculated in this way is 968. 【0108】 Thus, in the substrate processing apparatus management system 100 of the first modified example, the abnormality score is changed to a value weighted by the degree of deviation. As a result, maintenance workers of the substrate processing apparatus 1 can more appropriately grasp the degree of abnormality of each of the multiple chambers C1 to Cn provided by the substrate processing apparatus 1 by referring to the abnormality score calculated from the weighted degree of deviation. 【0109】 7. Second variation Figure 10 is a block diagram illustrating an example of the functional configuration of a substrate processing apparatus management system in a second modified example. The substrate processing apparatus management system 100 in the second modified example automatically determines a representative model. Referring to Figure 10, the differences from the block diagram shown in Figure 5 are that the representative information receiving unit 31 has been changed to an information receiving unit 31A, and a representative determination unit 37 has been added. The configuration of other functions is the same as the configuration shown in Figure 5, so it will not be explained again here. 【0110】 The information receiving unit 31A receives processing information PI1 to PIn corresponding to each of the multiple chambers C1 to Cn. The information receiving unit 31A outputs the processing information PI1 to PIn to the representative determination unit 37. 【0111】 The representative determination unit 37 determines a representative chamber from among multiple chambers C1 to Cn based on the processing information PI1 to PIn. Specifically, the average (average processing information) of multiple chambers C1 to Cn is calculated for each of the "a" data to "i" data included in the processing information PI1 to PIn. The representative determination unit 37 generates an average model that defines the correlation between two different combinations of the average of the "a" data and the average of the "b" data to the average of the "i" data by using machine learning on the average of the "a" data and the average of the "b" data to the average of the "i" data. 【0112】 Furthermore, the representative determination unit 37 selects the model that is closest to the average model from among the multiple chambers C1 to Cn as the representative model. Specifically, it uses the average model to calculate anomaly scores from each of the processing information PI1 to PIn. For example, regarding processing information PI1, the representative determination unit 37 calculates predicted values ​​for each of the "a" data to "i" data included in processing information PI1 using the average model, and calculates the deviation degree as the difference between the "a" data to "i" data and the predicted values. The representative determination unit 37 calculates the anomaly score as the sum of the deviation degrees calculated for multiple combinations of "a" data to "i" data included in the processing information. The representative determination unit 37 selects the chamber with the smallest anomaly score from among the multiple chambers C1 to Cn as the representative chamber. 【0113】 Figure 11 is a flowchart showing an example of the representative chamber determination process. The representative chamber determination process is performed by the CPU of the information analysis device 3 in step S23 of the model generation process shown in Figure 7, by executing a model generation program stored in memory. Referring to Figure 11, the CPU of the information analysis device 3 calculates average processing information from the processing information PI1 to PIn received in step S22 (step S51), and proceeds to step S52. Each of the processing information PI1 to PIn includes "a" data to "i" data. The average processing information is the average of multiple chambers C1 to Cn calculated for each of the "a" data to "i" data. 【0114】 In step S52, an average model is generated by machine learning the average processing information. Specifically, the correlation between two different sets of average processing information is machine-learned to generate an average model. In step S53, chamber Ck is selected from among multiple chambers C1 to Cn to be processed, and the process proceeds to step S54. Here, k is an integer from 1 to n. In step S54, processing information PIk corresponding to chamber Ck selected as the target for processing in step S53 is selected from among processing information PI1 to PIn. 【0115】 In the next step, S55, the degree of deviation between the chamber Ck selected for processing and the average model is calculated. Specifically, for each "a" data to "i" data included in the processing information PIk corresponding to the chamber Ck selected for processing, a predicted value is obtained using the average model, and the difference between the predicted value and the actual data is calculated as the degree of deviation. In the next step, S56, the anomaly score is calculated. The sum of the degree of deviations calculated for each "a" data to "i" data included in the processing information PIk in step S55 is calculated as the anomaly score. 【0116】 In step S57, it is determined whether there are any chambers that were not selected for processing in step S53. If there are unselected chambers, the process returns to step S53; otherwise, the process proceeds to step S58. In step S58, the chamber with the smallest anomaly score is determined to be the representative chamber, and the process returns to the representative model generation process. 