Quality Data Analysis Using ML for Defect Factor Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional quality data analysis systems face challenges in efficiently analyzing product quality data due to process factor bias, requiring effective measures to select and adjust process factors that cause defects, and existing machine learning approaches are time-consuming and require expertise, limiting their practical application.
Innovation Solution
A machine learning-based quality data analysis system that trains an inference model using accumulated quality data, selects adjustment process factors, and adjusts quality control standards based on optimal factor values, utilizing a User Interface for data selection and visualization, and employs algorithms like XGBoost and LightGBM to minimize defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning is used for quality data analysis, then analysis accuracy is improved, but analysis time and complexity increase
Solution Approach 1:
The system performs preliminary actions by pre-training the machine learning inference model with historical quality data and process factor information before actual quality analysis is needed. This pre-processing enables the model to be ready for rapid deployment, reducing the time required for real-time quality analysis while maintaining high accuracy through pre-learned patterns and correlations.
2Reliability
If machine learning models are trained and deployed, then quality analysis capability is improved, but system complexity and expert dependency increase
Solution Approach 1:
The system enables self-service by designing an automated workflow where the machine learning model independently performs quality data analysis without requiring expert intervention. The inference model automatically processes quality data, correlates process factors with defects, and generates analysis results, allowing non-expert users to perform sophisticated quality analysis through a simplified interface that hides the underlying complexity.
3Manufacturing precision
If process factors are adjusted based on analysis, then defect reduction is improved, but data integrity and analysis reliability worsen due to process factor bias
Solution Approach 1:
The system implements feedback by continuously monitoring the relationship between process factors and quality outcomes, using the machine learning model to identify biased process factor values that may have been manually adjusted without proper documentation. The model provides feedback on which process factors are most strongly correlated with defects, enabling data-driven adjustments that improve manufacturing precision while maintaining data integrity through automated correlation analysis rather than relying on potentially biased manual adjustments.
Data Source
AI summary
A quality data analysis apparatus and method for reducing time for product quality analysis and the quality cost by reducing the occurrence of product defects, the apparatus includes an input configured to obtain quality data on a product for process factors occurring in a production of the product, a data pre-processor to pre-process the quality data by encoding the process factors for each data types and setting the process factors that are lost, to a preset value, a determiner configured to determine whether the product is acceptable based on the process factors using machine learning, a data visualizer configured to generate an analysis report on a quality of the product based on the process factors and the determination, and a trainer configured to train the machine learning model using the quality data for learning and a first label relevant to the quality data for learning.