A Quantile Regression Approach for Remaining Useful Life Estimation with State Space Models

2025-06-20

AI Summary

Addressing high-precision, low-uncertainty remaining useful life (RUL) estimation for predictive maintenance in Industry 4.0/5.0—particularly under long-horizon degradation modeling challenges—this work proposes the first framework integrating synchronous quantile regression (SQR) into state-space models (SSMs). The SQR-SSM architecture enables joint multi-quantile forecasting, delivering well-calibrated uncertainty quantification without compromising point-estimate accuracy. Evaluated on the C-MAPSS benchmark, it outperforms leading sequence models—including LSTM, Transformer, and Informer—reducing average mean absolute error (MAE) by 12.6%–23.4% while accelerating inference by 3–8×, thus enabling real-time industrial deployment. The core contribution lies in the novel coupling paradigm of SQR and SSM, which jointly ensures interpretability, robustness to distributional shifts, and computational efficiency—advancing both theoretical modeling and practical applicability in prognostics.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Predictive Maintenance (PdM) is pivotal in Industry 4.0 and 5.0, proactively enhancing efficiency through accurate equipment Remaining Useful Life (RUL) prediction, thus optimizing maintenance scheduling and reducing unexpected failures and premature interventions. This paper introduces a novel RUL estimation approach leveraging State Space Models (SSM) for efficient long-term sequence modeling. To handle model uncertainty, Simoultaneous Quantile Regression (SQR) is integrated into the SSM, enabling multiple quantile estimations. The proposed method is benchmarked against traditional sequence modelling techniques (LSTM, Transformer, Informer) using the C-MAPSS dataset. Results demonstrate superior accuracy and computational efficiency of SSM models, underscoring their potential for high-stakes industrial applications.

Problem

• Estimating Remaining Useful Life (RUL) accurately for predictive maintenance
• Handling model uncertainty in State Space Models (SSM) with Quantile Regression
• Benchmarking SSM against traditional sequence modeling techniques for efficiency

Innovation

• State Space Models for long-term sequence modeling
• Simultaneous Quantile Regression for uncertainty handling
• Benchmarked against LSTM, Transformer, Informer