Saved in:
Bibliographic Details
Main Authors: De Silva, Devin Yasith, Patel, Dhaval, Constantinides, Christodoulos, Lin, Shuxin, Zhou, Nianjun, Adams, Paul J, Rosato, Sal, Constantinides, Nicolas, McGuinness, Deborah L., Kalagnanam, Jayant
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.08614
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915996147646464
author De Silva, Devin Yasith
Patel, Dhaval
Constantinides, Christodoulos
Lin, Shuxin
Zhou, Nianjun
Adams, Paul J
Rosato, Sal
Constantinides, Nicolas
McGuinness, Deborah L.
Kalagnanam, Jayant
author_facet De Silva, Devin Yasith
Patel, Dhaval
Constantinides, Christodoulos
Lin, Shuxin
Zhou, Nianjun
Adams, Paul J
Rosato, Sal
Constantinides, Nicolas
McGuinness, Deborah L.
Kalagnanam, Jayant
contents Monitoring complex industrial assets relies on engineer-authored symbolic rules that trigger based on sensor conditions and prompt technicians to perform corrective actions. The bottleneck is not detection but response: translating rules into maintenance steps requires asset-specific knowledge gained through years of practice. We investigate whether LLMs can serve as decision support for this rule-to-action step and introduce \ours{}, a benchmark of 6{,}690 expert-validated multiple-choice questions from 118 rule-action pairs across 16 asset types. We contribute (i) a symbolic-to-MCQA pipeline normalizing rules to Disjunctive Normal Form with embedding-based distractor sampling, (ii) five variants probing distinct failure modes (Pro, Pert, Verbose, Aug, Rationale), and (iii) a benchmark of 29 LLMs and 4 embedding baselines. A human evaluation (9 practitioners, mean 45.0\%) confirms \ours{} requires specialist knowledge beyond operational experience. Three findings stand out. The frontier has closed: the top three LLMs lie within one Macro point, with Bradley-Terry Elo placing claude-opus-4-6 30 points above the next model. Yet \ours{}\,Pro exposes brittleness, with every model losing 13--60\% relative accuracy under distractor expansion. \ours{}\,Aug exposes pattern-matching: under condition inversion, frontier models still select the original answer 49--63\% of the time. The deployment bottleneck is not capability but calibration: frontier models handle template-style fault detection but break under structural perturbation.
format Preprint
id arxiv_https___arxiv_org_abs_2605_08614
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules
De Silva, Devin Yasith
Patel, Dhaval
Constantinides, Christodoulos
Lin, Shuxin
Zhou, Nianjun
Adams, Paul J
Rosato, Sal
Constantinides, Nicolas
McGuinness, Deborah L.
Kalagnanam, Jayant
Artificial Intelligence
Monitoring complex industrial assets relies on engineer-authored symbolic rules that trigger based on sensor conditions and prompt technicians to perform corrective actions. The bottleneck is not detection but response: translating rules into maintenance steps requires asset-specific knowledge gained through years of practice. We investigate whether LLMs can serve as decision support for this rule-to-action step and introduce \ours{}, a benchmark of 6{,}690 expert-validated multiple-choice questions from 118 rule-action pairs across 16 asset types. We contribute (i) a symbolic-to-MCQA pipeline normalizing rules to Disjunctive Normal Form with embedding-based distractor sampling, (ii) five variants probing distinct failure modes (Pro, Pert, Verbose, Aug, Rationale), and (iii) a benchmark of 29 LLMs and 4 embedding baselines. A human evaluation (9 practitioners, mean 45.0\%) confirms \ours{} requires specialist knowledge beyond operational experience. Three findings stand out. The frontier has closed: the top three LLMs lie within one Macro point, with Bradley-Terry Elo placing claude-opus-4-6 30 points above the next model. Yet \ours{}\,Pro exposes brittleness, with every model losing 13--60\% relative accuracy under distractor expansion. \ours{}\,Aug exposes pattern-matching: under condition inversion, frontier models still select the original answer 49--63\% of the time. The deployment bottleneck is not capability but calibration: frontier models handle template-style fault detection but break under structural perturbation.
title DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules
topic Artificial Intelligence
url https://arxiv.org/abs/2605.08614