Defective root cause analysis method and system based on comprehensive analysis framework

Abstract

The invention provides a bad root cause analysis method and system based on a comprehensive analysis framework, and relates to the technical field of bad root cause analysis, and the method comprises the steps: obtaining bad data and processing process data; performing comprehensive analysis on the bad data and the processing process data according to the comprehensive analysis framework so as to position intermediate link factors related to the bad product; performing root cause positioning analysis on the processing process data according to the bad data so as to preliminarily position a root cause station and root cause machine equipment; root cause positioning analysis is carried out on the processing process data according to the intermediate link factors related to the bad products so as to further position root cause machine equipment; and performing fusion judgment on the root cause site and the root cause machine equipment which are preliminarily positioned and the root cause machine equipment which is further positioned to obtain a comprehensive root cause positioning result. The root cause positioning of the processing equipment is carried out based on the comprehensive analysis framework, and the problem that comprehensive analysis and judgment cannot be realized due to isolation between different bad root cause analysis means is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of root cause analysis of defects, and in particular to a root cause analysis method and system based on a comprehensive analysis framework. Background Art

[0002] The processing technology of display panels is complex and the processing flow is cumbersome. Therefore, when a display panel fails, it is often difficult to quickly and accurately locate the machine equipment that causes the failure. For the root cause location of defective products during panel processing, the existing defect root cause analysis methods mainly include the following: (1) Analysis based on expert experience: First, manually collect the original data of the defective products, and preliminarily locate the possible causes of the failure based on expert experience, then calculate the difference in yield between the equipment, and finally locate the root cause site that causes the failure based on the defect rate ranking; (2) Analysis by statistical methods: First, use statistical methods to analyze the differences between parallel equipment and parallel chambers at each site, and then locate the root cause site; (3) Analysis by machine learning algorithm: First, use machine learning algorithms to fit the equipment, parameters, etc. that the panel product passes through, predict the defect rate of the panel product, and then obtain the root cause results based on feature importance.

[0003] The above defective root cause analysis methods have the following defects: ① Different defective root cause analysis methods are often isolated from each other, and they are all independent analyses, which cannot achieve comprehensive analysis and judgment. ② Different defective root cause analyses are limited to analyzing the correlation between the sampling results of the intermediate links and the defects, in order to strengthen the monitoring of the sampling results of the intermediate links, and the analysis of the defective root causes is not detailed and comprehensive enough. ③ In the process of analyzing the differences in different equipment groups, the existing defective root cause analysis methods directly locate the root cause equipment, and the accuracy rate is low. ④ The existing defective root cause analysis methods often directly use the variance analysis method, but ignore the requirements of the variance analysis method for normality and homogeneity. In practice, the film passing volume between different equipment groups varies greatly, and it is difficult to meet this prerequisite requirement. Summary of the invention

[0004] The present invention provides a method and system for analyzing root causes of defects based on a comprehensive analysis framework, which solves the problem that different root cause analysis methods of existing root causes of defects are often isolated from each other and cannot achieve comprehensive analysis and judgment.

[0005] In a first aspect, an embodiment of the present invention provides a method for root cause analysis based on a comprehensive analysis framework, the method comprising the following process:

[0006] Obtain defect data and processing data of panel products;

[0007] Comprehensively analyze defective data and processing data based on the comprehensive analysis framework to locate intermediate factors related to product defects;

[0008] Conduct root cause analysis on the processing data based on the defect data to preliminarily locate the root cause site and root cause equipment;

[0009] Conduct root cause analysis on processing data based on intermediate factors related to product defects to further locate the root cause machine equipment;

[0010] The initially located root cause sites and root cause equipment as well as the further located root cause equipment are integrated and judged to obtain a comprehensive root cause location result.

[0011] In the above embodiment, the present invention effectively combines multiple analysis methods such as defect density analysis, processing time analysis, and measurement value analysis based on a comprehensive analysis framework to locate the root cause of the processing equipment, which can effectively improve the accuracy and effectiveness of equipment root cause location. At the same time, it solves the problem that different defect root cause analysis methods are isolated and cannot achieve comprehensive analysis and judgment.

[0012] As some optional implementation modes of the present application, the defective data includes the defective rate and category labels of defective products; the processing data of the panel products includes the production history data, intermediate sampling data and intermediate measurement data of the defective products.

