Multi-Field Hybrid Retrieval-Augmented Generation for Maritime Accident Root Cause Analysis

2026-06-11 · Source: Artificial Intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Maritime & Shipping · Depth: Expert, quick

Summary

A multi-field hybrid retrieval-augmented generation (RAG) framework is proposed for automated maritime accident root cause analysis (RCA). This system utilizes a comprehensive dataset of 13,329 Korea Maritime Safety Tribunal (KMST) reports spanning 1971-2025. The framework transforms raw adjudications into structured "incident cards" with distinct Summary, Causes, and Disposition fields, alongside a hierarchical L1/L2 cause taxonomy. Its retrieval strategy employs a field-aware hybrid approach, fusing sparse and dense rankings via Reciprocal Rank Fusion (RRF). Experimental results show the proposed retrieval significantly outperforms baselines, improving NormRecall@100 from 0.18 to 0.55. Furthermore, grounding the generator on retrieved precedents enhances RCA generation quality, increasing the LLM-as-a-judge score from 3.34 to 3.72 over an LLM-only baseline.

Key takeaway

For AI Scientists or Machine Learning Engineers developing RAG systems for specialized domains like legal or medical reports, this research demonstrates the value of a multi-field hybrid approach. You should consider structuring complex documents into field-specific "incident cards" and employing Reciprocal Rank Fusion for retrieval. This method can significantly improve both retrieval accuracy and the quality of generated analyses, streamlining workflows for evidence-based report drafting.

Key insights

Field-aware hybrid RAG significantly improves maritime accident root cause analysis by structuring reports and enhancing retrieval and generation.

Principles

• Structuring unstructured reports into "incident cards" improves retrieval.
• Hybrid retrieval fusing sparse and dense rankings is effective.
• Field-aware indexing enhances relevance in complex documents.

Method

The framework transforms raw adjudications into structured "incident cards" with Summary, Causes, and Disposition fields, indexed with a hierarchical cause taxonomy. It then uses a field-aware hybrid retrieval strategy via Reciprocal Rank Fusion.

In practice

• Structure legal documents into "incident cards" for RAG.
• Use Reciprocal Rank Fusion for hybrid retrieval.
• Index specific document fields for enhanced relevance.

Topics

• Retrieval-Augmented Generation
• Maritime Accident Analysis
• Hybrid Retrieval
• Reciprocal Rank Fusion
• Knowledge Base Structuring
• Large Language Models

Best for: AI Scientist, Research Scientist, Machine Learning Engineer

Related on AIssential

• Multi-Field Hybrid Retrieval-Augmented Generation for Maritime Accident Root Cause Analysis — Structuring RAG architectures around domain-specific document fields significantly enhances downstream generation quality and factual consistency.
• Advanced Retrieval Pipeline for RAG (HyDE, Hybrid Search, Reranking) | Build 100% Local Retrieval — Combining HyDE, hybrid search, and reranking significantly enhances RAG retrieval accuracy and relevance.
• Why Retrieval-Augmented Generation Fails: A Graph Perspective — RAG failures stem from shallow, fragmented internal information flow, not just poor retrieval.
• Beyond the Blank Stare: Injecting Medical Expertise into LLMs Through RAG-Powered Knowledge — RAG systems enhance LLM reliability in medicine by grounding responses in verifiable, external knowledge bases.
• Self-Conditioned Positional HNSW for Overlap-Aware Retrieval in Chunked-Document RAG Systems: Method and Industrial Evidence-Quality Audit — SCP-HNSW enhances RAG by using positional codes and a two-pass query to avoid redundant, overlapping chunks, confirmed by industrial evidence audits.
• Retrieval-Augmented Generation as a Bridge Between Classical NLP Pipelines and Modern LLM Systems — Retrieval-augmented generation (RAG) addresses the limitation of static knowledge in large language models (LLMs) by connecting them to external, up-to-date knowledge sources.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.