# Unsupervised Process Anomaly Detection and Identification Using the Leave-One-Variable-Out Approach

## TL;DR

This paper introduces a new unsupervised machine learning model for detecting and identifying anomalies in industrial systems without needing prior failure data.

## Contribution

The novel LOVO model masks one variable at a time to learn process correlations without requiring a latent space.

## Key findings

- The LOVO model outperformed comparative models in synthetic data detection.
- Comparative models outperformed LOVO in experimental data detection but required optimal latent size selection.
- LOVO showed strong identification performance with synthetic data despite minor interpretability issues.

## Abstract

Automated anomaly detection and identification can signal equipment issues and pinpoint causes in large-scale industrial systems. For systems with limited failure history, unsupervised machine learning methods can be utilized as they do not require past failures. This study introduces the leave-one-variable-out (LOVO) model, which masks one variable at a time to predict the others, learning underlying process correlations. Detection performance was assessed with synthetic and experimental data, while identification performance used only synthetic data due to its ability to generate labeled anomaly types. For detection using synthetic data, the LOVO model generally outperformed comparative models; while using experimental data, the comparative methods outperformed the LOVO model. However, the comparative methods required selecting a latent size, and these conclusions pertain to using the optimal size. In practice, it would not be feasible to always select the optimal value, and incorrect selections impacted performance. In contrast, the LOVO model does not require a latent space. For identification using synthetic data, the LOVO model was slightly outperformed in interpretability and repeatability but still demonstrated impressive results. These outcomes suggest that the LOVO model is an effective model and may be more easily implemented without the challenging tuning process of selecting a latent size.