[0013] In the above embodiment, the present invention obtains different types of processing process data, which facilitates the subsequent selection of different analysis methods to further trace back and analyze the preceding processing factors related to the intermediate sampling results, thereby further locating the root cause to the preceding processing site and processing machine equipment, thereby achieving a more detailed and comprehensive analysis of the defect root cause.

[0014] As some optional implementation modes of the present application, after obtaining the defective data and processing data of the panel product, the defective data and processing data need to be preprocessed; wherein, the preprocessing includes removing null values, calculating the defective rate and removing abnormal values.

[0015] In the above embodiment, the present invention pre-processes the data to make the data more accurate, which is convenient for subsequent root cause location analysis.

[0016] As some optional implementation methods of the present application, the process of comprehensive analysis of bad data and processing data according to the comprehensive analysis framework is as follows:

[0017] Conduct forward-looking root cause location analysis or direct root cause location analysis based on bad data;

[0018] If the forward root cause location analysis is adopted, the corresponding analysis means are selected according to the analysis object, and the correlation coefficient or sequence consistency coefficient is calculated for the production history data, intermediate sampling data and intermediate measurement data according to the selected analysis means to locate the intermediate link factors related to product defects, wherein the analysis means include defect density analysis, processing time analysis and measurement value analysis;

[0019] If direct root cause location analysis is used, root cause location analysis is performed directly on the processing data based on the defective data to preliminarily locate the root cause site and root cause machine equipment.

[0020] In the above embodiment, the comprehensive analysis framework adopted by the present invention has strong flexibility. Users can flexibly select analysis methods such as defect density analysis, processing time analysis, and measurement value analysis, as well as analysis paths such as forward root cause location analysis or direct root cause location analysis according to business needs to achieve any combination.

[0021] As some optional implementation methods of the present application, the process of performing root cause location analysis on processing data based on bad data is as follows:

[0022] Group the production history data according to the defect data to obtain the defect data of products in different equipment groups;

[0023] Perform difference analysis on the defective data of products in different device groups to determine whether there are differences between the devices in different groups currently being analyzed;

[0024] If there are differences between different device groups, locate the device in the group with the largest difference to obtain the preliminary located root cause site and root cause machine device.

[0025] In the above embodiment, the present invention first locates and analyzes the suspicious site related to the bad situation based on the suspicious site range preliminarily determined by the user, and then locates and analyzes the processing equipment under the site, which reduces the difficulty of directly locating the root cause equipment and improves the hit rate of the root cause equipment.

[0026] As some optional implementation methods of the present application, the process of performing root cause location analysis on processing data based on intermediate factors related to product defects is as follows:

[0027] Group production history data according to intermediate factors related to product defects to obtain defect data of products in different equipment groups;

[0028] Perform difference analysis on the defective data of products in different equipment groups to determine whether there are differences between the different grouped equipment currently being analyzed;

[0029] If there are differences between different device groups, locate the device in the group with the largest difference to obtain the root cause device for further location.

[0030] In the above embodiment, the present invention further traces back and analyzes the preceding processing factors related to the intermediate sampling results based on the defect analysis results of the intermediate sampling links such as defect density analysis, processing time analysis, and measurement value analysis, thereby further locating the root cause to the preceding processing site and processing machine equipment, thereby achieving a more detailed and comprehensive analysis of the defect root cause.

[0031] As some optional implementation methods of the present application, the process of fusion judgment on the initially located root cause site and root cause machine equipment and the further located root cause machine equipment is as follows:

[0032] A weighted average algorithm or a counting algorithm is used to comprehensively rank the initially located root cause sites and root cause equipment and the further located root cause equipment, and the root cause sites and root cause equipment are comprehensively located according to the comprehensive ranking;

[0033] The root cause sites and root cause equipment of the comprehensive positioning are screened and processed to obtain the comprehensive root cause positioning results.

[0034] In the above embodiment, the present invention uses a counting method to assign points to the root cause rankings given, so as to obtain the most likely root causes of defects. Moreover, the user can conduct further targeted analysis and investigation based on the root cause sites and root cause equipment located after comprehensive judgment, and make relevant adjustments in a timely manner to quickly adjust the defects and improve the yield.

