Manufacturing Defect Factor Search Using Mixture Distribution Components
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods fail to precisely search for manufacturing defect factors in manufacturing lines where multiple devices perform the same task, leading to variations in data and making it difficult to identify quality defects, especially with small production volumes and complex statistical distributions.
Innovation Solution
A method involving the classification of manufacturing monitoring data into non-defective and defective products, estimation of mixture distribution functions, and resolution into components to generate a list of items strongly related to manufacturing quality defects, using a manufacturing defect factor searching apparatus.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional statistical distribution functions are used to analyze manufacturing monitoring data, then the analysis method is simple, but the precision of defect factor identification deteriorates when multiple manufacturing devices are involved
Solution Approach 1:
The patent segments the complex manufacturing monitoring data into multiple components using mixture distribution functions. Each component represents a specific manufacturing device or process stage, allowing individual analysis of data from multiple devices while maintaining overall system visibility. This segmentation resolves the contradiction by breaking down the complex multi-device data distribution into manageable, analyzable segments.
Solution Approach 2:
The patent changes the statistical parameters used for analysis by employing mixture distribution functions with multiple components, each having its own parameters. This allows the system to adapt to the specific data characteristics of different manufacturing devices, improving defect factor identification precision while accounting for the complexity introduced by multiple devices through parameter customization.
2Measurement precision
If stratification processing is performed to divide manufacturing monitoring data by manufacturing attribute elements, then defect factor search precision improves, but the data accumulation time and response speed deteriorate
Solution Approach 1:
The patent performs preliminary classification of manufacturing monitoring data into defective and non-defective sets based on inspection results before conducting statistical analysis. This preliminary action prepares the data in advance, enabling faster and more precise defect factor search when needed, without requiring extensive additional data accumulation time.
Solution Approach 2:
The patent changes the approach from traditional stratification by manufacturing attributes to stratification by defect status (defective vs. non-defective). This parameter change enables the system to achieve high precision defect factor search while reducing data accumulation requirements, as the classification is based on clear defect outcomes rather than requiring long-term accumulation of attribute-based stratified data.
3Measurement precision
If mixture distribution functions are estimated and resolved into components, then defect sensitivity identification precision improves, but the computational complexity increases
Solution Approach 1:
The patent segments the mixture distribution function into multiple components, where each component corresponds to a specific defect source or manufacturing condition. This segmentation enables precise identification of defect sensitivity by analyzing each component separately, while the modular nature of component analysis keeps computational complexity manageable through systematic breakdown.
Solution Approach 2:
The patent applies partial action by focusing the component resolution on the most significant components that contribute to defectiveness. Rather than analyzing all possible components in equal detail, the system identifies and prioritizes the key components that have the greatest impact on defect sensitivity, reducing unnecessary computational complexity while maintaining high identification precision.
Data Source
AI summary
A manufacturing defect factor searching method includes: classifying manufacturing monitoring data into a set of non-defective products having an inspection result indicating a non-defective product and a set of defective products having the inspection result indicating a defective product, in accordance with a correspondence relationship between the manufacturing monitoring data and product inspection data indicating the inspection result of the product manufactured in the manufacturing line, the manufacturing monitoring data being collected from a manufacturing line of a product and being multivariate; estimating, for each item of the manufacturing monitoring data, a mixture distribution function approximating to a statistical distribution of each of the set of non-defective products and the set of defective products; resolving the mixture distribution function into components; and generating a list of items including a resolved component having a correlation with a manufacturing quality defect from among items of the manufacturing monitoring data.