LAST-RAG: Literature-Anchored Stochastic Trajectory Retrieval-Augmented Generation for Knowledge-Conditioned Degradation Model Selection

2026-05-18

AI Summary

This study addresses the limitations of existing stochastic process–based degradation modeling approaches, which rely solely on statistical fitting and are prone to selecting models inconsistent with true underlying mechanisms, particularly under short observation windows or high noise levels. To overcome this, the authors propose LAST-RAG, a novel framework that reframes model selection as a knowledge-guided decision-making problem integrating observational data with domain knowledge. Leveraging retrieval-augmented generation (RAG), LAST-RAG retrieves mechanistic evidence from a local knowledge base to hierarchically constrain the candidate model space. Furthermore, it incorporates a rule-based confidence reasoning mechanism under uncertainty (RCRUS) to manage decision ambiguity. Evaluated on Wiener/Gamma process family classification and fine-grained degradation model selection tasks, LAST-RAG significantly outperforms established baselines, including purely statistical, predictive, and uncertainty-aware methods.

Abstract

Stochastic-process-based degradation modeling is a core approach for estimating the distribution of remaining useful life (RUL); however, the selection of an appropriate stochastic process has not been sufficiently addressed. Existing model selection methods mainly rely on the statistical fit of the observed health indicator (HI) trajectory, but this approach may select a model that is inconsistent with the underlying degradation mechanism when the observation window is short or the signal is highly noisy. To address this issue, this paper proposes Literature-Anchored Stochastic Trajectory Retrieval-Augmented Generation (LAST-RAG). The proposed method uses both the observed HI trajectory and domain-specific context, and hierarchically conditions the candidate degradation model space based on theoretical and mechanical evidence retrieved from a local evidence bank. In addition, Rule-based Confidence Reasoning with Uncertain State (RCRUS) is introduced to prevent candidate models from being prematurely eliminated when hierarchical decisions are uncertain. Simulation-based experiments demonstrate that the proposed method outperforms statistical, prognostic, and uncertainty-aware baselines in both Wiener/gamma family classification and detailed degradation model classification. Ultimately, this study reframes degradation model selection from a purely statistical goodness-of-fit problem into a knowledge-conditioned decision-making problem that integrates observed data with domain knowledge.

Problem

• degradation model selection
• stochastic process
• remaining useful life
• knowledge-conditioned decision-making
• health indicator

Innovation

• Retrieval-Augmented Generation
• Stochastic Degradation Modeling
• Knowledge-Conditioned Model Selection
• Remaining Useful Life
• Domain Knowledge Integration