【0117】 Substrate processing in the second modified example equipment management In this system, a representative chamber is automatically determined, eliminating the need for administrators or maintenance personnel to determine the representative chamber by referring to processing results. This allows for easy detection of differences between chambers. In particular, a representative chamber can be determined at a stage where the processing results of the substrate after processing cannot be obtained, or at a stage where the processing state of the substrate processing apparatus 1 after maintenance cannot be evaluated. Subsequently, at a stage where the processing results of the substrate after processing can be obtained, administrators or maintenance personnel can select a representative chamber by referring to the processing results. 【0118】 8. Third variation Substrate processing in the third modified example equipment management The system doubles the processing information of multiple processes and compares the correlation between the multiple processing information in each of the multiple chambers with the correlation between the multiple processing information in a representative model. Furthermore, the substrate processing in the third modified example equipment management The system performs logarithmic calculations on multiple pieces of information to determine the correlation between them. 【0119】 Figure 12 is a block diagram illustrating an example of the functional configuration of the substrate processing apparatus management system in the third modified example. Referring to Figure 12, the differences from the block diagram shown in Figure 5 are that the information analysis device 3 is equipped with a first doubling unit 51 and a first logarithmic calculation unit 53, and the support device 4 is equipped with a second doubling unit 61 and a second logarithmic calculation unit 63. The configuration of the other functions is the same as that shown in Figure 5, so we will not repeat the explanation here. 【0120】 The first doubling unit 51 receives processing information PIrp corresponding to the representative chamber Crp from the representative information receiving unit 31. The first doubling unit 51 performs a doubling process on the processing information PIrp. The doubling process is a process in which a predetermined reference information is combined with the multiple processing pieces as new processing information for at least one target processing piece among the multiple processing pieces. 【0121】 The first duplication unit 51 classifies the multiple processing information PIrp corresponding to the representative chamber Crp input from the representative information receiving unit 31 into a target group and a reference group. The reference group includes at least one of the multiple processing information PIrp. For example, one of the multiple processing information PIrp input from the representative information receiving unit 31 may be classified into the reference group, or multiple of the multiple processing information PIrp may be classified into the reference group. Multiple processing information PIrp that were not classified into the reference group from the multiple processing information PIrp input from the representative information receiving unit 31 are classified into the target group. 【0122】 The processing information PIrp includes "a" data to "i" data. The first doubling unit 51 generates processing information CPIrp by combining the "a" data to "i" data included in the processing information PIrp classified as a target group with at least one reference piece of the "a" data to "i" data included in the processing information PIrp classified as a reference group, as new processing information. 【0123】 The target processing information is one or more of the "a" data to "i" data included in the processing information PIrp that are subject to the doubling process. and It is predetermined. The target processing information is preferably the processing information from "a" data to "i" data included in the processing information PIrp that has no correlation or a small correlation with any other processing information. Hereinafter, the data included in the processing information PIrp classified into the reference group is referred to as "a ’ Let the data be "i'" data. 【0124】 In this embodiment, we will explain using the example where all of the "a" data to "i" data included in the processing information PIrp classified as a target group are the target processing information subject to the doubling process. In this case, the "a" data included in the processing information PIrp classified as a reference group ’ All of the data from "a" to "i'" is reference information. Therefore, the processing information CPIrp after doubling the processing information PIrp is "a" data, "a ’ "Data ~ "i" data, "i ’ This includes 18 types of processing information for the data. 【0125】 The first logarithmic calculation unit 53 receives the processing information CPIrp after the doubling process from the first doubling unit 51. The first logarithmic calculation unit 53 processes the "a" data, "a" included in the processing information CPIrp. ’ "Data ~ "i" data, "i ’ The logarithms obtained by logarithmically transforming each of the "a" data are output to the model generation unit 33. Note that the "a" data included in the processing information CPIrp, "a ’ "Data ~ "i" data, "i ’ Without logarithmically transforming all of the data, the "a" data, "a ’ "Data ~ "i" data, "i ’ Only the predetermined processing information from the data may be logarithmically transformed. 【0126】 The model generation unit 33 generates a representative model by machine learning multiple logarithms corresponding to the processing information CPIrp input from the first logarithmic calculation unit 53. 