[0035] In a second aspect, the present invention provides a root cause analysis system based on a comprehensive analysis framework, the system comprising:

[0036] A data acquisition unit, the data acquisition unit is used to acquire defect data and processing data of panel products;

[0037] A comprehensive analysis unit, which is used to conduct a comprehensive analysis on the defect data and the processing data according to a comprehensive analysis framework to locate the intermediate factors related to the product defect;

[0038] A primary positioning unit, the primary positioning unit is used to perform root cause positioning analysis on the processing data according to the defect data, so as to preliminarily locate the root cause site and the root cause machine equipment;

[0039] A secondary positioning unit, which is used to perform root cause positioning analysis on processing data based on intermediate factors related to product defects, so as to further locate the root cause machine equipment;

[0040] The fusion judgment unit is used to perform fusion judgment on the initially located root cause site and root cause machine equipment and the further located root cause machine equipment to obtain a comprehensive root cause positioning result.

[0041] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for analyzing root causes of defects based on a comprehensive analysis framework when executing the computer program.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for analyzing root causes of defects based on a comprehensive analysis framework is implemented.

[0043] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method for analyzing root causes of defects based on a comprehensive analysis framework.

[0044] The beneficial effects of the present invention are as follows:

[0045] 1. Based on the comprehensive analysis framework, the present invention effectively combines multiple analysis methods such as defect density analysis, processing time analysis and measurement value analysis to locate the root cause of processing equipment, which can effectively improve the accuracy and effectiveness of equipment root cause location. The comprehensive analysis framework has strong flexibility. Users can flexibly select analysis methods and analysis paths according to business needs to achieve any combination.

[0046] 2. The present invention further traces back and analyzes the preceding processing factors related to the intermediate sampling results based on the defect analysis results of the intermediate sampling links such as defect density analysis, processing time analysis and measurement value analysis, thereby further locating the root cause to the preceding processing site and processing machine equipment, achieving a more detailed and comprehensive analysis of the defect root cause.

[0047] 3. The present invention first locates and analyzes the suspicious sites related to bad conditions based on the scope of suspicious sites preliminarily determined by the user, and then locates and analyzes the processing equipment under the site, which reduces the difficulty of directly locating the root cause equipment and improves the hit rate of the root cause equipment; at the same time, the analyzed equipment level is divided into equipment level, sub-equipment level, and chamber level, which is more comprehensively designed and can flexibly determine the granularity of the analysis as needed.

[0048] 4. The present invention helps to quickly screen and determine the root cause of defects. The principle is clear and easy to understand and grasp, the calculation is simple, and there is no need to build a model or collect other data. It can achieve the effect of quick judgment and rapid screening. It can not only be used for root cause inference of defective products, but also can be promoted and applied to other root cause inference scenarios, such as determining the link or equipment component where the fault occurs in operation and maintenance or fault repair scenarios, and has strong versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 is a schematic diagram of the computer device structure of the hardware operating environment described in the embodiment of the present invention;

[0051] Figure 2 is a flow chart of the defect root cause analysis method according to an embodiment of the present invention;

[0052] Figure 3 is a schematic diagram of analysis and inference of the comprehensive analysis framework described in an embodiment of the present invention;

[0053] Figure 4 It is a structural block diagram of the defect root cause analysis system described in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] 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.

[0055] In order to solve the problem that different root cause analysis methods of existing bad root causes are often isolated from each other and cannot achieve comprehensive analysis and judgment, the embodiment of the present invention provides a bad root cause analysis method and system based on a comprehensive analysis framework. Before introducing the specific technical solution of the present application, the hardware operating environment involved in the embodiment of the present application is first introduced.

[0056] See also Figure 1 , Figure 1 The figure is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present invention.

[0057] like Figure 1As shown, the computer device may include: a processor, such as a central processing unit (CPU), a communication bus, a user interface, a network interface, and a memory. Among them, the communication bus is used to realize the connection and communication between these components. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), and optionally, the user interface may also include a standard wired interface and a wireless interface. The network interface may be a wired interface or a wireless interface (such as a wireless fidelity interface). The memory may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory may optionally be a storage device independent of the aforementioned processor.

[0058] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0059] like Figure 1 As shown, the memory as a storage medium may include an operating system module, a network communication module, a user interface module and a data storage module.

