Research

Paper

AI LLM March 25, 2026

Evaluating Chunking Strategies For Retrieval-Augmented Generation in Oil and Gas Enterprise Documents

Authors

Samuel Taiwo, Mohd Amaluddin Yusoff

Abstract

Retrieval-Augmented Generation (RAG) has emerged as a framework to address the constraints of Large Language Models (LLMs). Yet, its effectiveness fundamentally hinges on document chunking - an often-overlooked determinant of its quality. This paper presents an empirical study quantifying performance differences across four chunking strategies: fixed-size sliding window, recursive, breakpoint-based semantic, and structure-aware. We evaluated these methods using a proprietary corpus of oil and gas enterprise documents, including text-heavy manuals, table-heavy specifications, and piping and instrumentation diagrams (P and IDs). Our findings show that structure-aware chunking yields higher overall retrieval effectiveness, particularly in top-K metrics, and incurs significantly lower computational costs than semantic or baseline strategies. Crucially, all four methods demonstrated limited effectiveness on P and IDs, underscoring a core limitation of purely text-based RAG within visually and spatially encoded documents. We conclude that while explicit structure preservation is essential for specialised domains, future work must integrate multimodal models to overcome current limitations.

Metadata

arXiv ID: 2603.24556
Provider: ARXIV
Primary Category: cs.IR
Published: 2026-03-25
Fetched: 2026-03-26 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.24556v1</id>\n    <title>Evaluating Chunking Strategies For Retrieval-Augmented Generation in Oil and Gas Enterprise Documents</title>\n    <updated>2026-03-25T17:35:24Z</updated>\n    <link href='https://arxiv.org/abs/2603.24556v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.24556v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Retrieval-Augmented Generation (RAG) has emerged as a framework to address the constraints of Large Language Models (LLMs). Yet, its effectiveness fundamentally hinges on document chunking - an often-overlooked determinant of its quality. This paper presents an empirical study quantifying performance differences across four chunking strategies: fixed-size sliding window, recursive, breakpoint-based semantic, and structure-aware. We evaluated these methods using a proprietary corpus of oil and gas enterprise documents, including text-heavy manuals, table-heavy specifications, and piping and instrumentation diagrams (P and IDs). Our findings show that structure-aware chunking yields higher overall retrieval effectiveness, particularly in top-K metrics, and incurs significantly lower computational costs than semantic or baseline strategies. Crucially, all four methods demonstrated limited effectiveness on P and IDs, underscoring a core limitation of purely text-based RAG within visually and spatially encoded documents. We conclude that while explicit structure preservation is essential for specialised domains, future work must integrate multimodal models to overcome current limitations.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.IR'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-25T17:35:24Z</published>\n    <arxiv:comment>Presented at CCSEIT 2026. This version matches the published proceedings</arxiv:comment>\n    <arxiv:primary_category term='cs.IR'/>\n    <arxiv:journal_ref>Computer Science and Information Technology (CS and IT), pp. 49-67, 2026</arxiv:journal_ref>\n    <author>\n      <name>Samuel Taiwo</name>\n    </author>\n    <author>\n      <name>Mohd Amaluddin Yusoff</name>\n    </author>\n    <arxiv:doi>10.5121/csit.2026.160507</arxiv:doi>\n    <link href='https://doi.org/10.5121/csit.2026.160507' rel='related' title='doi'/>\n  </entry>"
}