【0127】 The second doubling unit 61 receives multiple processing information PI1 to PIn corresponding to multiple chambers C1 to Cn from the information collection unit 41. The second doubling unit 61 performs doubling processing on each of the multiple processing information PI1 to PIn. Since the doubling process for each of the processing information PI1 to PIn is the same, the doubling process for processing information PI1 will be described here. 【0128】 The second duplication unit 61 combines a predetermined reference information with the "a" data to "i" data included in the processing information PI1 as new processing information for at least one target processing information from the "a" data to "i" data. The target processing information is one or more of the "a" data to "i" data included in the processing information PI1 that have been predetermined as targets for duplication processing. The target processing information is the same as the target processing information determined when the representative model was generated. The reference information is the processing information PIrp of the representative chamber Crp. The reference information may be, for example, the processing information PIrp of the representative chamber Crp classified into a reference group determined by the first duplication unit 51. 【0129】 In this embodiment, when generating a representative model, all "a" data to "i" data included in the processing information PIrp classified into the target group are treated as target processing information to be doubled. Therefore, the second doubled unit 61 treats all "a" data to "i" data included in the processing information PI1 as target processing information to be doubled. Furthermore, the second doubled unit 61 treats all "a" data to "i" data included in the processing information PIrp of the representative chamber Crp as target processing information. ’ All data from "data" to "i' data are used as reference information. 【0130】 The second doubling unit 61 combines the "a" data to "i" data contained in the processing information PI1 with the "a" data contained in the processing information PIrp. ’ The processing information CPI1 is generated by combining the "a" data and the "i'" data. The processing information CPI1 consists of the "a" data and the "a ’ "Data ~ "i" data, "i ’ This includes 18 types of processing information for the data. 【0131】 The second logarithmic calculation unit 63 receives the processing information CPI1 after the doubling process from the second doubling unit 61. The second logarithmic calculation unit 63 processes the "a" data, "a'" data, "i" data, and "i" data contained in the processing information CPI1. ’ The logarithms obtained by logarithmically transforming each of the data are output to the comparison unit 43. 【0132】 The comparison unit 43 compares the correlation between each of the processing information CPI1 to CPIn corresponding to each of the multiple chambers C1 to Cn with the correlation between the representative model and outputs the comparison result to the target determination unit 45. For example, let's describe the processing information CPI1 corresponding to chamber C1. The processing information CPI1 consists of "a" data, "a'" data to "i" data, and "i ’ The comparison unit 43 uses a representative model to process the data included in the processing information CPI1, including "a" data, "a'" data, "i" data, and "i ’ For each of the multiple logarithms corresponding to the data, the logarithm prediction The unit calculates a value and calculates the deviation degree as the difference between its logarithm and the corresponding predicted value. Furthermore, the comparison unit 43 calculates the anomaly score as the sum of the multiple deviation degrees calculated in accordance with the processing information CPI1. 【0133】 Figure 13 shows an example of a deviation degree table in the third modified example. The deviation degree table is a table that shows the deviation degree for all combinations of processing information. The support device 4 calculates the above deviation degree for all combinations of processing information. Referring to Figure 13, the processing information "a" data, "a ’ The multiple values ​​listed in the right-hand row of each of the "i" data and "i'" data are the processing information in the horizontal column above, "a" data and "a ’ This represents the degree of discrepancy between the processing information predicted from the "i" data and "i'" data and the processing information actually obtained. The processing information in the horizontal column above the degree of discrepancy table is the "a" data, "a ’ The multiple values ​​listed in the lower columns of each of the "i" data and "i'" data correspond to the processing information in the left vertical column, "a" data and "a’ The "data" represents the degree of discrepancy between the processing information predicted from the "i" data and the "i'" data, respectively, and the processing information actually obtained. 【0134】 Figure 14 is a flowchart showing an example of the model generation process in the third modified example. Referring to Figure 14, the difference from the model generation process shown in Figure 7 is that steps S61 and S62 are added between steps S22 and S23. The other processes are the same as those shown in Figure 7, so they will not be explained again here. 【0135】 In step S61, the doubling process is performed, and the process proceeds to step S62. The multiple processing information PIrp received in step S22 are classified into a target group and a reference group. Then, the "a" data to "i" data included in the processing information PIrp classified into the target group are added as new processing information to the "a" data included in the processing information PIrp classified into the reference group. ’ The "i'" data and the "i'" data are combined. 