Wafer process data analysis method based on Ftest

Abstract

The invention discloses a wafer process data analysis method based on Ftest inspection, and belongs to the technical field of wafer process data analysis. The method is realized based on a wafer process data analysis system, and the system comprises a data acquisition module, a database, a data processing module, a critical value confirmation module, a calculation and analysis module and an intelligent adjustment module. According to the method, logic grouping is carried out on detection parameters after wafer processing is finished, and Ftest inspection analysis is carried out, so that the problem of accurately identifying and analyzing a complex relationship between groups in massive wafer processing data is effectively solved, and the integrity and comprehensiveness of wafer processing data analysis are reflected; according to specific requirements of the wafer manufacturing industry, the variance of different detection parameter groups and the F statistical magnitude between the variance of the different detection parameter groups are calculated, abnormal setting parameters influencing the wafer performance can be rapidly identified, a wafer processing machine table is controlled to automatically correct the abnormal setting parameters, and the production and processing quality of wafers is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wafer process data analysis, and in particular to a wafer process data analysis method based on Ftest inspection. Background Art

[0002] Wafer process data refers to a series of data generated and collected during the semiconductor wafer manufacturing process, which covers information from raw material preparation, wafer processing, testing to packaging and other links; by analyzing wafer process data, problems in the wafer processing production process can be effectively discovered, so as to make rapid adjustments; among them, Ftest, as a classic statistical test method, is crucial for evaluating the consistency and significance of variances between wafer process data; in the prior art, due to reliance on single-machine computing resources, with the surge in data volume, the analysis process of wafer process data cannot be efficiently parallelized; and there are differences in wafer processing machines, and the ability to analyze and mine wafer process data is insufficient, which can easily cause inaccurate processing parameters and affect wafer quality. Summary of the invention

[0003] The purpose of the present invention is to provide a wafer process data analysis method based on Ftest inspection to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a wafer process data analysis method based on Ftest inspection, the method comprising the following steps:

[0005] S1. Processing a wafer using a wafer processing machine and collecting wafer process data, wherein the wafer data includes setting parameters of the wafer processing machine during wafer processing and detection parameters after wafer processing is completed;

[0006] S2, processing the collected wafer process data, and logically grouping the detection parameters according to the wafer processing machine ID;

[0007] S3, perform Ftest test on the detection parameters after logical grouping, calculate the variance of different detection parameter groups, and calculate the F statistic between the variances of different detection parameter groups; compare the calculated F statistic with the critical value of the F distribution to determine whether there is an abnormal detection parameter group; when there is an abnormal detection parameter group, execute step S4;

[0008] S4. Determine corresponding abnormal setting parameters according to the abnormal detection parameter group, and automatically correct the abnormal setting parameters in the wafer processing machine.

[0009] A wafer process data analysis system based on Ftest inspection, the system comprising: a data acquisition module, a database, a data processing module, a critical value confirmation module, a calculation and analysis module and an intelligent adjustment module;

[0010] The data acquisition module is used to collect wafer process data;

[0011] The database is used to store the collected wafer process data as historical data and continuously update the database;

[0012] The data processing module is used to process the collected wafer process data and logically group the wafer process data;

[0013] The critical value confirmation module is used to determine the critical value of the F distribution under different detection parameters;

[0014] The calculation and analysis module is used to perform an Ftest test on the detection parameters after logical grouping, calculate the variance of different detection parameter groups, and calculate the F statistic between the variances of different detection parameter groups; determine whether there is an abnormal detection parameter group;

[0015] The intelligent adjustment module is used to automatically correct abnormal setting parameters in the wafer processing machine.

