Cabinet fault detection method and device, computer equipment and storage medium

Abstract

The invention relates to a machine fault detection method and device, computer equipment and a storage medium, and the method comprises the steps: determining the wafer data of a target wafer, and determining the wafer grouping information of the target wafer according to the wafer data; acquiring process step information and machine equipment information of the target wafer, and grouping the machine equipment information of the target wafer according to process steps to form machine grouping information; according to the machine grouping information and the wafer grouping information, determining a process step having correlation with wafer abnormity; and determining a fault machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer. According to the scheme, automatic fault detection of the machine equipment can be realized, the fault detection cost of the machine equipment is saved, the fault detection efficiency of the machine equipment is improved, and thus the yield of produced chips can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of mechanical equipment detection, and in particular to a machine fault detection method, device, computer equipment and storage medium. Background Art

[0002] The process steps involved in the semiconductor manufacturing process are relatively complicated, and there are many machines involved. When low yield or product defects occur in the manufacturing process, it is necessary to find the problematic machine. When the existing technology performs fault detection on the machine, it often relies on the experience of engineers to unravel the huge amount of data and analyze them one by one, which takes a lot of time. Faced with a large number of machines and complicated chip manufacturing process steps, it is difficult for engineers to locate the faulty machine in a short time, which affects the chip manufacturing efficiency and increases the chip manufacturing cost. Therefore, how to improve the efficiency of locating faulty machines, reduce the cost of troubleshooting faulty machines, and thus improve chip manufacturing efficiency, is a problem that needs to be solved. Summary of the invention

[0003] Based on this, it is necessary to provide a machine fault detection method, device, computer equipment and storage medium that can improve the efficiency of locating faulty machines and reduce the cost of troubleshooting faulty machines in response to the above technical problems.

[0004] In a first aspect, the present application provides a machine fault detection method, the method comprising:

[0005] Determine wafer data of a target wafer, and determine wafer grouping information of the target wafer according to the wafer data;

[0006] Acquire process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0007] Determine a process step that is relevant to wafer abnormality according to the machine grouping information and the wafer grouping information;

[0008] A faulty machine among the machine equipment is determined according to the machine equipment information associated with the process step and the wafer data of the target wafer.

[0009] In one embodiment, the step of determining wafer data of a target wafer and determining wafer grouping information of the target wafer according to the wafer data includes:

[0010] Select a data type and determine a target wafer, and obtain wafer data of the target wafer under the data type;

[0011] A data threshold for characterizing wafer abnormality under the data type is determined based on the wafer data, and wafers are grouped according to the data threshold to form wafer grouping information.

[0012] In one of the embodiments, performing wafer grouping based on the wafer data and a data threshold representing wafer abnormality under the data type to form wafer grouping information includes:

[0013] Forming preset statistical data according to the wafer data, and determining a data threshold for characterizing wafer abnormality under the data type according to the statistical data;

[0014] The statistical data are grouped according to the data threshold to obtain at least two wafer groups, forming wafer grouping information.

[0015] In one embodiment, determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data, includes:

[0016] Determine wafer data of a target wafer, and determine a data dimension of the target wafer according to the wafer data;

[0017] Wafers are grouped according to the data dimension.

[0018] In one embodiment, the step of obtaining the process step information and machine equipment information of the target wafer includes:

[0019] Obtaining a selected wafer data type, and selecting process step information before generating wafer data of the wafer data type;

[0020] Based on the process step information, obtain the machine equipment information of each process step.

[0021] In one embodiment, the step of determining the process steps correlated with wafer abnormality according to the machine grouping information and the wafer grouping information includes:

[0022] Determine a two-dimensional observation table corresponding to each of the process steps according to the machine grouping information and the wafer grouping information, and determine a chi-square statistic and a degree of freedom corresponding to the process step based on the two-dimensional observation table;

[0023] A process step correlated with wafer abnormality is determined based on the chi-square statistic and the degrees of freedom.

[0024] In one embodiment, determining a faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer includes:

[0025] Obtain wafer data of a target wafer associated with the process step and / or statistical data of the wafer data, and group the wafer data according to the machine equipment information and wafer grouping information to form at least one type of grouping data;

[0026] At least one visualization chart is formed for the at least one grouping data according to at least one data dimension to determine a faulty machine among the machine equipment.

[0027] In one embodiment, determining a faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer includes:

[0028] Acquiring wafer data of the target wafer associated with the process step;

[0029] Generate statistical data of wafer data of the target wafer passing through each machine device based on the machine device information;

[0030] The statistical data are grouped according to the machine equipment information and the wafer grouping information to form grouped data, and the faulty machine among the machine equipment is determined based on the grouped data.

[0031] In one embodiment, determining a faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer includes:

[0032] Acquiring wafer data of the target wafer associated with the process step;

[0033] Grouping the wafer data according to the machine equipment information and the wafer grouping information to form at least one type of grouped data;

[0034] At least one visualization chart is formed for the at least one grouping data according to at least one data dimension to determine a faulty machine among the machine equipment.

[0035] In one embodiment, determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data, includes:

[0036] Reading sample wafer data of a sample wafer and sample grouping information corresponding to the sample wafer chip from a database;

[0037] Training a neural network model according to the sample wafer data and the sample grouping information to determine a wafer grouping model;

[0038] Determining wafer data of the target wafer according to wafer production data of the target wafer;

[0039] The wafer data is input into the wafer grouping model, and the wafer grouping information of the target wafer is determined according to the output data of the wafer grouping model.