【0136】 In step S62, the data "a" included in the processing information CPIrp after the doubling process, "a ’ "Data ~ "i" data, "i ’ The data is logarithmically transformed, and the process proceeds to step S23. In step S23, the representative chamber Crp is determined, and the process proceeds to step S24. 【0137】 In step S24, multiple logarithms are machine-learned by logarithmically transforming the processing information CPIrp, which is obtained by doubling the processing information PIrp corresponding to the representative chamber Crp. The "a" data, "a ’ For the logarithms of the "" data, ~"i" data, and "i'" data, the correlation between two distinct sets of data is learned, and a representative model is generated. 【0138】 Figure 15 is a flowchart showing an example of the inter-chamber performance comparison process in the third modified example. Referring to Figure 15, the difference from the inter-chamber performance comparison process shown in Figure 8 is that steps S71 and S72 are added between steps S34 and S35. The other processes are the same as the model generation process shown in Figure 8, so they will not be explained again here. 【0139】 In step S71, the doubling process is performed, and the process proceeds to step S72. The processing information PIk of the chamber Ck selected in step S34 is doubled. The "a" data to "i" data contained in the processing information PIk are combined with reference information as new processing information. The reference information is the "a" data contained in the processing information PIrp of the representative chamber Crp. ’ The data is "i'" data. 【0140】 In step S72, the data "a" included in the processing information after the doubling process, "a ’ "Data ~ "i" data, "i ’ The data is logarithmically transformed, and the process proceeds to step S35. The processing information CPIk, which is obtained by doubling the processing information PIk corresponding to chamber Ck, is logarithmically transformed. 【0141】 In step S35, the degree of deviation between the chamber Ck selected for processing and the representative model is calculated. Specifically, the processing information PIk corresponding to the chamber Ck selected for processing is doubled, and the "a" data, "a" data included in the processing information CPIk after the double processing are calculated. ’ "Data", ~"i"Data, "i'"Data Taso For each logarithm, a predicted value is obtained using a representative model, and the difference between that logarithm and the predicted value is calculated as the degree of deviation. 【0142】 9. Effects of the Embodiment The support device 4 according to the above embodiment supports maintenance work on a substrate processing apparatus 1 equipped with a plurality of chambers C1 to Cn for processing a substrate W. The support device 4 acquires processing information PI1 to PIn indicating the operation or state related to the processing of the substrate W in each of the plurality of chambers C1 to Cn, and generates support information regarding maintenance work on other chambers based on comparison information obtained by comparing the correlation between "a" data to "i" data included in the processing information of a predetermined representative chamber Crp from among the plurality of chambers C1 to Cn with the correlation between "a" data to "i" data included in the processing information of other chambers. Since the correlation between "a" data to "i" data included in the processing information is compared between the representative chamber Crp and other chambers, it is possible to detect differences in operation or state related to the processing of the substrate W between the representative chamber Crp and other chambers. Furthermore, since support information regarding maintenance work on other chambers is generated based on comparison information obtained by comparing the correlation between "a" data to "i" data included in the processing information between the representative chamber Crp and other chambers, it is possible to generate information indicating the performance difference between other chambers and the representative chamber. As a result, it becomes possible to notify support information indicating the performance differences between the multiple chambers C1 to Cn provided by the substrate processing device 1. 【0143】 Furthermore, when the processing results of the substrate W are good in all of the multiple chambers C1 to Cn, it is preferable to adjust the multiple chambers C1 to Cn so that the performance of the chambers with poor performance approaches that of the chambers with good performance. In the support device 4, by selecting the chamber with good performance as the representative chamber Crp, chamber identification information and deviation degree tables for chambers with poorer performance than the representative chamber are generated as support information. Therefore, support information can be generated to bring the performance of chambers with poorer performance than the representative chamber closer to that of the representative chamber. 【0144】 Furthermore, in the support device 4 relating to the first modified example, the comparison information is information compared by weighting it according to the correlation between the "a" data to "i" data included in the processing information. Therefore, since the correlation between the "a" data to "i" data included in the processing information that indicates the operation or state related to the processing of the substrate W is weighted, it is possible to detect the difference with the representative chamber Crp of the operation or state that has a significant impact on the performance of processing the substrate W. 