[0016] Compared with the prior art, the beneficial effects achieved by the present invention are: logically grouping the detection parameters after wafer processing is completed, and through Ftest inspection and analysis, the difficult problem of accurately identifying and analyzing the complex relationship between groups in massive wafer process data is effectively solved, reflecting the integrity and comprehensiveness of wafer process data analysis; according to the specific needs of the wafer manufacturing industry, the variance of different detection parameter groups and the F statistic between the variances of different detection parameter groups are calculated, which can quickly identify abnormal setting parameters that affect wafer performance and control the wafer processing machine to automatically correct the abnormal setting parameters, thereby improving the production and processing quality of wafers. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural schematic diagram of a wafer process data analysis system based on Ftest inspection of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1, the present invention provides a technical solution:

[0020] See also Figure 1 In this embodiment, a wafer process data analysis method based on Ftest inspection is provided, and the method includes the following steps:

[0021] S1. Processing a wafer using a wafer processing machine and collecting wafer process data, wherein the wafer data includes setting parameters of the wafer processing machine during wafer processing and detection parameters after wafer processing is completed.

[0022] It should be noted that the wafer processing machine is used to process the wafer production process, including but not limited to wafer polishing and wafer cutting; the setting parameters and detection parameters are parameters corresponding to the type of action of the wafer processing machine on the wafer.

[0023] In this implementation, a wafer processing machine is used to cut wafers, at which time the setting parameters represent the cutting depth of the wafer processing machine for the wafer, and the detection parameters represent the actual cutting depth of the wafer material; n wafer processing machines are used to process the same batch of wafers, and the initial setting parameters and other condition parameters of each wafer processing machine are the same; in the process of collecting the detection parameter group, the wafer process data stored in the Parquet format is read from the Hadoop Distributed FileSystem or a similar distributed file system through the PySpark API; the Parquet file format is particularly suitable for fast reading in big data scenarios due to its columnar storage and efficient compression characteristics.

[0024] S2. Process the collected wafer process data and logically group the detection parameters according to the wafer processing machine ID.

[0025] Specifically, the method steps for logically grouping wafer process data are as follows:

[0026] S21, determining the number n of wafer processing machines, and determining different IDs of the n wafer processing machines;

[0027] S22, using a wafer processing machine to process the wafers of the same batch, and logically grouping the detection parameters after wafer processing under the same setting parameters according to the wafer processing machine ID to obtain a detection parameter group after logical grouping, which is recorded as a set H; H = {H1, H2, ..., H n}; H i ={h i1 、h i2 ,...,h imi};

[0028] Among them, H1, H2, ..., H nThey represent the detection parameter groups after wafer processing of different wafer processing machines; H i represents the detection parameter group after the wafer processing of the i-th wafer processing machine is completed; i = 1, 2, ..., n; h i1 、h i2 ,...,h imi Respectively represent different detection parameters in the i-th detection parameter group; m i represents the number of detection parameters in the i-th detection parameter group; the number of detection parameters in the detection parameter group is the same as the number of wafers in the wafer processing machine.

[0029] It should be noted that the set H is the detection parameters logically grouped under the same setting parameters during the processing of the same batch of wafers; the number of detection parameters in the detection parameter group is the same as the number of wafers in the wafer processing machine, so that each wafer has a corresponding detection parameter.

[0030] In this implementation, PySpark's groupby operation is used to logically group the wafer process data according to the pre-set wafer processing machine ID, and prepare detection parameter groups of different wafer processing machines for subsequent Ftest inspection. By logically grouping the wafer process data and performing Ftest inspection and analysis, the difficult problem of accurately identifying and parsing the complex relationship between groups in massive wafer process data is effectively solved. Therefore, according to the actual cutting depth of the wafer material in the detection parameters, the cutting depth of the wafer by the wafer processing machine in the set parameters is adjusted, thereby improving the analysis capability of the wafer process data and reflecting the integrity and comprehensiveness of the wafer process data analysis.

[0031] S3. Perform Ftest on the detection parameters after logical grouping, calculate the variance of different detection parameter groups, and calculate the F statistic between the variances of different detection parameter groups; compare the calculated F statistic with the critical value of the F distribution to determine whether there is an abnormal detection parameter group; when there is an abnormal detection parameter group, execute step S4.

