Evidence-Driven Reasoning for Industrial Maintenance Using Heterogeneous Data
Abstract
Industrial maintenance platforms contain rich but fragmented evidence, including free-text work orders, heterogeneous operational sensors or indicators, and structured failure knowledge. These sources are often analyzed in isolation, producing alerts or forecasts that do not support conditional decision-making: given this asset history and behavior, what is happening and what action is warranted? We present Condition Insight Agent, a deployed decision-support framework that integrates maintenance language, behavioral abstractions of operational data, and engineering failure semantics to produce evidence-grounded explanations and advisory actions. The system constrains reasoning through deterministic evidence construction and structured failure knowledge, and applies a rule-based verification loop to suppress unsupported conclusions. Case studies from production CMMS deployments show that this verification-first design operates reliably under heterogeneous and incomplete data while preserving human oversight. Our results demonstrate how constrained LLM-based reasoning can function as a governed decision-support layer for industrial maintenance.
Cite
@article{arxiv.2603.08171,
title = {Evidence-Driven Reasoning for Industrial Maintenance Using Heterogeneous Data},
author = {Fearghal O'Donncha and Nianjun Zhou and Natalia Martinez and James T Rayfield and Fenno F. Heath and Abigail Langbridge and Roman Vaculin},
journal= {arXiv preprint arXiv:2603.08171},
year = {2026}
}