[0040] In a second aspect, the present application further provides a machine fault detection device, the machine fault detection device comprising:

[0041] A wafer grouping information determination module, used to determine wafer data of a target wafer, and determine wafer grouping information of the target wafer according to the wafer data;

[0042] A machine grouping information determination module is used to obtain process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0043] A correlation risk model, used to determine the process steps that are correlated with wafer abnormalities according to the machine grouping information and the wafer grouping information;

[0044] The faulty machine determining module is used to determine the faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer.

[0045] In a third aspect, the present application further provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0046] Determine wafer data of a target wafer, and determine wafer grouping information of the target wafer according to the wafer data;

[0047] Acquire process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0048] Determine a process step that is relevant to wafer abnormality according to the machine grouping information and the wafer grouping information;

[0049] A faulty machine among the machine equipment is determined according to the machine equipment information associated with the process step and the wafer data of the target wafer.

[0050] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0051] Determine wafer data of a target wafer, and determine wafer grouping information of the target wafer according to the wafer data;

[0052] Acquire process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0053] Determine a process step that is relevant to wafer abnormality according to the machine grouping information and the wafer grouping information;

[0054] A faulty machine among the machine equipment is determined according to the machine equipment information associated with the process step and the wafer data of the target wafer.

[0055] The above-mentioned machine fault detection method, device, computer equipment and storage medium determine the wafer data of the target wafer, determine the wafer grouping information of the target wafer according to the wafer data; obtain the process step information and machine equipment information of the target wafer, group the machine equipment information of the target wafer according to the process step to form machine grouping information; determine the process step related to the wafer abnormality according to the machine grouping information and wafer grouping information; determine the faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer. This solves the problem that when performing fault detection on machine equipment, it is necessary to rely on the experience of engineers to analyze one by one in the huge amount of data, resulting in low detection efficiency of the machine, and it is impossible to quickly and accurately locate the faulty machine, which affects the chip manufacturing efficiency, and the cost of troubleshooting the machine equipment is high.

[0056] The above scheme groups the target wafers according to the wafer data, groups the machines and equipment for producing chips based on the process steps, and based on the grouping results of the target wafers and the grouping results of the machines and equipment, determines the process steps that may cause the failure of the machines and equipment through the machine grouping information and the wafer grouping information, and locates the faulty machines in the machines and equipment according to the process steps, which can realize automatic fault detection of the machines and equipment, save the fault detection cost of the machines and equipment, and improve the fault detection efficiency of the machines and equipment, thereby improving the yield of the produced chips. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is an application environment diagram of a machine fault detection method in an embodiment;

[0058] Figure 2 A schematic diagram of a process flow of a machine fault detection method in one embodiment;

[0059] Figure 3 A schematic diagram of a flow chart of a machine fault detection method in another embodiment;

[0060] Figure 4 A schematic diagram of a flow chart of a machine fault detection method in another embodiment;

[0061] Figure 5 is a structural block diagram of a machine fault detection device in one embodiment;

[0062] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0064] The machine fault detection method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The server 104 determines the wafer data of the target wafer based on the target wafer data, and determines the wafer grouping information of the target wafer according to the wafer data; obtains the process step information and machine equipment information of the target wafer, and groups the machine equipment information of the target wafer according to the process step to form machine grouping information; determines the process step related to the wafer abnormality according to the machine grouping information and the wafer grouping information; determines the faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer, and sends the machine identification of the faulty machine to the terminal 102 through the communication network. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices, and the Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. The portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0065] In one embodiment, Figure 2 As shown, a machine fault detection method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server.

[0066] In this embodiment, the method includes the following steps:

[0067] S210 , determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data.

[0068] Among them, the wafer data can be any one or more of the defect data of the target wafer, electrical wafer data measurement data and stability wafer data, and yield data. The type and quantity of the target wafer can be selected according to actual needs. Defect data can include the type of defects existing in the target wafer and the number of defects existing in the target wafer; electrical wafer data can be obtained by probing each die on the wafer, such as CP (Chip Probing) or WAT (Wafer Acceptance Test) and other wafer-level wafer data; wafer measurement refers to the process of performing various parameter tests and precision calibration on wafer products, and measurement data refers to the data collected and stored when performing parameter tests and precision calibration on the wafer; stability wafer data, namely SP data, can reflect the stability and overclocking capability of the wafer.

[0069] Specifically, production data of the target wafer is obtained, wafer data of the target wafer is obtained according to the production data of the target wafer, the target wafer is grouped according to the target wafer data, the target wafer is divided into at least two groups, and the target wafer groups are determined. The target wafer groups are sorted to determine wafer grouping information of the target wafer.

[0070] Exemplarily, assuming that the target wafer can be divided into an excellent group and an inferior group, the user can determine the grouping guidance data from the wafer data of the target wafer according to actual needs, and group the target wafer according to the grouping guidance data. The data type of the grouping guidance data is the data type determined by the user according to actual needs for correlation analysis with the machine. If the grouping guidance data is CP data, the target wafer with a yield value less than a preset yield threshold can be divided into an inferior group, and the target wafer with a yield value greater than or equal to the yield threshold can be divided into an excellent group. If the grouping guidance data is defect data, the target wafer with a relevant statistical value such as defect type / number greater than a preset defect critical value can be divided into an inferior group, and the target wafer with a defect value less than or equal to the defect critical value can be divided into an excellent group. If the grouping guidance data is electrical wafer data, the data range of the expected electrical data can be pre-set, and if the electrical wafer data falls within the data range of the expected electrical data, the target wafer is divided into an excellent group, and if the electrical wafer data does not fall within the data range of the expected electrical data, the target wafer is divided into an inferior group. If the grouping guidance data is measurement data, the data range of the expected measurement data can be preset. If the measurement data falls within the data range of the expected measurement data, the target wafer is divided into the superior group. If the measurement data does not fall within the data range of the expected measurement data, the target wafer is divided into the inferior group. If the grouping guidance data is stability wafer data, the target wafer whose stability wafer data is greater than a preset stability threshold can be divided into the superior group, and the target wafer whose stability wafer data is less than or equal to the stability threshold can be divided into the inferior group.