[0032] It should be noted that the critical value of the F distribution is a series of values ​​generated by statistical software or mathematical functions, which depends on the significance level set by the manager and the degrees of freedom in the test parameters; the significance level indicates the risk of falsely rejecting the null hypothesis that the manager can accept; the manager determines the significance level of the Ftest based on historical wafer process data, and at a given significance level and degrees of freedom, determines whether the calculated F statistic is extreme enough through the critical value of the F distribution, thereby deciding whether to reject the null hypothesis; when the calculated F statistic is greater than the critical value of the F distribution, the null hypothesis is rejected, and it is considered that the test involved in the calculation is There are significant differences in the parameter groups, which are abnormal detection parameter groups; when the calculated F statistic is less than or equal to the critical value of the F distribution, the null hypothesis is not rejected, and it is considered that there is insufficient evidence to show that there are significant differences in the means of the detection parameter groups, which are normal detection parameter groups; by splitting the calculation process of an Ftest into multiple detection parameter groups, the variances of different detection parameter groups and the F statistics between the variances of different detection parameter groups are obtained, and highly optimized data processing procedures and algorithms are designed, which can quickly identify abnormal setting parameters that affect wafer performance and control the wafer processing machine to automatically correct the abnormal setting parameters, thereby improving the production and processing quality of wafers.

[0033] In this implementation, multi-threading technology is used to submit multiple detection parameter group calculation tasks of Ftest inspection to the Spark cluster in parallel, and multiple independent Ftest jobs can be submitted to the Spark cluster at the same time, so that the computing resources of the entire cluster are more fully utilized and the overall processing speed is significantly improved; in each detection parameter group, a customized UDF is applied to execute the Ftest inspection, and this UDF encapsulates the data processing flow and algorithm of Ftest; when the variances of the detection parameters in different detection parameter groups are compared pairwise, the subprocesses are started through Python's multiprocessing module for parallel execution, and multiple subprocesses share the calculation tasks of these detection parameter groups.

[0034] Specifically, the method steps are:

[0035] S31, according to different detection parameters in different detection parameter groups, respectively calculate the variance of the detection parameter groups after wafer processing of different wafer processing machines is completed According to the calculation formula:

[0036]

[0037] in, represents the variance of the detection parameters in the i-th detection parameter group; h ij represents the jth detection parameter in the i-th detection parameter group; Indicates the mth detection parameter group in the i iThe average value of the detection parameters; j = 1, 2, ..., m i ;

[0038] S32, comparing the variances of different detection parameter groups in pairs, and calculating the F statistic between the variances of different detection parameter groups, according to the calculation formula:

[0039]

[0040] Among them, F ab The F statistic representing the variance of the ath detection parameter group and the variance of the bth detection parameter group; represents the variance of the detection parameters in the ath detection parameter group; represents the variance of the detection parameters in the b-th detection parameter group; a, b = 1, 2, ..., n, and a ≠ b;

[0041] S33, according to the significance level set by the management, F ab The numerator and denominator degrees of freedom in the F distribution are used to determine the critical value K corresponding to the F distribution. ab , K ab With F ab For comparison; when K ab ≥F ab When K ab <F ab When , the detection parameter groups of the ath and bth wafer processing machines are used as abnormal detection parameter groups.

[0042] In this embodiment, the molecular degree of freedom df a =m a -1; the denominator degree of freedom df b =m b -1; where m a Indicates the number of detection parameters in the ath detection parameter group; m b Represents the number of detection parameters in the bth detection parameter group; through the determined significance level, the molecular degree of freedom df a and the denominator degrees of freedom df b , determine the critical value K corresponding to the F distribution ab , so as to determine whether the variance of the detection parameter groups of the ath and bth wafer processing machines is significantly different.

[0043] Furthermore, the collected wafer process data is stored as historical data in a database, and the database is continuously updated; based on the historical wafer process data stored in the database, the management personnel set the significance level of the detection parameters, and determine the critical value of the F distribution under different detection parameters according to the degrees of freedom in the detection parameters.

