Query Extraction Engine for Semiconductor Data Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The semiconductor manufacturing process, which involves multiple discrete steps across different facilities, faces challenges in tracking semiconductor devices due to the vast amount of data generated, leading to difficulties in querying and extracting relevant information efficiently, particularly due to the use of wildcards in data queries which consume significant processing resources and time.
Innovation Solution
A method and system that define a general query with ambiguous parameters, utilize metadata to identify candidate values, and generate atomic queries to extract data from multiple data stores, reducing ambiguity and processing time by employing a modular extraction engine to execute these queries in parallel.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If wildcards are used in data queries to increase flexibility, then query adaptability is improved, but processing time and resource consumption increase significantly
Solution Approach 1:
The patent segments a complex wildcard query into multiple atomic queries, each with specific parameter values. The query processing system divides the original query containing wildcards into several simpler queries that can be executed efficiently in parallel, resolving the contradiction between query flexibility and processing speed.
Solution Approach 2:
The system performs preliminary actions by pre-compiling and caching metadata about available data parameters and their possible values. This preliminary preparation enables the query system to quickly generate atomic queries without performing expensive full-scan operations during actual query execution.
2Loss of information
If comprehensive data is collected from multiple facilities to improve tracking accuracy, then data completeness is improved, but processing complexity increases
Solution Approach 1:
The patent segments the data collection process by facility or data source, creating separate atomic queries for each data repository. This segmentation maintains data completeness across multiple facilities while simplifying the processing complexity by handling each data source independently through standardized query templates.
Solution Approach 2:
The system employs a universal query interface that can access multiple different data stores and facilities through a common methodology. This universal approach maintains data completeness from diverse sources while reducing processing complexity by applying the same atomic query generation logic across all data repositories.
3Measurement precision
If detailed metadata is analyzed to resolve ambiguous parameters, then query precision is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by analyzing only the necessary portion of metadata required to resolve ambiguous parameters in the query. Instead of performing exhaustive metadata analysis, the system selectively processes only the metadata relevant to the specific ambiguous parameters, maintaining precision while improving efficiency.
Solution Approach 2:
The system performs preliminary metadata analysis and caching, pre-processing metadata to identify and store information about parameter relationships and possible values. This preliminary action enables faster resolution of ambiguous parameters during actual query execution without sacrificing precision.
Data Source
AI summary
A method includes defining a general query for extracting data from at least one data store operable to store workpiece data associated with the processing of workpieces in a manufacturing system. The general query specifies at least one ambiguous parameter having a plurality of potential values. Metadata associated with the workpiece data is accessed. The metadata is employed to identify a plurality of candidate values for the at least one ambiguous parameter. A plurality of atomic queries is generated. Each atomic query is associated with one of the candidate values. The plurality of atomic queries is executed to extract data from the at least one data store and generate an output report including the extracted data.