Automated Defect Analysis for Semiconductor Manufacturing Machines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual analysis of semiconductor manufacturing processes is inefficient and inaccurate in identifying problematic manufacturing machines and processing workstations causing defects, due to reliance on human skill and lack of systematic data analysis.
Innovation Solution
A method and electronic device that collect and analyze processing history, processing factor, and quality information from databases to determine problem manufacturing machines and influencing factors, using modules for data collection, storage, and statistical testing to identify defective machines and factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If manual analysis is used to identify problematic manufacturing machines and processing workstations, then the method is simple to implement, but the efficiency and accuracy of defect analysis deteriorates
Solution Approach 1:
The patent replaces manual mechanical analysis with an automated computer-based system that collects processing history data, performs statistical testing (chi-square tests), and automatically identifies problematic machines and workstations. This substitution of manual analysis with automated computational methods directly resolves the contradiction by dramatically improving analysis efficiency while maintaining ease of implementation through software automation.
Solution Approach 2:
The system performs self-service by automatically collecting data from databases, conducting statistical analyses, and generating identification results without requiring manual intervention for each analysis case. The automated pipeline handles data collection, processing, and interpretation independently, thereby improving productivity while keeping the system simple to deploy.
2Device complexity
If manual analysis is used to identify problematic manufacturing machines and processing workstations, then the system complexity is low, but the accuracy of analysis deteriorates due to dependence on analyst skill
Solution Approach 1:
The patent replaces subjective manual analysis with objective computer-based statistical testing. The system automatically performs chi-square tests and other statistical analyses to identify correlations between processing factors and defects, eliminating dependence on individual analyst skills. This substitution maintains relatively low system complexity while dramatically improving measurement precision and reproducibility.
Solution Approach 2:
The system transforms qualitative manual judgment into quantitative statistical parameters by collecting numerical processing history data and applying statistical tests. This parameter transformation from subjective assessment to objective numerical analysis improves accuracy while keeping the overall system structure relatively simple and manageable.
3Measurement precision
If comprehensive data collection and statistical testing are performed to identify problem machines, then the accuracy of identification improves, but the device complexity and time consumption increases
Solution Approach 1:
The patent creates a universal automated system that handles multiple functions: data collection from databases, statistical testing, and identification of problematic machines and workstations. This multi-functional integrated system improves identification accuracy while managing complexity through consolidation rather than separate specialized systems for each function.
Solution Approach 2:
The system replaces complex manual data collection and analysis processes with automated computational methods. By substituting manual efforts with computer-based data gathering and statistical testing, the system achieves high identification accuracy while the complexity is managed through automation rather than requiring complex manual procedures.
4Measurement precision
If comprehensive data collection and statistical testing are performed to identify problem machines, then the accuracy of identification improves, but the time consumption increases
Solution Approach 1:
The patent replaces time-consuming manual analysis with automated computer-based statistical testing. The system automatically collects processing history data and performs chi-square tests and other statistical analyses, achieving high identification accuracy while significantly reducing the time required compared to manual methods. The automation handles the time-consuming data processing tasks efficiently.
Solution Approach 2:
The system enables continuous automated data collection and analysis, performing statistical testing continuously on processing history data without interruption. This continuous automated operation maintains high identification accuracy while reducing overall time consumption by eliminating gaps between data collection, analysis, and result generation that occur in manual processes.
Data Source
AI summary
A method for finding manufactured or to-be-manufactured products for defects includes obtaining basic information on processing history of products passed by each manufacturing machine at each processing workstation, and obtaining processing factor information of same, where each product is passed by a defect detection workstation after the product is passed by at least one processing workstation. The method includes obtaining quality information detected by each defect detection workstation. The method determines one or more problem manufacturing machines at one or more problem processing workstations according to the basic information on processing history and the quality information of the products. The method further determines one or more processing factors which influence the problem manufacturing machines according to the processing factor information of each manufacturing machine and the quality information of the products. A related electronic device and non-transitory storage medium are also provided.