[0071] Optionally, the user can also group the target wafers according to other dimensions such as chip production batch or chip category based on actual needs.

[0072] S220 , obtaining process step information and machine equipment information of the target wafer, and grouping the machine equipment information of the target wafer according to the process steps to form machine grouping information.

[0073] The process step information of the target wafer refers to the production process steps required or passed in the production process of the target wafer. A process step information may be completed by a machine independently or by two or more machines in collaboration. The machine equipment information may include the equipment identification of the machine equipment, and the machine grouping information includes the equipment identification of the machine equipment and the group corresponding to the machine equipment.

[0074] Specifically, the production process of the target wafer is determined, and a visualization list is generated according to the production process of the target wafer. According to the process step selection information sent by the user, the process step information that needs to be paid attention to is selected from the visualization list. The selected process step information can be one, two or more. The relevant information of the machine equipment can be imported into the database in advance through the manufacturing execution system or other data import system. The relevant information of the machine equipment can include the equipment identification of the machine equipment, the chip manufacturing process flow that the machine equipment is responsible for, and the chip type that the machine equipment is responsible for producing and processing. The machine equipment involved in the execution process of each process step can be read from the database separately, so as to group the machine equipment based on the process step information and the machine equipment information, that is, the machine equipment that performs the same process step is divided into a group, and the grouping results are integrated to determine the machine grouping data. The machine grouping information can be a table that records the machine equipment identification and the corresponding group of the machine equipment obtained after the grouping results are integrated.

[0075] S230 , determining a process step that is correlated with wafer abnormality according to the machine grouping information and the wafer grouping information.

[0076] Among them, wafer abnormality refers to a reference item for judging whether the wafer data is normal / reasonable under each wafer data type, which can be a statistical value of wafer data under each wafer data type, such as the yield of CP data, the number of defects in defect data, and other statistical values.

[0077] S240 , determining a faulty machine among the machines according to the machine equipment information associated with the process steps and the wafer data of the target wafer.

[0078] In the above machine fault detection method, the wafer data of the target wafer is determined, and the wafer grouping information of the target wafer is determined based on the wafer data; the process step information and machine equipment information of the target wafer are obtained, and the machine equipment information of the target wafer is grouped according to the process steps to form machine grouping information; the process steps that are relevant to the wafer abnormality are determined based on the machine grouping information and the wafer grouping information; the faulty machine in the machine equipment is determined based on the machine equipment information associated with the process steps and the wafer data of the target wafer. This solves the problem that when performing fault detection on machine equipment, it is necessary to rely on the experience of engineers to analyze one by one in the huge amount of data, resulting in low detection efficiency of the machine, and the inability to quickly and accurately locate the faulty machine, which affects the chip manufacturing efficiency, and the high cost of troubleshooting the machine equipment. The above scheme groups the target wafers according to the wafer data, groups the machines and equipment for producing chips based on the process steps, and based on the grouping results of the target wafers and the grouping results of the machines and equipment, determines the process steps that may cause the failure of the machines and equipment through the machine grouping information and the wafer grouping information, and locates the faulty machines in the machines and equipment according to the process steps, which can realize automatic fault detection of the machines and equipment, save the fault detection cost of the machines and equipment, and improve the fault detection efficiency of the machines and equipment, thereby improving the yield of the produced chips.

[0079] In one embodiment, determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data, includes:

[0080] Select a data type and determine a target wafer, and obtain wafer data of the target wafer under the data type; determine a data threshold that characterizes wafer abnormality under the data type based on the wafer data, group the wafers according to the data threshold, and form wafer grouping information.

[0081] Among them, the data type is the data type of the wafer, which can be determined according to the wafer data such as CP, WAT or defect obtained when testing the wafer. Each type of wafer data will have a corresponding grouping judgment standard. For example, if the wafer data is CP data, it is generally grouped according to the yield. If the wafer data is the electrical test value, it can be grouped according to high or low or qualified intervals. The wafers with electrical test values ​​within the preset interval can be set as qualified. If the wafer data is defec data, it can be grouped according to statistics such as the number and density of defects. The core idea is to group according to the good or bad standards under different data types. Of course, other dimensions can also be viewed at the same time. For example, if two batches of wafer data are selected, if the wafers in one batch perform better, the entire batch may be grouped as a group. The main grouping is based on the good or bad under the wafer dimension.

[0082] The above solution provides a method for grouping target wafers, which can achieve flexibility in grouping target wafers.

[0083] In one embodiment, wafer grouping is performed based on wafer data and a data threshold that characterizes wafer abnormality under a data type to form wafer grouping information, including:

[0084] Preset statistical data is formed according to the wafer data, and a data threshold for characterizing wafer abnormality under the data type is determined according to the statistical data; the statistical data is grouped according to the data threshold to obtain at least two wafer groups to form wafer grouping information.

[0085] Among them, statistical data refers to the statistical value of wafer data / test results. For example, during the test process, an electrical test is performed on a bare die, that is, a small piece of the wafer. There are multiple test values ​​for one wafer. This solution performs analysis based on a wafer dimension, so multiple test values ​​on the entire wafer will be counted, such as generating an average value to represent this wafer.