[0044] It should be noted that the method for determining the critical value of F distribution under different detection parameters is: according to the molecular degrees of freedom df a , denominator degrees of freedom df b , taking the F statistic as the independent variable, we get the probability density function of the F distribution According to the determined significance level, the critical value of the F distribution is determined; wherein the critical value of the F distribution makes the right side area of ​​the probability density function curve equal to the determined significance level.

[0045] S4. Determine corresponding abnormal setting parameters according to the abnormal detection parameter group, and automatically correct the abnormal setting parameters in the wafer processing machine.

[0046] Specifically, the method steps are:

[0047] S41, analyzing the abnormal detection parameter group, determining that the detection parameter group corresponding to the abnormal detection parameter group after pairwise comparison with other detection parameter groups in steps S32-S33, and using the detection parameter group as adjustment analysis data, determining the wafer processing machine corresponding to the adjustment analysis data, and using the setting parameters of the wafer processing machine as the abnormal setting parameters;

[0048] S42: Determine the total average value of the detection parameters according to the detection parameter group set H after logical grouping. Among them, h ij represents the jth detection parameter in the i-th detection parameter group;

[0049] S43, adjusting the detection parameters in the analysis data and Compare and determine if greater than The detection parameter set X1 and less than The detection parameter set X2 is calculated based on the square difference between the detection parameters in the detection parameter sets X1 and X2. The discrete degrees L1 and L2 are determined, and the abnormal setting parameters in the wafer processing machine are adjusted according to the difference between L1 and L2.

[0050] In this embodiment, a detection parameter group that is compared with other detection parameter groups and is all corresponding to the abnormal detection parameter group is determined, and the detection parameter group is used as adjustment analysis data; for example, when analyzing the abnormal detection parameter group, it is found that after comparing the detection parameter group in the second wafer processing machine with the detection parameter groups of other wafer processing machines, they are all abnormal detection parameter groups, so it is determined that the setting parameters in the second wafer processing machine are inaccurate, the detection parameter group in the second wafer processing machine is used as adjustment analysis data, and the setting parameters in the second wafer processing machine are used as abnormal setting parameters; the detection parameters in the second detection parameter group are respectively compared with the detection parameter groups of other wafer processing machines. Compare and filter out those greater than The detection parameter set X1 and less than The detection parameter set X2 is combined with the detection parameters in the set X1 Sum the squared differences to obtain the discreteness L1; compare the detection parameters in set X2 with The square difference is summed to obtain the discreteness L2; ​​since the setting parameters in this embodiment represent the cutting depth of the wafer by the wafer processing machine, and the detection parameters represent the actual cutting depth of the wafer material, the differences in the wafer processing machines make it impossible for the actual cutting depth of the wafer material to reach the ideal condition, by calculating the difference between L1 and L2, when L1-L2 is greater than 0, the actual cutting depth of the wafer material by the wafer processing machine is too high, and the setting parameters in the second wafer processing machine are corrected to reduce the cutting depth of the wafer, and the larger L1-L2 is, the more the cutting depth is reduced; when L1-L2 is less than 0, the actual cutting depth of the wafer material by the wafer processing machine is too low, and the setting parameters in the second wafer processing machine are corrected to increase the cutting depth of the wafer by the wafer processing machine, and the smaller L1-L2 is, the more the cutting depth is increased; by automatically adjusting the setting parameters of the wafer processing machine, the production quality of the wafer is improved.

[0051] It should be noted that steps S1-S4 are a process of automatically correcting the setting parameters in the wafer processing machine after processing a batch of wafers. After processing the next batch of wafers, steps S1-S4 are repeated until the wafer processing machine is turned off.