[0060] exist Figure 1 In the computer device shown, the network interface is mainly used for data communication with the network server; the user interface is mainly used for data interaction with the user; the processor and memory in the computer device of the present application can be set in the computer device, and the computer device calls the computer program product stored in the data storage module through the processor, and executes the root cause analysis method based on the comprehensive analysis framework provided in the embodiment of the present application.

[0061] Based on the hardware environment of the aforementioned embodiment, the embodiment of the present application provides a method for root cause analysis based on a comprehensive analysis framework, please refer to Figure 2 , Figure 2 The flowchart of the defect root cause analysis method is as follows:

[0062] (1) During the production process of panel products, defects of panel products are monitored, and defect data and processing data of panel products are timely obtained according to defect reports fed back by monitoring, and the defect data and processing data are pre-processed.

[0063] In the embodiment of the present invention, the defective data is not limited to the defective rate and category label obtained by the final quality inspection of the panel product. It can also be combined with business logic to flexibly define defective data for parameter anomalies in the production monitoring process, so as to realize the timely discovery and processing of production anomalies. Therefore, the relevant defective data can also be the monitoring data in the intermediate process. Specifically, the defective data includes but is not limited to: ① The ID of the defective product; ② The site where the defect is detected; ③ The number of defective products; ④ The input amount of products in the same batch; ⑤ The time when the defect is detected.

[0064] In the embodiment of the present invention, the processing data of the panel product includes but is not limited to production history data of defective products, intermediate sampling data, and intermediate measurement data.

[0065] In the embodiment of the present invention, the preprocessing includes but is not limited to removing null values, calculating defective rates, and removing abnormal values, etc. The defective rate is the proportion of defective products in the quality inspection results of a batch of panel products.

[0066] It should be emphasized that before performing root cause location analysis based on the embodiments of the present invention, the sites of the processing process can be filtered, including but not limited to: ① Based on business knowledge, remove measurement sites and sampling sites that are not related to processing in the production history data, and only retain processing sites and rework sites in the product history data; ② Obtain the results of important sites, and give priority to analyzing the root cause machines and equipment that are highly correlated with defects at these important sites.

[0067] (2) Comprehensively analyze defect data and processing data based on the comprehensive analysis framework to locate intermediate factors related to product defects.

[0068] In the traditional analysis process, different analysis methods such as defect density analysis, processing time analysis, and measurement value analysis are often separated. Therefore, the embodiment of the present invention performs comprehensive analysis based on a comprehensive analysis framework to solve the problem that the existing defect root cause location method is often isolated from different defect root cause analysis methods and cannot achieve comprehensive analysis and judgment.

[0069] In the embodiment of the present invention, the comprehensive analysis framework includes three layers: the top layer is the defective data of the panel products, which can be either numerical values ​​(such as defective rate, defective value) or category labels (such as whether it is normal, OK or NG); the second layer is the analysis means, including but not limited to defect density analysis, processing time analysis, and measurement value analysis, etc.; the third layer is the analysis path, including but not limited to forward root cause location analysis and direct root cause location analysis. Specifically, the third layer performs forward analysis based on the analysis results of the analysis means selected in the second layer or directly performs root cause location analysis based on the defective data of the top layer, and directly jumps to the next step, please refer to Figure 3, Figure 3 This is a schematic diagram of the analytical inference of the comprehensive analytical framework.

[0070] Specifically, the process of comprehensive analysis of defective data and processing data according to the comprehensive analysis framework is as follows:

[0071] (2.1) Select an analysis path based on the bad data and conduct forward root cause location analysis or direct root cause location analysis.

[0072] (2.2) Based on business needs, if forward-tracing root cause location analysis is used, the corresponding analysis method is selected according to the analysis object, and the production history data, intermediate sampling data and intermediate measurement data are correlated or sequence consistency coefficients are calculated based on the selected analysis method to locate the intermediate factors related to product defects, wherein the analysis methods include but are not limited to defect density analysis, processing time analysis and measurement value analysis.

[0073] Specifically, although there are differences in the data analyzed by different analysis methods such as defect density analysis, processing time analysis, and measurement value analysis, the analysis goals and essence are the same, which is to locate the factors that are highly correlated with product defects. Since the defect data of panel products can be either numerical or category label type, the two different data forms determine different analysis methods, such as

[0074] As shown in Table 1.