[0086] The above solution provides another method for grouping target wafers, which can achieve reliability in grouping target wafers.

[0087] In one embodiment, determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data, includes:

[0088] Determine wafer data of a target wafer, determine data dimensions of the target wafer according to the wafer data; and group the wafers according to the data dimensions.

[0089] Determine wafer data of a target wafer, determine data dimensions of the target wafer according to the wafer data, determine production batch information of the target wafer according to the data dimensions, group the target wafers according to the production batch information of the target wafer, and determine wafer grouping information of the target wafer.

[0090] The above scheme groups the target wafers based on the wafer data and / or statistical data corresponding to the target wafers, which can improve the accuracy of the target wafer grouping. Selecting a data dimension for measuring the group to which the target wafer belongs from the list of wafer data and / or statistical data corresponding to the target wafer, and grouping the target wafers based on the selected dimension can further improve the flexibility of the target wafer grouping.

[0091] In one embodiment, obtaining process step information and machine equipment information of the target wafer includes:

[0092] The selected wafer data type is obtained, and process step information before the wafer data of the wafer data type is generated is selected; based on the process step information, the machine equipment information of each process step is obtained.

[0093] It should be noted that when selecting different types of wafer data to judge the quality of the target wafer, the process step information before the judgment data needs to be used as the judgment standard, so as to ensure that the selected process step information has an impact on the judgment data. For example, if there are 100 process steps, and the generation link of the selected wafer data type is the 80th step, then when performing correlation analysis, only the process step information of steps 1 to 79 will be selected for analysis.

[0094] The above solution can improve the completeness and accuracy of the acquired machine equipment information.

[0095] In one embodiment, Figure 3 As shown, the process steps related to wafer abnormality are determined according to the machine grouping information and the wafer grouping information, including:

[0096] S310, determining a two-dimensional observation table corresponding to each of the process steps according to the machine grouping information and the wafer grouping information, and determining a chi-square statistic and a degree of freedom corresponding to the process step based on the two-dimensional observation table.

[0097] The header row data of the two-dimensional observation table are the groups into which the target wafers are divided, and the header column data of the two-dimensional observation table represent the equipment identification of the machine equipment under the corresponding process step. The chi-square statistic is used to measure the difference between the observed data and the expected data. The chi-square statistic is calculated by squaring the difference between each observed value and the corresponding expected value, dividing it by the expected value, and then adding up these values. The larger the chi-square statistic, the greater the difference between the observed data and the expected data. Degrees of freedom refer to the number of parameters that can be freely changed when performing a chi-square test.

[0098] Specifically, the observation frequencies of the equipment corresponding to the process steps in different groups are determined according to the equipment grouping data and wafer grouping information, and the two-dimensional observation table corresponding to the process steps is generated according to the observation frequencies of the equipment corresponding to the process steps in different groups. The chi-square statistic and degree of freedom corresponding to each process step are calculated according to the two-dimensional observation table corresponding to each process step.

[0099] Exemplarily, a method for determining the chi-square statistic and degrees of freedom corresponding to a process step based on a two-dimensional observation table may be: the calculation formula for the expected observation frequency of the i-th row and j-th column in the two-dimensional observation table is shown in formula (1):

[0100]

[0101] Among them, E ij is the expected observation frequency in the i-th row and j-th column of the two-dimensional observation table, n i is the sum of the observation frequencies of the i-th row in the two-dimensional observation table, mj is the sum of the observation frequencies in the jth column of the two-dimensional observation table, and N is the sum of all observation frequencies in the two-dimensional observation table.

[0102] For example, in Table 1, n 1 is the sum of the observation frequencies in the first row of the two-dimensional observation table, that is, a+b, n 2 is the sum of the observation frequencies in the second row of the two-dimensional observation table, that is, c+d; m 1 is the sum of the observation frequencies in the first column of the two-dimensional observation table, that is, a+c, m 2 is the sum of the observation frequencies in the second column of the two-dimensional observation table, that is, b+d; N is a+b+c+d.

[0103] The calculation formula of the chi-square statistic is shown in formula (2):

[0104]

[0105] Among them, χ2 is the chi-square statistic, Q ij represents the observation frequency of the i-th row and j-th column in the two-dimensional observation table, R represents the total number of rows in the two-dimensional observation table, and C represents the total number of columns in the two-dimensional observation table.

[0106] The calculation formula of degrees of freedom is shown in formula (3):

[0107] df=(R-1)(C-1) (3)

[0108] Where df is the degrees of freedom.

[0109] Exemplarily, the two-dimensional observation table corresponding to the process steps is shown in Table 1:

[0110] Table 1

[0111] Machine Group Excellent Group Inferior Group sum Equipment1 a b a+b Equipment2 c d c+d sum a+c b+d n

[0112] Among them, Equipment1 and Equipment2 are the equipment identifiers of a group of machine equipment corresponding to the process step, a is the observation frequency of the target wafer being in the excellent group for the machine equipment with equipment identifier Equipment1 under the process step, b is the observation frequency of the target wafer being in the inferior group for the machine equipment with equipment identifier Equipment1 under the process step, c is the observation frequency of the target wafer being in the excellent group for the machine equipment with equipment identifier Equipment2 under the process step, d is the observation frequency of the target wafer being in the inferior group for the machine equipment with equipment identifier Equipment1 under the process step, and n is the sum of a, b, c and d.

[0113] S320. Determine the process steps that are correlated with wafer abnormality based on the chi-square statistic and the degrees of freedom.