[0052] A wafer process data analysis system based on Ftest inspection, the system comprising: a data acquisition module, a database, a data processing module, a critical value confirmation module, a calculation and analysis module and an intelligent adjustment module;

[0053] The data acquisition module is used to collect wafer process data;

[0054] The database is used to store the collected wafer process data as historical data and continuously update the database;

[0055] The data processing module is used to process the collected wafer process data and logically group the wafer process data;

[0056] The critical value confirmation module is used to determine the critical value of the F distribution under different detection parameters;

[0057] The calculation and analysis module is used to perform an Ftest test on the detection parameters after logical grouping, calculate the variance of different detection parameter groups, and calculate the F statistic between the variances of different detection parameter groups; determine whether there is an abnormal detection parameter group;

[0058] The intelligent adjustment module is used to automatically correct abnormal setting parameters in the wafer processing machine.

[0059] Furthermore, the intelligent adjustment module includes a human-computer interaction platform for digitally displaying the collected wafer process data, the variances of different detection parameter groups calculated, and the F statistics between the variances of different detection parameter groups; wherein, the management personnel can manually adjust the setting parameters of the wafer processing machine through the human-computer interaction platform.

[0060] In this embodiment:

[0061] The wafer is processed using a wafer processing machine, and the data acquisition module collects the wafer process data, and sends the collected wafer process data to the database and the data processing module;

[0062] The data processing module processes the collected wafer process data, logically groups the detection parameters according to the wafer processing machine ID, and sends the logically grouped detection parameters to the calculation and analysis module;

[0063] The critical value confirmation module determines the critical value of the F distribution under different detection parameters based on the historical wafer process data stored in the database and the significance level set by the management personnel and the degrees of freedom in the detection parameters;

[0064] The calculation and analysis module performs Ftest test on the detection parameters after logical grouping, calculates the variance of different detection parameter groups, and calculates the F statistic between the variances of different detection parameter groups; according to the critical value of the F distribution under different detection parameters determined in the critical value confirmation module, the calculated F statistic is compared with the critical value of the F distribution to determine whether there is an abnormal detection parameter group; when there is an abnormal detection parameter group, the abnormal detection parameter group is sent to the intelligent adjustment module; if the detection parameter group is normal, no operation is performed;

[0065] The intelligent adjustment module determines the corresponding abnormal setting parameters according to the abnormal detection parameter group, and automatically corrects the abnormal setting parameters in the wafer processing machine;

[0066] Among them, when the management personnel manually adjust the setting parameters of the wafer processing machine through the human-computer interaction platform, the intelligent adjustment module keeps the setting parameters in the wafer processing machine fixed.

[0067] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A wafer process data analysis method based on Ftest inspection, characterized in that: The method comprises the following steps: S1. Processing a wafer using a wafer processing machine and collecting wafer process data, wherein the wafer data includes setting parameters of the wafer processing machine during wafer processing and detection parameters after wafer processing is completed; S2, processing the collected wafer process data, and logically grouping the detection parameters according to the wafer processing machine ID; S3, perform Ftest test on the detection parameters after logical grouping, calculate the variance of different detection parameter groups, and calculate the F statistic between the variances of different detection parameter groups; compare the calculated F statistic with the critical value of the F distribution to determine whether there is an abnormal detection parameter group; when there is an abnormal detection parameter group, execute step S4; S4. Determine corresponding abnormal setting parameters according to the abnormal detection parameter group, and automatically correct the abnormal setting parameters in the wafer processing machine.

2. The wafer process data analysis method based on Ftest inspection according to claim 1 is characterized in that: The method steps for logically grouping wafer process data are as follows: S21, determining the number n of wafer processing machines, and determining different IDs of the n wafer processing machines; S22, using a wafer processing machine to process the wafers of the same batch, and logically grouping the detection parameters after wafer processing under the same setting parameters according to the wafer processing machine ID to obtain a detection parameter group after logical grouping, which is recorded as a set H; H = {H1, H2, ..., H n }; H i ={h i1 、h i2 ,...,h imi }; Among them, H1, H2, ..., H n They represent the detection parameter groups after wafer processing of different wafer processing machines; H i represents the detection parameter group after the wafer processing of the i-th wafer processing machine is completed; i = 1, 2, ..., n; h i1 、h i2 ,...,h imi Respectively represent different detection parameters in the i-th detection parameter group; m i represents the number of detection parameters in the i-th detection parameter group; the number of detection parameters in the detection parameter group is the same as the number of wafers in the wafer processing machine.