【0145】 Furthermore, in the support device 4 relating to the second modified example, one of the multiple chambers C1 to Cn having the closest correlation to the correlation between the averages of the "a" data to "i" data included in the processing information across the multiple chambers C1 to Cn is determined as the representative chamber. Therefore, since the chamber whose operation or state related to the processing of the substrate W is closest to the average is determined as the representative chamber Crp from among the multiple chambers C1 to Cn, the representative chamber Crp can be determined automatically. In particular, the representative chamber Crp can be determined even at a stage where processing results from inspecting the substrate W processed by the multiple chambers C1 to Cn have not yet been obtained. 【0146】 Furthermore, each of the multiple chambers C1 to Cn includes a chemical valve 113a for discharging the chemical solution and a flow meter 124 for measuring the flow rate of the liquid discharged from the chemical valve 113a. The "a" data to "i" data included in the processing information include "h. Inlet valve opening," which indicates the amount of opening input to the discharge valve to control the amount of opening of the chemical valve 113a, and "b. Chemical flow rate," which indicates the flow rate measured by the flow meter 124. Therefore, support information regarding maintenance work on the chemical valve 113a of other chambers can be generated from the difference in the correlation between "h. Inlet valve opening" and "b. Chemical flow rate" between the representative chamber Crp and other chambers. The support information includes a deviation degree table. Maintenance workers can look at the deviation degree table to determine whether "b. Chemical flow rate" is stable or not, and can also determine the operational delay of "h. Inlet valve opening." 【0147】 Furthermore, each of the multiple chambers C1 to Cn includes a chemical valve 113a for discharging the chemical solution and a concentration meter 122 for measuring the concentration of the chemical solution. The "a" data to "i" data included in the processing information include "h. Input valve opening," which indicates the amount of opening input to the discharge valve to control the amount of opening of the chemical valve 113a, and "a. Chemical solution concentration," which is measured by the concentration meter 122. Therefore, support information regarding maintenance work on other chambers can be generated from the difference in the correlation between "h. Input valve opening" and "a. Chemical solution concentration" between the representative chamber Crp and other chambers. The support information includes a deviation degree table. Maintenance workers can look at the deviation degree table to determine whether "a. Chemical solution concentration" is stable or not, and can also determine the operational delay of "h. Input valve opening." 【0148】 Furthermore, each of the multiple chambers C1 to Cn includes a concentration meter 122 for measuring the concentration of the chemical solution and a first thermometer 123 for measuring the temperature inside the chamber. The "a" data to "i" data included in the processing information include "a. Chemical solution concentration" and "c. Chamber temperature". Therefore, support information regarding maintenance work for other chambers is generated from the difference in the correlation between the chemical solution concentration and the chamber temperature between the representative chamber Crp and the other chambers. Ruko Maintenance workers can look at the deviation table and determine whether "a. chemical concentration" and "c. room temperature" are stable or not. 【0149】 Furthermore, each of the multiple chambers C1 to Cn includes a concentration meter 122 for measuring the concentration of the chemical solution and a second thermometer 123a for measuring the temperature of the chemical solution. The "a" data to "i" data included in the processing information include "a. Chemical solution concentration" and "d. Chemical solution temperature". Therefore, support information regarding maintenance work on other chambers is generated from the differences in the correlation between the chemical solution concentration and the chemical solution temperature between the representative chamber Crp and the other chambers. Ruko This allows maintenance workers to determine whether "a. Chemical concentration" and "d. Chemical temperature" are stable by looking at the deviation table. 【0150】 Furthermore, the support device 4 acquires a deviation table as comparison information, which includes a deviation degree indicating the degree of deviation between the predicted values ​​predicted from the processing information in other chambers and the processing information in other chambers, using the correlation between the "a" data to "i" data contained in the processing information in the representative chamber Crp. Therefore, since the predicted values ​​for each of the "a" data to "i" data contained in the processing information in other chambers are predicted using the correlation between the "a" data to "i" data contained in the processing information in the representative chamber Crp, the correlation between the "a" data to "i" data contained in the processing information in the representative chamber Crp can be compared between the representative chamber Crp and the other chambers. A deviation table including a deviation degree indicating the degree of deviation between multiple predicted values ​​and multiple "a" data to "i" data contained in the processing information in other chambers is acquired as comparison information, so the correlation between the processing information in the representative chamber Crp and the other chambers can be compared. Report This allows for the quantitative demonstration of the differences in correlation between the included "a" data and "i" data. 