[0114] Specifically, the significance level of each process step is determined from the chi-square distribution table according to the chi-square statistic and degree of freedom corresponding to each process step, and the process step correlated with the wafer abnormality is determined from each process step according to the significance level of each process step, and the machine equipment corresponding to the process step correlated with the wafer abnormality is determined, and the machine equipment corresponding to the process step correlated with the wafer abnormality is tested to determine the faulty machine in the machine equipment. Among them, the significance level is the P value, which is the probability of a more extreme result than the obtained sample observation result when the null hypothesis is true. For example, a p value of 0.05 means that under the null hypothesis, the observed data only appears 5% of the time.

[0115] The above scheme determines the significance level based on the chi-square statistic and the degrees of freedom, and determines the abnormal step from the process steps based on the significance level and the preset significance threshold, which can narrow the scope of screening for faulty machines and improve the screening efficiency for faulty machines. The faulty machines can be screened out from the abnormal machine grouping corresponding to the abnormal step based on the wafer data, which can improve the screening accuracy for faulty machines.

[0116] In one embodiment, determining a faulty tool in the tool equipment according to tool equipment information associated with the process steps and wafer data of the target wafer includes:

[0117] Obtain wafer data and / or statistical data of wafer data of target wafers associated with process steps, group them according to machine equipment information and wafer grouping information to form at least one grouping data; form at least one visualization chart for at least one grouping data according to at least one data dimension to determine the faulty machine in the machine equipment. The visualization charts are statistical charts of various data, wafer charts and trend charts, which may include box plots, scatter plots, CDF (Cumulative Distribution Function) charts and wafer MAP charts, as well as wafer stacking charts.

[0118] Specifically, a process visualization image of the target wafer is drawn according to the wafer data, abnormal information of the target wafer is determined according to the process visualization image, the abnormal information of the target wafer is analyzed, and the faulty machine is determined from the machine devices corresponding to the abnormal machine group.

[0119] Exemplarily, the chip yield corresponding to the process steps that the machine equipment in the abnormal machine group is responsible for when producing the target wafer can be determined based on the visualization chart, and the faulty machine can be determined from the machine equipment in the abnormal machine group based on the chip yield corresponding to the machine equipment.

[0120] Assuming that WAT electrical test data is selected, one of the visualization charts will show the distribution of the equipment under the selected process step in the group, that is, in which group the wafers processed by each equipment are located; a visualization chart will show a wafer stacking diagram, showing the test statistics of the associated wafers on each bare die (bare die with data), so that you can see what kind of problems and problem locations may occur when each device processes the wafer; a visualization chart will show a trend chart in at least one data dimension, and you can choose charts with different data dimensions (that is, change the data on the X-axis). Each point represents the data of a wafer, and the data dimensions can be processing time, wafer dimension, batch dimension and other dimensions. For example, you can analyze according to dimensions such as before and after processing time, each wafer, and each batch. If the processing time is used as the X-axis, you can see the trend of wafer problems and the time point when the abnormality occurs.

[0121] Exemplarily, in one embodiment, determining a faulty machine in the machine equipment according to the machine equipment information associated with the process steps and the wafer data of the target wafer includes:

[0122] Obtain wafer data of a target wafer associated with a process step under a selected wafer data type; group the wafer data according to machine equipment information and wafer grouping information to form at least one type of grouped data; form at least one visual chart according to at least one data dimension for the at least one type of grouped data to determine a faulty machine in the machine equipment.

[0123] The above scheme constructs a visualization chart of the target wafer according to the wafer data, and determines the faulty machine in the machine equipment based on the visualization chart, which can further improve the efficiency of determining the faulty machine. At the same time, it displays the visualization chart to the user, making it easier for the user to understand the abnormal situation of the target wafer based on the process visualization image.

[0124] In one embodiment, determining a faulty machine in the machine equipment according to the machine equipment information associated with the process steps and the wafer data of the target wafer includes:

[0125] Obtain wafer data of the target wafer associated with the process step; generate statistical data of the wafer data of the target wafer passing through each machine device based on the machine equipment information; group the statistical data according to the machine equipment information and wafer grouping information to form grouped data, and determine the faulty machine in the machine equipment based on the grouped data.

[0126] The above scheme groups the wafer data of the target wafer according to the machine equipment information and wafer grouping information through the statistical data of each machine equipment, so as to determine the faulty machine among the machine equipment according to the grouped data, thereby improving the detection efficiency of the faulty machine.

[0127] In one embodiment, Figure 4 As shown, determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data, including:

[0128] S410 , reading sample wafer data of a sample wafer and sample grouping information corresponding to the sample wafer chip from a database.

[0129] The sample wafer is a randomly selected wafer produced in the historical process. The sample wafer data may include defect data, electrical wafer data, measurement data and stability wafer data, and yield data of the sample chip. The sample grouping information corresponding to the sample wafer is the group information of the sample wafer after the sample wafer is grouped.

[0130] S420: Train the neural network model according to the sample wafer data and the sample grouping information to determine the wafer grouping model.

[0131] Specifically, the sample grouping information is used as label data, the neural network model is trained according to the sample wafer data and the sample grouping information, and the trained neural network model is used as the wafer grouping model.

[0132] S430 , determining wafer data of the target wafer according to wafer production data of the target wafer.

[0133] Exemplarily, the production data of the target wafer can be determined by parsing STDF (Standard Test Data File) and semiconductor manufacturing process data storage files, namely, klarf files, and the production data of the target wafer can also be read from the manufacturing execution system of the target wafer.