3. The wafer process data analysis method based on Ftest inspection according to claim 2 is characterized in that: The method steps of step S3 are: S31, according to different detection parameters in different detection parameter groups, respectively calculate the variance of the detection parameter groups after wafer processing of different wafer processing machines is completed According to the calculation formula: in, represents the variance of the detection parameters in the i-th detection parameter group; h ij represents the jth detection parameter in the i-th detection parameter group; Indicates the mth detection parameter group in the i i The average value of the detection parameters; j = 1, 2, ..., m i ; S32, comparing the variances of different detection parameter groups in pairs, and calculating the F statistic between the variances of different detection parameter groups, according to the calculation formula: Among them, F ab The F statistic representing the variance of the ath detection parameter group and the variance of the bth detection parameter group; represents the variance of the detection parameters in the ath detection parameter group; represents the variance of the detection parameters in the b-th detection parameter group; a, b = 1, 2, ..., n, and a ≠ b; S33, according to the significance level set by the management, F ab The numerator and denominator degrees of freedom in the F distribution are used to determine the critical value K corresponding to the F distribution. ab , K ab With F ab For comparison; when K ab ≥F ab When K ab <F ab When , the detection parameter groups of the ath and bth wafer processing machines are used as abnormal detection parameter groups.

4. The wafer process data analysis method based on Ftest inspection according to claim 3 is characterized in that: The collected wafer process data are stored as historical data in a database, and the database is continuously updated. Based on the historical wafer process data stored in the database, managers set the significance level of the detection parameters and determine the critical value of the F distribution under different detection parameters based on the degrees of freedom in the detection parameters.

5. The wafer process data analysis method based on Ftest inspection according to claim 3 is characterized in that: The method steps of step S4 are: S41, analyzing the abnormal detection parameter group, determining a detection parameter group that corresponds to the abnormal detection parameter group after pairwise comparison with other detection parameter groups, and using the detection parameter group as adjustment analysis data, determining a wafer processing machine corresponding to the adjustment analysis data, and using the setting parameters of the wafer processing machine as the abnormal setting parameters; S42: Determine the total average value of the detection parameters according to the detection parameter group set H after logical grouping. S43, adjusting the detection parameters in the analysis data and Compare and determine if greater than The detection parameter set X1 and less than The detection parameter set X2 is calculated based on the square difference between the detection parameters in the detection parameter sets X1 and X2. The discrete degrees L1 and L2 are determined, and the abnormal setting parameters in the wafer processing machine are adjusted according to the difference between L1 and L2.

6. A wafer process data analysis system based on Ftest inspection, using a wafer process data analysis method based on Ftest inspection as described in any one of claims 1 to 5, characterized in that: The system includes: a data acquisition module, a database, a data processing module, a critical value confirmation module, a calculation and analysis module and an intelligent adjustment module; The data acquisition module is used to collect wafer process data; The database is used to store the collected wafer process data as historical data and continuously update the database; The data processing module is used to process the collected wafer process data and logically group the wafer process data; The critical value confirmation module is used to determine the critical value of the F distribution under different detection parameters; The calculation and analysis module is used to perform an Ftest test on the detection parameters after logical grouping, calculate the variance of different detection parameter groups, and calculate the F statistic between the variances of different detection parameter groups; determine whether there is an abnormal detection parameter group; The intelligent adjustment module is used to automatically correct abnormal setting parameters in the wafer processing machine.

7. The wafer process data analysis method based on Ftest inspection according to claim 6 is characterized in that: The intelligent adjustment module includes a human-computer interaction platform, which is used to digitally display the collected wafer process data, the variances of different detection parameter groups calculated, and the F statistics between the variances of different detection parameter groups; wherein, the management personnel can manually adjust the setting parameters of the wafer processing machine through the human-computer interaction platform.

Images