CRAI Dependency Graphs for Faster Semiconductor Defect Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern semiconductor manufacturing generates vast amounts of data with natural dependencies, making it time-consuming to detect and diagnose manufacturing defects and perturbations caused by malicious actors.
Innovation Solution
A computer-implemented method and system using Causal Relational Artificial Intelligence (CRAI) and a risk framework to predict causal effects by identifying interventions, collecting process dependency data, creating intervention models, and training CRAI models to determine intervention efficacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional defect detection methods are used in semiconductor manufacturing, then comprehensive analysis can be performed on manufacturing data, but the process becomes extremely time-consuming and labor-intensive
Solution Approach 1:
The patent introduces an intermediary system comprising a dependency graph and causal relational AI model that mediates between raw manufacturing data and defect diagnosis. The dependency graph serves as an intermediary structure that pre-organizes relationships between manufacturing parameters, processes, and outcomes, allowing the CRAI model to efficiently query causal relationships without exhaustive analysis of all data points.
Solution Approach 2:
The patent performs preliminary action by pre-construction of the dependency graph and training of the CRAI model before actual defect detection occurs. The dependency graph is built in advance to capture all causal relationships in the manufacturing process, and the CRAI model is pre-trained on historical data, enabling rapid inference when defects are detected without performing time-consuming analysis during the actual diagnosis phase.
2Reliability
If comprehensive data analysis is performed to detect manufacturing defects, then detection accuracy improves, but the complexity of the analysis process increases
Solution Approach 1:
The patent segments the complex analysis system into distinct modular components: (1) data collection module that gathers manufacturing parameters, (2) dependency graph construction module that organizes relationships, (3) CRAI model training module that learns causal patterns, and (4) inference module that performs defect diagnosis. Each module handles a specific aspect of the analysis, reducing overall system complexity while maintaining comprehensive analysis capability.
Solution Approach 2:
The dependency graph acts as an intermediary structure that simplifies the complexity of analyzing vast manufacturing data. Instead of directly analyzing all raw data points, the system uses the pre-constructed dependency graph to represent causal relationships, which the CRAI model then queries efficiently. This intermediary representation reduces the computational complexity from exponential to polynomial time.
3Measurement precision
If traditional methods are used to diagnose manufacturing perturbations, then thorough investigation can be conducted, but the effort and resources required increase significantly
Solution Approach 1:
The patent replaces the mechanical/manual process of defect diagnosis with an automated intelligent system. Instead of human experts manually analyzing manufacturing data and determining causal relationships, the system uses CRAI models that automatically infer causal relationships from the dependency graph and historical data, dramatically increasing diagnosis throughput while maintaining or improving accuracy.
Solution Approach 2:
The system implements self-service by enabling the manufacturing process to diagnose its own defects autonomously. The CRAI model continuously monitors manufacturing parameters, automatically detects anomalies, identifies root causes through the dependency graph, and provides diagnosis results without requiring external human intervention for each defect case, thereby maximizing diagnosis throughput.
Data Source
AI summary
In an approach to CRAI and risk framework for manufacturing applications, a computer-implemented method for causal effect prediction includes identifying, by one or more computer processors, an intervention, where the intervention is selected from the group consisting of threats, failures, corrections, and relevant outputs; collecting, by the one or more computer processors, process dependency data; creating, by the one or more computer processors, an intervention model; combining, by the one or more computer processors, the process dependency data and the intervention model to create a combined process dependency graph; training, by the one or more computer processors, a causal relational artificial intelligence (CRAI) model; and determining, by the one or more computer processors, an estimate of an intervention efficacy.