[0134] S440 , inputting wafer data into a wafer grouping model, and determining wafer grouping information of a target wafer according to output data of the wafer grouping model.

[0135] The above scheme provides a method for training a neural network model to determine a chip grouping model. Through the chip grouping model, the target wafers are grouped according to the wafer data of the target wafers, which can improve the grouping efficiency of the target wafers and save labor costs.

[0136] Exemplarily, training a neural network model according to sample wafer data and sample grouping information to determine a wafer grouping model includes:

[0137] Perform data cleaning on the sample wafer data to determine the valid wafer data; perform principal component analysis on the valid wafer data to determine the feature data; determine the model training data and model wafer data based on the feature data and sample grouping information, and train the neural network model through the model training data to determine the candidate grouping model; use the model wafer data to test the candidate grouping model to determine the model loss function of the candidate grouping model; based on the model loss function, use the error back propagation method to update the model parameters of the candidate grouping model to determine the wafer grouping model.

[0138] The above scheme performs data cleaning on the sample wafer data to determine the effective wafer data for model training of the neural network model, which can improve the reliability and effectiveness of the model training data, perform principal component analysis on the effective wafer data to determine the feature data, and perform model training and model testing on the neural network model through the feature data and sample grouping information, which can improve the model training efficiency and improve the prediction accuracy of the wafer grouping model.

[0139] Exemplarily, based on the above embodiment, the machine fault detection method includes:

[0140] Acquire the production data of the target wafer, calculate the wafer data of the target wafer according to the production data of the target wafer, group the target wafer according to the target wafer data, divide the target wafer into at least two groups, and determine the group of the target wafer. Sort the groups of the target wafers to determine the wafer grouping information of the target wafer. The target wafers can be divided into an excellent group and an inferior group. The user can determine the grouping guidance data from the wafer data of the target wafer according to actual needs, and group the target wafers according to the grouping guidance data. The data type of the grouping guidance data is the data type determined by the user according to actual needs for correlation analysis with the machine. If the grouping guidance data is CP data, the target wafer with a yield value less than a preset yield threshold can be divided into an inferior group, and the target wafer with a yield value greater than or equal to the yield threshold can be divided into an excellent group. If the grouping guidance data is defect data, the target wafer with a relevant statistical value such as defect type / quantity greater than a preset defect critical value can be divided into an inferior group, and the target wafer with a defect value less than or equal to the defect critical value can be divided into an excellent group. If the grouping guidance data is electrical wafer data, the data range of the expected electrical data can be set in advance. If the electrical wafer data falls within the data range of the expected electrical data, the target wafer is divided into an excellent group. If the electrical wafer data does not fall within the data range of the expected electrical data, the target wafer is divided into an inferior group. If the grouping guidance data is measurement data, the data range of the expected measurement data can be set in advance. If the measurement data falls within the data range of the expected measurement data, the target wafer is divided into an excellent group. If the measurement data does not fall within the data range of the expected measurement data, the target wafer is divided into an inferior group. If the grouping guidance data is stability wafer data, the target wafer whose stability wafer data is greater than a preset stability threshold can be divided into an excellent group, and the target wafer whose stability wafer data is less than or equal to the stability threshold can be divided into an inferior group.

[0141] Determine the production process of the target wafer, and generate a visualization list based on the production process of the target wafer. According to the process step selection information sent by the user, select the process step that needs to be paid attention to from the visualization list. The selected process step can be one, two or more. The relevant information of the machine equipment can be imported into the database in advance through the manufacturing execution system or other data import system. The relevant information of the machine equipment can include the equipment identification of the machine equipment, the chip manufacturing process flow that the machine equipment is responsible for, and the type of chip that the machine equipment is responsible for producing and processing. The machine equipment involved in the execution of each process step can be read from the database separately to group the machine equipment based on the process step, that is, the machine equipment that executes the same process step is divided into a group, and the grouping results are integrated to determine the machine grouping data.

[0142] According to the machine grouping data and wafer grouping information, the observation frequencies of the machine equipment corresponding to the process steps in different groups are determined, and the two-dimensional observation table corresponding to the process steps is generated according to the observation frequencies of the machine equipment corresponding to the process steps in different groups. According to the observation frequencies corresponding to the row data and the observation frequencies corresponding to the column data in the two-dimensional observation table, the expected observation frequencies corresponding to the two-dimensional observation table are determined; according to the actual observation frequencies corresponding to the two-dimensional observation table, the expected observation frequencies, the observation frequencies corresponding to the row data in the two-dimensional observation table, the observation frequencies corresponding to the column data, the number of rows and columns in the two-dimensional observation table, the chi-square statistics corresponding to the process steps are determined; according to the number of rows and columns in the two-dimensional observation table, the degrees of freedom corresponding to the process steps are determined.

[0143] Determine the significance level of each process step from the chi-square distribution table according to the chi-square statistic and degree of freedom corresponding to each process step. If the significance level corresponding to the process step is less than the preset significance threshold, the process step is determined to be an abnormal step, and the machine group corresponding to the abnormal step is determined according to the machine grouping data, and the machine group corresponding to the abnormal step is used as the abnormal machine group. Draw a process visualization image of the target wafer according to the wafer data, determine the abnormal information of the target wafer according to the process visualization image, analyze the abnormal information of the target wafer, and determine the faulty machine from the machine equipment corresponding to the abnormal machine group.

