Process-aware graph-temporal framework for equipment prognostics with multimodal data at semiconductor final test

Key Points

• Equipment prognostics shows significant improvements in predictive accuracy, which can lead to decreased downtime.
• The framework integrates multimodal data, enhancing the reliability of failure prediction across semiconductor test processes.
• Methodological assessment of the graph-temporal framework allows for improved data processing and analysis for better outcomes.
• Significance is highlighted as the framework may enable more effective maintenance strategies in semiconductor manufacturing.