Integrating Survival-Based Aging Models with Data-Driven RUL Prognostics

2026-07-03

PHM Society European Conference

AI Summary

This study addresses the unreliability of remaining useful life (RUL) estimation in predictive maintenance caused by information limitations inherent in single-model approaches. To overcome this, we propose an uncertainty-aware probabilistic fusion framework. By explicitly formulating structural assumptions linking wear, latent health, and failure, principled combination rules are derived to adaptively integrate a degradation component based on parametric survival models with a sensor-driven component utilizing a 1D CNN and posterior uncertainty. Experimental evaluations on the N-CMAPSS dataset demonstrate that the proposed framework effectively transcends the limited complementarity of conventional methods. It improves point prediction accuracy, yields narrower and better-calibrated prediction intervals, and significantly enhances the robustness of RUL estimation.

Abstract

Predictive maintenance requires reliable remaining useful life (RUL) estimation. Existing methods mainly follow two paradigms: wear-based aging models that capture cumulative degradation and sensor-driven data models that reflect instantaneous health conditions, each providing only partial information. In this work, we propose a probabilistic fusion framework that integrates wear-based and sensor-based prognostic components through failure probability distributions. Based on explicit structural assumptions linking wear, latent health, sensor observations, and failure, we derive a principled combination rule that enables uncertainty-aware integration with adaptive weighting of the components. Experimentally, we assess this combination rule by learning the wear-based component using a parametric survival model and the sensor-based component using a 1D convolutional neural network (1D-CNN) with a post-hoc uncertainty model. Evaluation on multiple N-CMAPSS datasets demonstrates that the fused model improves point accuracy, preserves the C-index, and produces narrower yet well-calibrated prediction intervals compared to either component alone. The results highlight the complementary roles of wear-based survival model and sensor-based deep learning model, and show that their probabilistic integration provides a structured pathway toward more robust and consistent prognostics over the life-time.

Problem

• Remaining Useful Life
• Predictive Maintenance
• Prognostics
• Aging Models
• Data-Driven Models

Innovation

• Probabilistic Fusion Framework
• Remaining Useful Life (RUL)
• Survival Model
• 1D-CNN
• Uncertainty Quantification