【0151】 In the third modification, a representative model is generated that shows the correlation between multiple processing information in a representative chamber, and the correlation between multiple processing information in each of the multiple chambers is compared with the correlation between multiple processing information in the representative model. Therefore, it is possible to combine reference information that is correlated with processing information that is not correlated with other processing information or has a small correlation with other processing information. Thus, it is possible to provide information on the performance difference with the representative chamber with respect to processing information that is not correlated with other processing information or has a small correlation with other processing information. 【0152】 Furthermore, in the third modification, logarithmic calculation is performed on at least one of the multiple pieces of processing information, and then the correlation between the multiple pieces of processing information is calculated. This makes it possible to improve the accuracy of the correlation between processing information that changes little and other pieces of processing information. 【0153】 10. Other Embodiments (1) In the above embodiment, the substrate processing apparatus 1 is a substrate cleaning apparatus, but the present invention is not limited thereto. The substrate processing apparatus 1 may be a single-wafer type substrate cleaning apparatus rather than a batch type, as long as it has a plurality of chambers, and it may have a configuration that performs processing other than cleaning. Furthermore, the substrate processing apparatus 1 may be an apparatus that performs wet etching on the substrate W, or an apparatus that performs flash lamp annealing (FLA) processing which heats the surface of the substrate W with a flash of light. 【0154】 (2) 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 quantities 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. 【0155】 (3) In the above embodiment, the support device 4 includes a support unit 49, but the present invention is not limited thereto. For example, support information may be transmitted from the support device 4 to the substrate processing device 1, and the support information may be output by being displayed in the substrate processing device 1. In this case, maintenance workers will be able to view the support information while maintaining the substrate processing device 1. 【0156】 (4) 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 support 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. 【0157】 (5) In the above embodiment, the substrate processing apparatus 1 and the support apparatus 4 are provided separately, but the substrate processing apparatus 1 and the support apparatus 4 may be provided as an integrated unit. Also, in the above embodiment, the information analysis apparatus 3 and the support apparatus 4 are provided separately, but the information analysis apparatus 3 and the support apparatus 4 may be provided as an integrated unit. 【0158】 (6) In the support 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 support device 4, a representative model or an average model may be generated by using other machine learning methods such as deep learning. 【0159】 (7) In the above embodiment, processing information PI1 to PIn is transmitted from the processing information transmission unit 13 of the substrate processing apparatus 1 to the information collection unit 41 of the support device 4, but the present invention is not limited thereto. For example, processing information PI1 to PIn may be transmitted via a cloud on the internet or the like. In this case, the substrate processing apparatus 1 and the support device 4 do not need to communicate directly, so the communication load can be reduced. 【0160】 (8) In the above embodiment, the support device 4 includes a model receiving unit 47 and a comparison unit 43, but the present invention is not limited thereto. For example, the support device 4 may transmit each of the processing information PI1 to PIn collected by the information collection unit 41 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 support device 4. 【0161】 (9) The first modified example may be combined with the third modified example. 【0162】 (10) In the third modified example, a first doubling unit 51 and a first logarithmic calculation unit 53 are added to the information analysis device 3, and a second doubling unit 61 and a second logarithmic calculation unit 63 are added to the support device 4. 【0163】 Alternatively, a first doubling unit 51 may be added to the information analysis device 3, and a second doubling unit 61 may be added to the support device 4. In this case, the first logarithmic calculation unit 53 and the second logarithmic calculation unit 63 are unnecessary. The model generation unit 33 receives processing information CPIrp, which is obtained by doubling the processing information PIrp of the representative chamber Crp by the first doubling unit 51, and the comparison unit 43 receives processing information CPIk, which is obtained by doubling the processing information PIk of the chamber Ck by the second doubling unit 61. In this case, reference information that is correlated with processing information that is not correlated with other processing information or has a small correlation with other processing information can be combined. Therefore, information regarding the performance difference with the representative chamber can be provided with respect to processing information that is not correlated with other processing information or has a small correlation with other processing information. 