[0075]

[0076] Table 1

[0077] The calculation methods of the correlation coefficients in Table 1 include but are not limited to the Pearson correlation coefficient calculation method, the Kendall correlation coefficient calculation method, etc.; the calculation methods of the sequence consistency coefficient include but are not limited to the Spearman rank correlation coefficient calculation method, the Cohen's Kappa coefficient calculation method, etc.; the calculation methods of the feature importance include but are not limited to the Chi-square test analysis method, etc.

[0078] (2.3) If direct root cause location analysis is used, root cause location analysis is performed directly on the processing data based on the defect data to preliminarily locate the root cause site and root cause machine equipment.

[0079] (3) Perform root cause analysis on the processing data based on the defect data to preliminarily locate the root cause site and root cause machine equipment.

[0080] In the embodiment of the present invention, the defective data can be directly analyzed to find out that the defective heights of the processing sites, processing equipment and panel products in the production process are highly correlated, that is, the differences in the defective data of the panel products of different processing sites and processing equipment can be analyzed.

[0081] Specifically, due to the complexity of the production and processing of panel products, the entire production and processing process involves hundreds of processing sites, and each processing site often operates with multiple parallel processing equipment. A single device contains multiple parallel sub-equipment, and a sub-equipment often contains multiple parallel chambers. Multiple products are processed and transported in parallel chambers at the same time to improve efficiency. Therefore, when performing machine equipment analysis, it is generally necessary to first locate the root cause site of the defect, and then locate the machine equipment at the processing site.

[0082] Specifically, the process of performing root cause location analysis on processing data based on defective data is as follows:

[0083] (3.1) With respect to the defective data and production history data of a batch of panel products, the production history data are grouped according to the defective data of panel products produced by different processing equipment, so as to obtain the defective data of products in different equipment groups.

[0084] During the entire processing process, different processing sites are in an upstream-downstream relationship, while different equipment at the same processing site is in a parallel relationship, completing processing operations at the same time. If there is a significant difference in product defects between different parallel equipment at the processing site, it means that there is a strong correlation between the processing site and product defects, and it should be focused on; conversely, if there is no difference in product defects between different processing equipment at the processing site, it means that there is no difference between different processing equipment at the processing site, and the processing is relatively stable (it is almost impossible for all equipment at a site to have problems at the same time), then other suspicious sites can be considered.

[0085] Furthermore, it is necessary to locate and analyze the defective equipment under the suspicious site. Specifically, when locating processing equipment, users can locate the root cause of the defect at the equipment level according to business needs, or further fine-tune the location to the sub-equipment level, chamber level, etc., to achieve more fine-grained root cause investigation. Similarly, according to the dimensions of the equipment root cause location analysis selected by the user (such as locating the root cause at the equipment level, chamber level, etc.), the production history data and defective data are first grouped accordingly for parallel equipment. For example, if the root cause is located at the equipment level, each group contains the defective data of products passing through different parallel devices. Similarly, if the root cause is located at the chamber level, each group contains the defective data of products passing through different chambers.

[0086] (3.2) Perform a difference analysis on the defective data of products in different device groups to determine whether there are differences between the different grouped devices currently being analyzed.

[0087] In the embodiment of the present invention, statistical means are used to determine whether there is a significant difference in failure between different device groups, that is, the determination is made by calculating the p value of the statistical test. Specifically, the statistical test calculation method used includes but is not limited to: variance analysis method, weighted analysis method, etc.

