Time-Series Foundation Model Embeddings for Remaining Useful Life Estimation

2026-06-10

AI Summary

This work addresses the challenge in industrial predictive maintenance where remaining useful life (RUL) estimation typically relies on extensive labeled data and manual feature engineering. The authors propose, for the first time, leveraging the frozen pretrained time series foundation model Chronos-2 as a general-purpose backbone to extract contextual representations from multivariate sensor data, combined with a lightweight regression head for end-to-end fine-tuning. This approach eliminates the need for task-specific modeling or large-scale annotations, substantially improving data efficiency. Evaluated on two real-world industrial equipment datasets, the method consistently outperforms baseline models—including recurrent networks, convolutional architectures, Transformers, and gradient boosting—and demonstrates sustained performance gains as context length increases.

Abstract

Remaining Useful Life (RUL) prediction is essential for industrial predictive maintenance, yet many learning-based approaches rely on extensive feature engineering or large labeled datasets to train task-specific sequence models. In this work, we introduce a lightweight learning approach, in which we leverage a frozen pretrained time-series foundation model (TSFM) and combine it with a small regression head for RUL estimation from multivariate sensor streams. More specifically, we use Chronos-2 as a frozen backbone to extract context window features and train a lightweight regression neural network for RUL prediction. Experiments on real-world industrial sensor data from two device types show that Chronos-2 features consistently improve over recurrent, convolutional, Transformer-based, and gradient-boosting baselines under the same preprocessing and evaluation protocol. We further analyze the impact of context length and find that performance improves significantly with longer histories, indicating that TSFM representation offer a practical and data-efficient alternative for RUL estimation in industrial settings.

Problem

• Remaining Useful Life
• Predictive Maintenance
• Time-Series Foundation Model
• Feature Engineering
• Labeled Data

Innovation

• Time-Series Foundation Model
• Remaining Useful Life
• Chronos-2
• Frozen Backbone
• Lightweight Regression