Interpretable Predictive Maintenance: Combining Anomaly Detection with Quantitative Root Cause Analysis

doi:10.1007/978-3-032-05179-0_18

Interpretable Predictive Maintenance: Combining Anomaly Detection with Quantitative Root Cause Analysis

2026-01-01

0

PRE

AI

I

Inês de Carvalho Jerónimo Barbosa *

DOI：10.1007/978-3-032-05179-0_18

Original

Original request for help

Save

Abstract

En 中文

Predictive Maintenance (PdM) aims to prevent failures through early detection, yet lacks explainability to support decision-making. Current PdM models often identify failures, but fail to explain their root causes, especially in real-world scenarios, with complex and limited labeled data. This study proposes an interpretable framework that combines LSTM-based Anomaly Detection with a dual-layered Root Cause Analysis (RCA) based on SHAP attributions. Applied to a real-world dataset, the method detects degradation transitions, tracks failure patterns over time, and provides interpretable information without explicit root cause labels.

Keywords:

Predictive Maintenance

Anomaly Detection

Root Cause Analysis

Explainability

Interpretable Predictive Maintenance: Combining Anomaly Detection with Quantitative Root Cause Analysis

Abstract

Journal

Organization