Toward a robust approach to multivariate time series anomaly detection

BackgroundAnomaly detection in semiconductor manufacturing is vital for maintaining yield and reducing costs, particularly in high-volume production where inspections are time-consuming and expensive.AimWe aim to develop a robust, unsupervised deep learning framework for multivariate anomaly detection, addressing limitations in current fault detection and classification systems.ApproachWe propose a transformer-based model integrating aggregated z-normalization to mitigate distribution drift and peaks-over-threshold for adaptive thresholding, ensuring reliable performance across varying datasets.ResultsThe framework achieved an F1 score of 0.9827 on semiconductor datasets, with high precision (0.9866) and minimal false alarms, validated through extensive ablation studies.ConclusionsThe proposed solution is scalable and adaptable for industrial environments, with future work focused on improving the detection of single-spike anomalies and borderline cases to enhance yield and operational reliability.