[0088] In particular, if there is a bad data category label, the chi-square analysis method is used to calculate it to improve the accuracy of the analysis. That is, according to the p-value calculated by the significance and the preset significance threshold, the groups with significant differences can be determined. For example, if the calculated p-value is less than the preset significance level, it is considered that at least one group has a significant difference in the final product defect rate from other groups. The root cause of the product's defect is likely to be in this link. Other production inspection tools or means should be further combined to conduct in-depth inspections of each group at this level (such as different chambers or different equipment) in terms of the setting of production and processing parameters, the health status of processing equipment, etc. Whether there are problems. If the p-value is greater than the significance level, it is considered that different groups at this analysis level (such as different equipment) have no significant difference in the defect rate of the final product. It is recommended to consider other production and processing sites, or to analyze from other levels of production equipment (such as different chambers). The embodiment of the present invention selects corresponding statistical analysis methods according to the distribution of the data to be analyzed and the variance of the data, so as to better meet the data analysis needs in real application scenarios and make the analysis more accurate and reliable. At the same time, combined with the effect analysis theory, the differences of the links or factors with significant differences are further quantified. Compared with the traditional method of giving p-values ​​or significance judgment conclusions, the significant differences are quantified and multiple comparison tests are used to further analyze the differences in different factors, which helps to further determine the most likely root causes of the adverse effects and improve the accuracy and efficiency of the analysis.

[0089] (3.2) If there are differences between different device groups, locate the device in the group with the largest difference to obtain the preliminary located root cause site and root cause machine device.

[0090] In the above analysis process, if there is a group with a significant difference in product defect rate between the selected analysis dimension (equipment, chamber) and other groups, further analysis and calculation will be performed to determine the group with the most likely root cause of product defects in the analysis dimension. The size of the difference between different groups will be further quantified to locate the group with the largest difference from other groups.

[0091] It should be emphasized that traditional analysis methods are often based on hypothesis testing and can only determine that defects are significantly concentrated in a certain device grouping compared to other groups (that is, there are significant differences between the defects corresponding to a certain device grouping and those of other groups), but they fail to clearly indicate which device grouping it is.

[0092] Therefore, after determining that there are differences in product defective rates between different groups, it is necessary to further quantify the degree of difference between different groups, so as to locate and screen out the group with the greatest difference from other groups, thereby further reducing the scope of root cause screening.

[0093] The embodiment of the present invention uses a multiple comparison test method, adopts the form of a pairwise paired test, determines whether there is a significant difference between any two groups, and then calculates the difference effect size between the two groups with significant differences. Finally, for each group, the difference effect size between it and other groups is averaged, which is the average difference of the group relative to other groups. Then, the average difference is sorted to determine those processing groups (i.e., processing equipment, processing chambers) that are significantly different from other groups, providing a reference for further determining the scope of suspicious root cause screening. Specifically, there are many methods for pairwise multiple comparison tests, including but not limited to Tukey's honest significant difference test (applicable to the case where the data in each group meets the homogeneity of variance and normal distribution) and Duncan's test (applicable to the case of unequal variance). Preferably, the embodiment of the present invention uses Duncan's test to perform pairwise pairing for significance test, calculates the difference value between any two groups, and then obtains its average difference value for each group, and finally sorts the average difference value, so that the group with the largest difference in product defects from other groups can be screened out, so as to further troubleshoot the problem.

[0094] (4) Conduct root cause analysis on the processing data based on the intermediate factors related to product defects to further locate the root cause machine equipment.

[0095] In an embodiment of the present invention, after locating the intermediate link factors that are highly correlated with product defects, because there are upstream and downstream relationships in production, if there are problems with the results of the intermediate links (measurement, sampling, etc.), then the root cause of the problem can only be in the upstream links. Therefore, it is necessary to trace back further to find which previous production and processing equipment has affected the values ​​of the relevant factors in the intermediate links, that is, to analyze the differences in the values ​​of the intermediate link factors of different production equipment.

[0096] For example, based on the conclusion of defect density analysis, it is necessary to further trace which machines and equipment in the previous production links have affected the defect density values ​​of the intermediate links. The root cause location analysis of related processes is the same.

[0097] Specifically, the process of conducting root cause location analysis on processing data based on intermediate factors related to product defects is as follows:

[0098] (4.1) Group the production history data according to the intermediate factors related to product defects to obtain the defect data of products in different equipment groups.

[0099] (4.2) Perform difference analysis on the defective data of products in different equipment groups to determine whether there are differences between the different grouped devices currently analyzed.

[0100] (4.3) If there are differences between different device groups, locate the device in the group with the largest difference to obtain the root cause device for further location.

[0101] In an embodiment of the present invention, the principle of performing root cause location analysis on processing data based on intermediate link factors related to product defects is the same as the principle of performing root cause location analysis on processing data based on defective data, except that the defective data is replaced by intermediate link factors related to product defects. Therefore, the relevant steps of performing root cause location analysis on processing data based on intermediate link factors related to product defects can refer to the steps of performing root cause location analysis on processing data based on defective data.