【0164】 Furthermore, a first doubling unit 51 may be added to the information analysis device 3, and a second doubling unit 61 may be added to the support device 4, and these may be combined with the first modified example. 【0165】 Alternatively, a first logarithmic calculation unit 53 may be added to the information analysis device 3, and a second logarithmic calculation unit 63 may be added to the support device 4. In this case, the first doubling unit 51 and the second doubling unit 61 are unnecessary. The logarithm obtained by logarithmically transforming the processing information PIrp of the representative chamber Crp from the first logarithmic calculation unit 53 is input to the model generation unit 33, and the logarithm obtained by logarithmically transforming the processing information PIk of the chamber Ck from the second logarithmic calculation unit 63 is input to the comparison unit 43. In this case, the accuracy of the correlation between processing information with other processing information can be improved for processing information with small changes. 【0166】 Furthermore, a first logarithmic calculation unit 53 may be added to the information analysis device 3, and a second logarithmic calculation unit 63 may be added to the support device 4, and these may be combined with the first modified example. 【0167】 11. 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 component of the embodiment. In the above embodiment, the information collection unit 41 is an example of a processing information collection unit, the support unit 49 is an example of a support information generation unit, and the comparison unit 43 is an example of a deviation degree acquisition unit. Also, the representative determination unit 37 is an example of a representative chamber determination unit. Furthermore, the chemical solution valve 113a is an example of a discharge valve, the flow meter 127 is an example of a flow rate measurement unit, the concentration meter 122 is an example of a concentration measurement unit, the first thermometer 123 is an example of an indoor temperature measurement unit, and the second thermometer 123a is an example of a liquid temperature measurement unit. [Explanation of symbols] 【0168】 1 Substrate processing device, 3 Information analysis device, 4 Support device, 10 Control device, 11 Processing information acquisition unit, 13 Processing information transmission unit, 31 Representative information receiving unit, 31A Information receiving unit, 33 Model generation unit, 35 Model transmission unit, 37 Representative determination unit, 41 Information collection unit, 43 Comparison unit, 45 Target determination unit, 47 Model receiving unit, 49 Support unit, 100 Substrate processing device management system, 111 Processing tank, 112 Substrate holding unit, 112a Lifting device, 113 Chemical piping, 113a Chemical valve, 114 Pure water piping, 114a Pure water valve, 115 Bubbler tube, 116 Gas piping, 116a Gas valve, 117 Heater, 118 Heater drive unit, 118a Heater thermometer, 118b Outlet thermometer, 121 Liquid level gauge, 122 Concentration meter, 123 1st thermometer, 123a 2nd thermometer, 124,125,127 Flow meter.

Claims

[Claim 1] A support device for assisting maintenance work on a substrate processing apparatus equipped with multiple chambers for processing substrates, A processing information collection unit that acquires a plurality of processing information indicating operations or states related to the processing of substrates in each of the plurality of chambers, The system includes a support information generation unit that generates support information regarding maintenance work for the other chambers based on comparison information obtained by comparing the correlation between the multiple processing information in a predetermined representative chamber among the multiple chambers with the correlation between the multiple processing information in the other chambers, The support information generation unit performs a doubling process on at least one of the multiple processing pieces of processing information, combining it with predetermined reference information as new processing information, and then calculates the correlation between the multiple processing pieces of processing information after the doubling process. [Claim 2] The support device according to claim 1, wherein the comparison information is information obtained by weighting and comparing the correlations between the plurality of processing information. [Claim 3] A support device for assisting maintenance work on a substrate processing apparatus having a plurality of chambers for processing substrates, A processing information collection unit that acquires a plurality of processing information indicating operations or states related to the processing of substrates in each of the plurality of chambers, The system includes a support information generation unit that generates support information regarding maintenance work for the other chambers based on comparison information obtained by comparing the correlation between the plurality of processing information collected while a predetermined representative chamber processes a substrate with the correlation between the plurality of processing information in the other chambers, among the plurality of chambers, The support information generation unit performs logarithmic calculation on at least one of the multiple processing pieces of information and then calculates the correlation between the multiple processing pieces of information. The support information generation unit performs a doubling process on at least one of the multiple processing pieces of processing information, combining it with predetermined reference information as new processing information, and then calculates