Litcius/Paper detail

Scalable anomaly detection in manufacturing systems using an interpretable deep learning approach

Thomas Schlegl, Stefan Schlegl, Nikolai West, Jochen Deuse

2021Procedia CIRP16 citationsDOIOpen Access PDF

Abstract

Anomaly detection in manufacturing systems has great potential for the prevention of critical quality faults. In recent years, unsupervised deep learning has shown to frequently outperform conventional methods for anomaly detection. However, tuning, deploying and debugging deep learning models is a time-consuming task, limiting their practical applicability in manufacturing systems. We approach this problem by developing a deep learning model that learns interpretable shapes that can be used for anomaly detection in temporal process data. Application of the model to assembly tightening processes in the automotive industry shows a significant improvement in model interpretability and scalability.

Topics & Concepts

InterpretabilityAnomaly detectionScalabilityDeep learningArtificial intelligenceComputer scienceDebuggingAutomotive industryMachine learningProcess (computing)Anomaly (physics)LimitingTask (project management)Fault detection and isolationData miningEngineeringSystems engineeringProgramming languageMechanical engineeringDatabaseAerospace engineeringPhysicsCondensed matter physicsActuatorOperating systemAnomaly Detection Techniques and ApplicationsFault Detection and Control SystemsTime Series Analysis and Forecasting
Scalable anomaly detection in manufacturing systems using an interpretable deep learning approach | Litcius