[0144] In the above machine fault detection method, the wafer data of the target wafer is determined, and the wafer grouping information of the target wafer is determined according to the wafer data; the machine equipment is grouped according to the process steps of the target wafer, and the machine grouping data is determined according to the grouping results; the two-dimensional observation table corresponding to the process steps is determined according to the machine grouping data and the wafer grouping information, and the chi-square statistics and degrees of freedom corresponding to the process steps are determined based on the two-dimensional observation table; the faulty machine in the machine equipment is determined according to the chi-square statistics and degrees of freedom. This solves the problem that when performing fault detection on machine equipment, it is necessary to rely on the experience of engineers to analyze one by one in the huge amount of data, resulting in low detection efficiency of the machine, and it is impossible to quickly and accurately locate the faulty machine, which affects the chip manufacturing efficiency, and the cost of troubleshooting the machine equipment is high. The above scheme groups the target wafers according to the wafer data, groups the machines and equipment for producing chips based on the process steps, and uses the chi-square test to find the process steps with a lower significance level based on the grouping results of the target wafers and the grouping results of the machines and equipment. The faulty machines in the machines and equipment are located according to the chi-square test results, which can realize automatic fault detection of the machines and equipment, save the fault detection cost of the machines and equipment, and improve the fault detection efficiency of the machines and equipment, thereby improving the yield of the produced chips.

[0145] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0146] Based on the same inventive concept, the embodiment of the present application also provides a machine fault detection device for implementing the machine fault detection method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more machine fault detection device embodiments provided below can refer to the limitations of the machine fault detection method above, and will not be repeated here.

[0147] In one embodiment, Figure 5 As shown, a machine fault detection device is provided, including: a wafer grouping information determination module 501, a machine grouping information determination module 502, a correlation risk model 503 and a faulty machine determination module 504, wherein:

[0148] The wafer grouping information determination module 501 is used to determine the wafer data of the target wafer, and determine the wafer grouping information of the target wafer according to the wafer data;

[0149] The machine grouping information determination module 502 is used to obtain the process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process step to form machine grouping information;

[0150] A correlation risk model 503, used to determine the process steps that are correlated with wafer abnormalities according to the machine grouping information and the wafer grouping information;

[0151] The faulty machine determining module 504 is used to determine a faulty machine among the machine equipment according to the machine equipment information associated with the process steps and the wafer data of the target wafer.

[0152] Exemplarily, the wafer grouping information determination module 501 is specifically used for:

[0153] Select a data type and determine a target wafer, and obtain wafer data of the target wafer under the data type;

[0154] A data threshold for characterizing wafer abnormality under a data type is determined based on wafer data, and wafers are grouped according to the data threshold to form wafer grouping information.

[0155] Exemplarily, the wafer grouping information determining module 501 is further specifically configured to:

[0156] Forming preset statistical data according to the wafer data, and determining a data threshold for characterizing wafer abnormality under the data type according to the statistical data;

[0157] The statistical data are grouped according to a data threshold to obtain at least two wafer groups, forming wafer grouping information.

[0158] Exemplarily, the wafer grouping information determining module 501 is further specifically configured to:

[0159] Determine wafer data of a target wafer, and determine data dimensions of the target wafer according to the wafer data;

[0160] Wafers are grouped according to data dimensions.

[0161] Exemplarily, the machine grouping information determination module 502 is specifically used for:

[0162] Obtaining a selected wafer data type, selecting process step information before generating wafer data of the wafer data type;

[0163] Based on the process step information, obtain the machine equipment information of each process step.

[0164] Exemplarily, the correlation risk model 503 is specifically used for:

[0165] Determine a two-dimensional observation table corresponding to each process step according to the machine grouping information and the wafer grouping information, and determine a chi-square statistic and a degree of freedom corresponding to the process step based on the two-dimensional observation table;

[0166] The process steps associated with wafer anomalies were identified based on the chi-square statistic and degrees of freedom.

[0167] Exemplarily, the faulty machine determination module 504 is specifically used for:

[0168] Obtain wafer data and / or statistical data of the wafer data of the target wafer associated with the process step, and group them according to the machine equipment information and the wafer grouping information to form at least one type of grouping data;

[0169] At least one visual chart is formed for at least one grouped data according to at least one data dimension to determine a faulty machine among the machine equipment.

[0170] Exemplarily, the faulty machine determination module 504 is further specifically configured to:

[0171] Acquire wafer data of a target wafer associated with a process step;

[0172] Generate statistical data of wafer data of the target wafer passing through each machine device based on the machine device information;

[0173] The statistical data are grouped according to the machine equipment information and the wafer grouping information to form grouped data, and the faulty machine among the machine equipment is determined based on the grouped data.

[0174] Exemplarily, the faulty machine determination module 504 is further specifically configured to:

[0175] Acquire wafer data of a target wafer associated with a process step under a selected wafer data type;

[0176] Grouping the wafer data according to the machine equipment information and the wafer grouping information to form at least one type of grouping data;

[0177] At least one visual chart is formed for at least one grouped data according to at least one data dimension to determine a faulty machine among the machine equipment.

[0178] Exemplarily, the wafer grouping information determining module 501 is further specifically configured to:

[0179] Reading sample wafer data of the sample wafer and sample grouping information corresponding to the sample wafer chip from the database;

[0180] Training a neural network model according to sample wafer data and sample grouping information to determine a wafer grouping model;

[0181] determining wafer data of the target wafer according to wafer production data of the target wafer;

[0182] The wafer data is input into the wafer grouping model, and the wafer grouping information of the target wafer is determined according to the output data of the wafer grouping model.

[0183] Each module in the above-mentioned machine fault detection device can be implemented in whole or in part by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module above.