[0102] (5) Perform a fusion judgment on the initially located root cause site and root cause equipment as well as the further located root cause equipment to obtain a comprehensive root cause location result.

[0103] In the embodiment of the present invention, the process of fusion judgment is as follows:

[0104] (5.1) A weighted average algorithm or a counting algorithm is used to comprehensively rank the initially located root cause sites and root cause equipment and the further located root cause equipment, and the root cause sites and root cause equipment are comprehensively located based on the comprehensive ranking.

[0105] (5.2) Screen and process the root cause sites and root cause equipment of the comprehensive positioning to obtain the comprehensive root cause positioning results.

[0106] Specifically, the embodiment of the present invention first uses a counting method to assign points to the root cause rankings given. For example, if a total of N defective root causes are given, the first-ranked root cause scores N-1, the second-ranked score is N-2, and so on, the last-ranked candidate root cause scores 0; then for each defective root cause, its scores in different rankings are accumulated and sorted to obtain the most likely defective root cause. Based on the defective root cause processing sites and defective processing equipment located after comprehensive judgment, users can conduct further targeted analysis and investigation, and make relevant adjustments in a timely manner to quickly adjust defects and improve yields.

[0107] In summary, the present invention effectively combines multiple analysis methods such as defect density analysis, processing time analysis, and measurement value analysis to locate the root cause of processing equipment based on a comprehensive analysis framework, which can effectively improve the accuracy and effectiveness of equipment root cause location. The comprehensive analysis framework has strong flexibility. Users can flexibly select analysis methods and analysis paths according to business needs to achieve arbitrary combinations, which solves the problem that different defect root cause analysis methods are isolated and cannot achieve comprehensive analysis and judgment.

[0108] In addition, in one embodiment, based on the same inventive concept as the above embodiment, the embodiment of the present invention provides a root cause analysis system based on a comprehensive analysis framework, and the system corresponds to the method of embodiment 1 one by one, please refer to Figure 4 , Figure 4 The structure block diagram of the defect root cause analysis system is as follows:

[0109] A data acquisition unit, the data acquisition unit is used to acquire defect data and processing data of panel products;

[0110] A comprehensive analysis unit, which is used to conduct a comprehensive analysis on the defect data and the processing data according to a comprehensive analysis framework to locate the intermediate factors related to the product defect;

[0111] A primary positioning unit, the primary positioning unit is used to perform root cause positioning analysis on the processing data according to the defect data, so as to preliminarily locate the root cause site and the root cause machine equipment;

[0112] A secondary positioning unit, which is used to perform root cause positioning analysis on processing data based on intermediate factors related to product defects, so as to further locate the root cause machine equipment;

[0113] The fusion judgment unit is used to perform fusion judgment on the initially located root cause site and root cause machine equipment and the further located root cause machine equipment to obtain a comprehensive root cause positioning result.

[0114] It should be noted that each unit in the defect root cause analysis system based on the comprehensive analysis framework in this embodiment corresponds one-to-one to each step in the defect root cause analysis method based on the comprehensive analysis framework in the aforementioned embodiment. Therefore, the specific implementation method and technical effects achieved by this embodiment can refer to the implementation method of the aforementioned defect root cause analysis method based on the comprehensive analysis framework, which will not be repeated here.

[0115] In addition, in one embodiment, the present application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory, and the computer program implements the method in the aforementioned embodiment when executed by the processor.

[0116] In addition, in one embodiment, the present application further provides a computer storage medium, on which a computer program is stored, and the computer program implements the method in the aforementioned embodiment when executed by a processor.

[0117] In some embodiments, the computer readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or various devices including one or any combination of the above memories. The computer may be various computing devices including intelligent terminals and servers.

[0118] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0119] As an example, executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0120] By way of example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed across multiple sites and interconnected by a communication network.

