Generative PLM: Utilizing Large Language Models for Automated Engineering Change Management and Requirements Traceability: A Case Study in Smart Manufacturing

Abstract

Generative large language models are emerging as a promising enabler for digital transformation in product lifecycle management, particularly in engineering change management and requirements traceability. This paper presents a case study in smart manufacturing that explores how generative PLM can support the automation of change request interpretation, impact analysis, and bidirectional traceability between engineering requirements and design artifacts. The proposed approach integrates large language models with structured PLM workflows to assist engineers in extracting relevant entities, identifying affected components, and linking changes to upstream requirements and downstream implementation records. In the case study, the method is evaluated in a smart manufacturing environment where frequent design revisions and cross-disciplinary dependencies create substantial coordination overhead. Results indicate that LLM-assisted workflows can reduce manual effort, improve consistency in traceability maintenance, and accelerate response time for engineering changes, while also highlighting the need for human oversight in safety-critical and ambiguous cases. The findings suggest that generative PLM can complement existing engineering information systems by improving adaptability, knowledge reuse, and decision support across the product lifecycle. Overall, this study contributes a practical framework for applying large language models to engineering change management and requirements traceability in Industry 4.0 settings.