Defect Analysis Data Platform for Multi-Source Semiconductor Yield Tracing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The manufacturing of semiconductor electronics faces challenges in tracing and analyzing defects due to the complexity of the manufacturing process, relying heavily on manual data sorting and experience-based analysis, which is inefficient and time-consuming.
Innovation Solution
A data management platform and intelligent defect analysis system that extracts, transforms, and loads data from various sources into a distributed data storage, using ETL modules, data lakes, warehouses, and data marts, enabling algorithmic analysis and visualization of defects across multiple data sources, including biographical, parameter, and defect information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual data sorting and experience-based analysis are used, then analysis flexibility is maintained, but analysis efficiency and speed are significantly reduced
Solution Approach 1:
The data management system is segmented into multiple layers including data source layer, data access layer, data processing layer, and application layer. Each layer performs specific functions independently, allowing complex data management tasks to be broken down into manageable components that can be processed efficiently by algorithm servers.
Solution Approach 2:
The patent introduces a data access layer as an intermediary between data sources and algorithm servers. This layer provides standardized interfaces and data preprocessing, enabling algorithm servers to focus on defect analysis without dealing with raw data complexity, thus improving efficiency while managing system complexity.
2Loss of information
If data from multiple sources are integrated into a unified system, then defect analysis comprehensiveness is improved, but data processing complexity increases
Solution Approach 1:
The patent organizes multi-source data into a multi-layered hierarchical structure rather than a flat integration. Data from diverse sources (production data, inspection data, test data) are integrated vertically across layers with standardized schemas at each level, preserving information completeness while managing integration complexity through structured organization.
Solution Approach 2:
The data access layer provides universal interfaces that can access and standardize data from multiple different sources. This layer performs data cleansing, transformation, and validation uniformly across all data sources, enabling comprehensive defect information integration without proportionally increasing processing complexity.
3Loss of time
If real-time data processing is implemented, then defect identification speed is improved, but computational resource requirements increase
Solution Approach 1:
The system performs preliminary data processing including cleansing, standardization, and validation in the data access and processing layers before data reaches algorithm servers. This preprocessing reduces the computational burden on algorithm servers during real-time defect identification, enabling faster response with reduced resource consumption during critical analysis phases.
Solution Approach 2:
The patent implements selective real-time processing where only critical defect-related data undergoes intensive real-time analysis, while other data is processed asynchronously or in batches. This partial real-time approach maintains fast defect identification for critical issues while managing overall computational resource consumption.
Data Source
AI summary
A data management platform for intelligently managing data is provided. The data management platform includes an ETL module configured to extract, cleanse, transform, or load data; a data lake configured to store a first group of data formed by extracting raw data from a plurality of data sources by the ETL module; a data warehouse configured to store a second group of data formed by cleansing and standardizing on the first group of data; a general data layer configured to store a third group of data formed by subjecting the second group of data to data fusion; and a data mart configured to store a fourth group of data formed by transforming the third group of data by the ETL module. The general data layer is a distributed data storage storing information available for querying. The data mart is a database of NoSQL type storing information available for computational processing.