[0121] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0122] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0123] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0124] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A root cause analysis method based on a comprehensive analysis framework, characterized in that: The method comprises the following steps: Obtain defect data and processing data of panel products; Comprehensively analyze defective data and processing data according to the comprehensive analysis framework to locate the intermediate factors related to product defects; Conduct root cause analysis on the processing data based on the defect data to preliminarily locate the root cause site and root cause equipment; Conduct root cause analysis on processing data based on intermediate factors related to product defects to further locate the root cause machine equipment; The initially located root cause sites and root cause equipment as well as the further located root cause equipment are integrated and judged to obtain a comprehensive root cause location result.

2. The root cause analysis method based on a comprehensive analysis framework according to claim 1 is characterized in that: The defective data includes the defective rate and category labels of defective products; the processing data of the panel products includes the production history data, intermediate sampling data and intermediate measurement data of the defective products.

3. The root cause analysis method based on a comprehensive analysis framework according to claim 1, characterized in that: After obtaining the defective data and processing data of the panel product, the defective data and processing data need to be preprocessed; wherein the preprocessing includes removing null values, calculating the defective rate and removing abnormal values.

4. The root cause analysis method based on a comprehensive analysis framework according to claim 2 is characterized in that: The process of comprehensive analysis of defective data and processing data according to the comprehensive analysis framework is as follows: Conduct forward-looking root cause location analysis or direct root cause location analysis based on bad data; If the forward root cause location analysis is adopted, the corresponding analysis means are selected according to the analysis object, and the correlation coefficient or sequence consistency coefficient is calculated for the production history data, intermediate sampling data and intermediate measurement data according to the selected analysis means to locate the intermediate link factors related to the defective products, wherein the analysis means include defect density analysis, processing time analysis and measurement value analysis; If direct root cause location analysis is used, root cause location analysis is performed directly on the processing data based on the defective data to preliminarily locate the root cause site and root cause machine equipment.

5. The method for root cause analysis based on a comprehensive analysis framework according to claim 4, characterized in that: The process of root cause location analysis of processing data based on defective data is as follows: Group the production history data according to the defect data to obtain the defect data of products in different equipment groups; Perform difference analysis on the defective data of products in different device groups to determine whether there are differences between the devices in different groups currently being analyzed; If there are differences between different device groups, locate the device in the group with the largest difference to obtain the preliminary located root cause site and root cause machine device.

6. The method for root cause analysis based on a comprehensive analysis framework according to claim 2, characterized in that: The process of root cause location analysis of processing data based on intermediate factors related to product defects is as follows: Group production history data according to intermediate factors related to product defects to obtain defect data of products in different equipment groups; Perform difference analysis on the defective data of products in different equipment groups to determine whether there are differences between the different grouped equipment currently being analyzed; If there are differences between different device groups, locate the device in the group with the largest difference to obtain the root cause device for further location.

7. The root cause analysis method based on a comprehensive analysis framework according to claim 1 is characterized in that: The process of merging and judging the initially located root cause site and root cause equipment and the further located root cause equipment is as follows: A weighted average algorithm or a counting algorithm is used to comprehensively rank the initially located root cause sites and root cause equipment and the further located root cause equipment, and the root cause sites and root cause equipment are comprehensively located according to the comprehensive ranking; The root cause sites and root cause equipment of the comprehensive positioning are screened and processed to obtain the comprehensive root cause positioning results.

8. A root cause analysis system based on a comprehensive analysis framework, characterized in that: The system comprises: A data acquisition unit, the data acquisition unit is used to acquire defect data and processing data of panel products; A comprehensive analysis unit, which is used to conduct a comprehensive analysis on the defect data and the processing data according to a comprehensive analysis framework to locate the intermediate factors related to the product defect; A primary positioning unit, the primary positioning unit is used to perform root cause positioning analysis on the processing data according to the defect data, so as to preliminarily locate the root cause site and the root cause machine equipment; A secondary positioning unit, which is used to perform root cause positioning analysis on processing data based on intermediate factors related to product defects, so as to further locate the root cause machine equipment; The fusion judgment unit is used to perform fusion judgment on the initially located root cause site and root cause machine equipment and the further located root cause machine equipment to obtain a comprehensive root cause positioning result.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for analyzing the root cause of a defect based on a comprehensive analysis framework as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for analyzing the root cause of a defect based on a comprehensive analysis framework as described in any one of claims 1 to 7 is implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for analyzing the root cause of a defect based on a comprehensive analysis framework as described in any one of claims 1 to 7 is implemented.

Images