Causality-Based Fleet Matching for Manufacturing Sensor Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for determining root causes and corrective actions in manufacturing systems, particularly semiconductor manufacturing, rely on correlations rather than causations, leading to inaccurate assignments of root causes and ineffective corrective actions, resulting in increased downtime and costs due to the lack of consideration for product data and directed causal relationships between sensors and equipment constants.
Innovation Solution
A causality-based approach using a directed acyclic graph (DAG) to represent causal relationships between sensors, integrating product data and equipment constants, enabling accurate root cause analysis and effective corrective actions by determining causal strengths and directing corrective actions based on these relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If correlation-based methods are used to determine root causes, then the analysis process is simple, but the accuracy of root cause identification deteriorates
Solution Approach 1:
The patent introduces a causal strength index as an intermediary metric that quantifies the strength of causal relationships between sensors. This index serves as a mediator between the complex causal graph analysis and the final root cause identification, enabling accurate root cause determination through measurable causal strengths rather than simple correlations.
Solution Approach 2:
The patent replaces the traditional correlation-based statistical approach with a causality-based analytical framework. By substituting correlation coefficients with causal strength indices derived from causal graphs, the system achieves more accurate root cause identification while maintaining computational feasibility through structured causal relationship modeling.
2Reliability
If manual inspections and performance metrics are used for system health assessment, then the implementation is straightforward, but unplanned downtime increases
Solution Approach 1:
The system enables automated self-diagnosis by using causal graphs to automatically identify root causes of sensor anomalies. The causal strength index matrix autonomously ranks potential root causes, allowing the system to self-service without manual intervention and thereby reduce unplanned downtime while maintaining straightforward implementation through automated causal analysis.
Solution Approach 2:
The patent implements a feedback mechanism where sensor data continuously updates the causal graph analysis. When anomalies are detected, the system automatically queries the causal strength index matrix to identify root causes and recommends corrective actions, creating a closed-loop feedback system that improves reliability through automated monitoring and response.
3Ease of manufacture
If correlation-based root cause analysis is used, then the computational resources required are minimal, but corrective actions become ineffective
Solution Approach 1:
The patent performs preliminary construction of causal graphs and causal strength index matrices before actual root cause analysis is needed. By pre-establishing the causal relationships and their strengths among all sensors during system operation, the system prepares actionable causal information in advance, making corrective actions more effective when anomalies occur without requiring intensive real-time computational resources.
Data Source
AI summary
A method includes generating a causal graph based on a plurality of values, each value corresponding to a causal relationship between two or more sensors of a plurality of sensors in one or more manufacturing systems. The method further includes determining a causal strength index matrix. The method further includes responsive to identifying an anomalous behavior in at least one of the plurality of sensors, determining a root cause of the anomalous behavior using at least one of the causal strength index matrix or the causal graph. The method further includes causing a recommended corrective action to be issued based on the root cause of the anomalous behavior.