Systems and methods for determining relationships between defects

Abstract

Systems and methods are provided for identifying relationships between defects. The system may obtain defect items and associated information. Defect items may be compared to one another based on their attributes to determine how related they are. According to the comparisons, defect items may be grouped together into issue items for further analysis by a user. The system may further update a defect comparison model according to user interaction with defect items.

1 . A system comprising:
 at least one processor; and   at least one memory storing machine-readable instructions, wherein the at least one processor is configured to access the at least one memory and execute the machine-readable instructions to cause the system to:
 obtain a first defect data object and a second defect data object from a database; 
 determine, using a machine learning model, a defect similarity between the first defect data object and the second defect data object, wherein the defect similarity is determined based on a comparison metric with which to determine the defect similarity, wherein the comparison metric is based on a feature weight related to an attribute of a manufacturing or an industrial process; 
 determine, using the machine learning model, that the first defect data object and the second defect data object are related based on the comparison metric; 
 generate and store an issue data object comprising the first defect data object and the second defect data object in the database, the issue data object indicative of a diagnosis of a defect common to the first defect data object and the second defect data object; and 
 obtain one or more additional defect data objects in which respective comparison value scores indicative of a degree to which the additional defect data objects correspond to or are related to the issue data object satisfy a criterion, or remove one or more existing defect data objects in which respective comparison value scores indicative of a degree to which the existing defect data objects correspond to or are related to the issue data object or other existing defect data objects within the issue data objects fail to satisfy the criterion. 
   
     
     
         2 . The system of  claim 1 , wherein the machine-readable instructions, when executed, further cause the system to:
 adjust the feature weight or a different feature weight based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         3 . The system of  claim 1 , wherein the machine-readable instructions, when executed, further cause the system to:
 generate or provide training data for the machine learning model based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         4 . The system of  claim 1 , wherein the machine-readable instructions, when executed, further cause the system to:
 obtain an approval of the issue data object; and   generate or provide training data for the machine learning model based on the approval of the issue data object.   
     
     
         5 . The system of  claim 1 , wherein the determining of the defect similarity is based on the feature weight and an other feature weight corresponding to an other attribute, wherein the attribute has a higher feature weight compared to the other attribute and the feature weight is evaluated prior to the other feature weight. 
     
     
         6 . The system of  claim 5 , wherein the machine-readable instructions, when executed, further cause the system to:
 halt an evaluation of the other feature weight if the evaluation of the feature weight fails to satisfy the criterion or a different criterion.   
     
     
         7 . The system of  claim 1 , wherein the machine-readable instructions, when executed, further cause the system to:
 train the machine learning model based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         8 . The system of  claim 1 , wherein the machine-readable instructions, when executed, further cause the system to:
 generate a graph of the first defect data object, the second defect data object, and one or more other defect data objects within the issue data object, wherein the graph indicates a degree of connectedness among the first defect data object, the second defect data object, and the one or more other defect data objects.   
     
     
         9 . A computer-implemented method performed on a computer system having at least one physical processor programmed with machine-readable instructions that, when executed by the at least one physical processor, cause the computer system to perform the method, the method comprising:
 obtaining a first defect data object and a second defect data object from a database;   determining, using a machine learning model, a defect similarity between the first defect data object and the second defect data object, wherein the defect similarity is determined based on a comparison metric with which to determine the defect similarity, wherein the comparison metric is based on a feature weight related to an attribute of a manufacturing or an industrial process;   determining, using the machine learning model, that the first defect data object and the second defect data object are related based on the comparison metric;   generating and storing an issue data object comprising the first defect data object and the second defect data object in the database, the issue data object indicative of a diagnosis of a defect common to the first defect data object and the second defect data object; and   obtaining one or more additional defect data objects in which respective comparison value scores indicative of a degree to which the additional defect data objects correspond to or are related to the issue data object satisfy a criterion, or remove one or more existing defect data objects in which respective comparison value scores indicative of a degree to which the existing defect data objects correspond to or are related to the issue data object or other existing defect data objects within the issue data objects fail to satisfy the criterion.   
     
     
         10 . The computer-implemented method of  claim 9 , further comprising:
 adjusting the feature weight or a different feature weight based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         11 . The computer-implemented method of  claim 9 , further comprising:
 generating or providing training data for the machine learning model based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         12 . The computer-implemented method of  claim 9 , further comprising:
 obtaining an approval of the issue data object; and   generating or providing training data for the machine learning model based on the approval of the issue data object.   
     
     
         13 . The computer-implemented method of  claim 9 , wherein the determining of the defect similarity is based on the feature weight and an other feature weight corresponding to an other attribute, wherein the attribute has a higher feature weight compared to the other attribute and the feature weight is evaluated prior to the other feature weight. 
     
     
         14 . The computer-implemented method of  claim 13 , further comprising:
 halting an evaluation of the other feature weight if the evaluation of the feature weight fails to satisfy the criterion or a different criterion.   
     
     
         15 . The computer-implemented method of  claim 9 , further comprising:
 training the machine learning model based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         16 . A non-transitory computer readable medium comprising instructions that, when executed, cause one or more processors to perform a method, the method comprising:
 obtaining a first defect data object and a second defect data object from a database;   determining, using a machine learning model, a defect similarity between the first defect data object and the second defect data object, wherein the defect similarity is determined based on a comparison metric with which to determine the defect similarity, wherein the comparison metric is based on a feature weight related to an attribute of a manufacturing or an industrial process;   determining, using the machine learning model, that the first defect data object and the second defect data object are related based on the comparison metric;   generating and storing an issue data object comprising the first defect data object and the second defect data object in the database, the issue data object indicative of a diagnosis of a defect common to the first defect data object and the second defect data object; and   obtaining one or more additional defect data objects in which respective comparison value scores indicative of a degree to which the additional defect data objects correspond to or are related to the issue data object satisfy a criterion, or removing one or more existing defect data objects in which respective comparison value scores indicative of a degree to which the existing defect data objects correspond to or are related to the issue data object or other existing defect data objects within the issue data objects fail to satisfy the criterion.   
     
     
         17 . The non-transitory computer readable medium of  claim 16 , wherein the method further comprises:
 adjusting the feature weight or a different feature weight based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         18 . The non-transitory computer readable medium of  claim 16 , wherein the method further comprises:
 generating or providing training data for the machine learning model based on the generated issue data object, the additional defect data objects, or the removed existing defect data objects.   
     
     
         19 . The non-transitory computer readable medium of  claim 16 , wherein the method further comprises:
 obtaining an approval of the issue data object; and   generating or providing training data for the machine learning model based on the approval of the issue data object.   
     
     
         20 . The non-transitory computer readable medium of  claim 16 , wherein the determining of the defect similarity is based on the feature weight and an other feature weight corresponding to an other attribute, wherein the attribute has a higher feature weight compared to the other attribute and the feature weight is evaluated prior to the other feature weight.