the correlation between the multiple processing pieces of processing information after the doubling process. [Claim 4] The support device according to claim 3, wherein the comparison information is information obtained by comparing it with weights according to the correlation between the plurality of processing information. [Claim 5] The support device according to any one of claims 3 to 4, further comprising a representative chamber determination unit that determines one of the plurality of chambers having the correlation closest to the correlation between the averages of the plurality of processing information of each of the plurality of chambers as the representative chamber. [Claim 6] Each of the aforementioned plurality of chambers is equipped with a discharge valve for discharging liquid, It includes a flow rate measuring unit for measuring the flow rate of liquid discharged from the discharge valve, The support device according to any one of claims 3 to 4, wherein each of the plurality of processing information includes an open signal indicating the amount of opening input to the discharge valve in order to control the amount of opening of the discharge valve, and a flow rate measured by the flow rate measuring unit. [Claim 7] Each of the aforementioned plurality of chambers is equipped with a discharge valve for discharging liquid, It includes a concentration measuring unit for measuring the concentration of a liquid, Each of the plurality of processing information includes an open signal indicating the amount of opening input to the discharge valve in order to control the amount of opening of the discharge valve, and a concentration measured by the concentration measuring unit, the support device according to any one of claims 1 to 4. [Claim 8] Each of the aforementioned plurality of chambers includes a concentration measuring unit for measuring the concentration of a liquid, It includes an indoor temperature measuring unit that measures the temperature of the chamber, The support device according to any one of claims 1 to 4, wherein each of the plurality of processing information includes a concentration measured by the concentration measuring unit and a temperature measured by the room temperature measuring unit. [Claim 9] Each of the aforementioned plurality of chambers includes a concentration measuring unit for measuring the concentration of a liquid, It includes a liquid temperature measuring unit for measuring the temperature of a liquid, The support device according to any one of claims 1 to 4, wherein each of the plurality of processing information includes a concentration measured by the concentration measuring unit and a temperature measured by the liquid temperature measuring unit. [Claim 10] The support device according to any one of claims 1 to 4, further comprising a deviation degree acquisition unit that acquires deviation degree information as comparison information, which indicates the degree of deviation between a plurality of predicted values ​​predicted from the plurality of processing information in the other chambers, using the correlation between the plurality of processing information in the representative chamber. [Claim 11] A substrate processing apparatus comprising the support device described in any one of claims 1 to 4. [Claim 12] A method for comparing the performance of inter-chambers, performed by an auxiliary device that assists in maintenance work on a substrate processing apparatus having multiple chambers for processing substrates, A process for acquiring a plurality of processing information indicating an operation or state related to the processing of the substrate in each of the plurality of chambers, The process includes generating support information for maintenance work in the other chambers based on comparison information obtained by comparing the correlation between the multiple processing information in a predetermined representative chamber among the multiple chambers with the correlation between the multiple processing information in the other chambers, A chamber performance comparison method comprising the process of generating the support information, which includes performing a doubling process to combine predetermined reference information as new processing information with at least one target processing information among the multiple processing information, and then calculating the correlation between the multiple processing information after the doubling process. [Claim 13] A chamber-to-chamber performance comparison program, which is executed on a computer in a support device that assists in maintenance work for a substrate processing apparatus having multiple chambers for processing substrates, A process for acquiring a plurality of processing information indicating an operation or state related to the processing of the substrate in each of the plurality of chambers, The computer is instructed to perform a process that generates support information regarding maintenance work for the other chambers based on comparison information obtained by comparing the correlation between the multiple processing information in a predetermined representative chamber among the multiple chambers with the correlation between the multiple processing information in the other chambers. The process for generating the support information is a chamber-to-chamber performance comparison program that includes performing a doubling process on at least one of the multiple processing pieces of processing information, combining predetermined reference information as new processing information with the multiple processing pieces of processing information, and then calculating the correlation between the multiple processing pieces of processing information after the doubling process.

Images