LLM system aids explainable defect analysis in laser powder bed fusion

Researchers have developed a new decision-support system that combines structured knowledge about defects with large language models (LLMs) to analyze and guide mitigation strategies in laser powder bed fusion (LPBF) manufacturing. The system utilizes an ontology-integrated LLM, incorporating a knowledge base of 27 defect types and their causal relationships. It supports natural language queries for defect explanations and mitigation advice, and includes a multimodal image-assessment module for interpreting defect images. Evaluations showed the integrated system achieved a macro-average F1 score of 0.808, demonstrating improved consistency and interpretability. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT This system could improve manufacturing quality control by providing explainable defect analysis and mitigation guidance.

RANK_REASON This is a research paper detailing a novel system for defect analysis in manufacturing. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• LLM
• Laser Powder Bed Fusion
• LPBF
• ontology
• foundation models
• Cohen's kappa
• Md Habibor Rahman
• arXiv

• paper
• other