[0184] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a machine fault detection method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0185] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0186] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0187] Step 1: determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data;

[0188] Step 2: Obtain process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0189] Step 3: Determine the process steps that are relevant to the wafer abnormality according to the machine grouping information and the wafer grouping information;

[0190] Step 4: According to the machine equipment information associated with the process steps and the wafer data of the target wafer, the faulty machine among the machines is determined.

[0191] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0192] Step 1: determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data;

[0193] Step 2: Obtain process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0194] Step 3: Determine the process steps that are relevant to the wafer abnormality according to the machine grouping information and the wafer grouping information;

[0195] Step 4: According to the machine equipment information associated with the process steps and the wafer data of the target wafer, the faulty machine among the machines is determined.

[0196] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0197] Step 1: determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data;

[0198] Step 2: Obtain process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information;

[0199] Step 3: Determine the process steps that are relevant to the wafer abnormality according to the machine grouping information and the wafer grouping information;

[0200] Step 4: According to the machine equipment information associated with the process steps and the wafer data of the target wafer, the faulty machine among the machines is determined.

[0201] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0202] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0203] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0204] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A machine fault detection method, characterized in that: include: Determine wafer data of a target wafer, and determine wafer grouping information of the target wafer according to the wafer data; Acquire process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information; Determine a process step that is relevant to wafer abnormality according to the machine grouping information and the wafer grouping information; A faulty machine among the machine equipment is determined according to the machine equipment information associated with the process step and the wafer data of the target wafer.

2. The method according to claim 1, characterized in that The step of determining wafer data of a target wafer and determining wafer grouping information of the target wafer according to the wafer data includes: Select a data type and determine a target wafer, and obtain wafer data of the target wafer under the data type; A data threshold for characterizing wafer abnormality under the data type is determined based on the wafer data, and wafers are grouped according to the data threshold to form wafer grouping information.

3. The method according to claim 2, characterized in that The step of performing wafer grouping based on the wafer data and a data threshold for characterizing wafer abnormality under the data type to form wafer grouping information includes: Forming preset statistical data according to the wafer data, and determining a data threshold for characterizing wafer abnormality under the data type according to the statistical data; The statistical data are grouped according to the data threshold to obtain at least two wafer groups, forming wafer grouping information.

4. The method according to claim 1, characterized in that The step of determining wafer data of a target wafer and determining wafer grouping information of the target wafer according to the wafer data includes: Determine wafer data of a target wafer, and determine a data dimension of the target wafer according to the wafer data; Wafers are grouped according to the data dimension.

5. The method according to claim 1, characterized in that The step of obtaining the process step information and machine equipment information of the target wafer includes: Obtaining a selected wafer data type, and selecting process step information before generating wafer data of the wafer data type; Based on the process step information, obtain the machine equipment information of each process step.

6. The method according to claim 1, characterized in that The process steps of determining the process steps related to wafer abnormality according to the machine grouping information and the wafer grouping information include: Determine a two-dimensional observation table corresponding to each of the process steps according to the machine grouping information and the wafer grouping information, and determine a chi-square statistic and a degree of freedom corresponding to the process step based on the two-dimensional observation table; A process step correlated with wafer abnormality is determined based on the chi-square statistic and the degrees of freedom.

7. The method according to claim 1, characterized in that According to the machine equipment information associated with the process step and the wafer data of the target wafer, a faulty machine in the machine equipment is determined, including: Obtain wafer data of a target wafer associated with the process step and / or statistical data of the wafer data, and group the wafer data according to the machine equipment information and wafer grouping information to form at least one type of grouping data; At least one visualization chart is formed for the at least one grouping data according to at least one data dimension to determine a faulty machine among the machine equipment.

8. The method according to claim 7, characterized in that According to the machine equipment information associated with the process step and the wafer data of the target wafer, a faulty machine in the machine equipment is determined, including: Acquiring wafer data of the target wafer associated with the process step; Generate statistical data of wafer data of the target wafer passing through each machine device based on the machine device information; The statistical data are grouped according to the machine equipment information and the wafer grouping information to form grouped data, and the faulty machine among the machine equipment is determined based on the grouped data.

9. The method according to claim 7, characterized in that: According to the machine equipment information associated with the process step and the wafer data of the target wafer, a faulty machine in the machine equipment is determined, including: Acquiring wafer data of the target wafer associated with the process step; Grouping the wafer data according to the machine equipment information and the wafer grouping information to form at least one type of grouped data; At least one visualization chart is formed for the at least one grouping data according to at least one data dimension to determine a faulty machine among the machine equipment.

10. The method according to claim 1, characterized in that Determining wafer data of a target wafer, and determining wafer grouping information of the target wafer according to the wafer data, including: Reading sample wafer data of a sample wafer and sample grouping information corresponding to the sample wafer chip from a database; Training a neural network model according to the sample wafer data and the sample grouping information to determine a wafer grouping model; Determining wafer data of the target wafer according to wafer production data of the target wafer; The wafer data is input into the wafer grouping model, and the wafer grouping information of the target wafer is determined according to the output data of the wafer grouping model.

11. A machine fault detection device, characterized in that: The machine fault detection device comprises: A wafer grouping information determination module, used to determine wafer data of a target wafer, and determine wafer grouping information of the target wafer according to the wafer data; A machine grouping information determination module is used to obtain process step information and machine equipment information of the target wafer, and group the machine equipment information of the target wafer according to the process steps to form machine grouping information; A correlation risk model, used to determine the process steps that are correlated with wafer abnormalities according to the machine grouping information and the wafer grouping information; The faulty machine determining module is used to determine the faulty machine in the machine equipment according to the machine equipment information associated with the process step and the wafer data of the target wafer.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 11 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

Images