Fabian Hartung, Billy Joe Franks, Tobias Michels, Dennis Wagner, Philipp Liznerski, Steffen Reithermann, Sophie Fellenz, Fabian Jirasek, Maja Rudolph, Daniel Neider, Heike Leitte, Chen Song, Benjamin Klopper, Stephan Mandt, Michael Bortz, Jakob Burger, Hans Hasse, Marius Kloft

• This paper provides the first comprehensive evaluation and analysis of modern (deep-learning-based) unsupervised anomaly detection methods for chemical process data. We focus on the Tennessee Eastman process dataset, a standard litmus test to benchmark anomaly detection methods for nearly three decades. Our extensive study will facilitate choosing appropriate anomaly detection methods in industrial applications. From the benchmark, we conclude that reconstruction-based methods are the methods of choice, followed by generative and forecasting-based methods.