Causal Graph Diagnostics for Multi-Module Root Cause Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lithographic apparatus diagnostics face challenges due to high complexity and the scarcity of experts with detailed system dynamics knowledge, leading to time-consuming root-cause identification and non-scalable diagnostic solutions that are prone to information loss and require explicit encoding of domain knowledge.
Innovation Solution
A computer-implemented method using a causal graph to generate a reasoning tool that augments diagnostics knowledge based on historical relations, statistical, and causal characteristics, enabling the identification of the most likely root-cause module by analyzing sensor data and expert knowledge, even for non-domain experts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expert knowledge is used to diagnose system problems, then diagnostic accuracy is improved, but diagnostic time increases and scalability decreases
Solution Approach 1:
The system performs preliminary encoding of expert diagnostic knowledge into structured causal models and diagnostic rules during system design. This pre-processing of knowledge allows the diagnostic system to quickly retrieve and apply established diagnostic logic without requiring real-time expert intervention, thereby reducing diagnostic time while maintaining accuracy.
Solution Approach 2:
The patent introduces an intelligent diagnostic system as an intermediary between experts and the lithographic apparatus. This intermediary automatically collects sensor data, applies encoded diagnostic knowledge, and generates diagnoses, eliminating the need for direct expert involvement in routine diagnostics while preserving diagnostic quality.
2Measurement precision
If custom diagnostic solutions are created for each system, then diagnostic accuracy is improved, but scalability and adaptability decrease
Solution Approach 1:
The patent creates a universal diagnostic platform that can be applied across multiple lithographic apparatus systems. The system uses standardized sensor interfaces, common diagnostic models, and reusable knowledge bases that can adapt to different system configurations through parameter adjustment rather than complete re-development, enabling both accuracy and scalability.
Solution Approach 2:
The diagnostic system allows customization through parameter adjustment rather than structural modification. By changing operational parameters, sensor thresholds, and diagnostic rules within a unified framework, the system can adapt to different lithographic apparatus configurations while maintaining the same core diagnostic engine, thus achieving scalability with adaptability.
3Reliability
If domain knowledge is implicitly encoded in diagnostic algorithms, then diagnostic capability is improved, but collaboration and verification become difficult
Solution Approach 1:
The patent segments domain knowledge into distinct, organized components including causal models, diagnostic rules, sensor data structures, and failure mode databases. This segmentation allows different experts to work on specific knowledge components independently, facilitates verification of individual knowledge elements, and enables collaborative development without requiring all experts to understand the entire diagnostic system.
Solution Approach 2:
The system uses explicit knowledge representations that can be copied, shared, and reviewed among multiple experts. Diagnostic rules, causal models, and knowledge bases are stored in standardized formats that allow transparent examination and collaborative refinement, replacing implicit encoded knowledge with shareable explicit representations that enhance collaboration while maintaining diagnostic capability.
Data Source
AI summary
A computer implemented method for diagnosing a system includes: receiving a causal graph, the causal graph defining (i) a plurality of nodes each representing a module of a plurality of modules of a system, wherein each module is characterized by one or more signals; and (ii) edges connected between the nodes, the edges representing propagation of performance between modules; generating a reasoning tool by augmenting the causal graph with diagnostics knowledge based on historically determined relations between performance, statistical and causal characteristics of at least one module out of the plurality of modules; obtaining a health metric of the at least one module, wherein the health metric is associated with the one or more signals associated with the at least one module; and using the health metric as an input to the reasoning tool to identify a module that is the most likely cause of the behavior.