Applications of LLMs in Offshore Engineering Consulting and Design: Opportunities, Challenges, and Solutions
Abstract The rapid development of large language models (LLMs) has brought new opportunities to the offshore engineering consulting and design industry. This paper aims to explore the specific application scenarios and potential value of LLMs in engineering work. First, combined with the multidisciplinary characteristics of offshore engineering, the role of LLMs in knowledge management, technical document generation and design optimization is analyzed. For example, through natural language processing technology, LLMs can efficiently process design specifications, standards and historical engineering data, and provide fast and accurate query services. At the same time, its powerful language generation ability can assist in the preparation of complex engineering reports, design specifications and bidding documents, thereby improving work efficiency. In addition, LLMs can also provide innovative design suggestions by learning from the company’s existing experience and industry data to ensure the optimization of complex engineering solutions. In engineering consulting services, LLMs can support cross-team collaboration, promote knowledge sharing and reduce communication costs. However, this paper also discusses its challenges in professionalism, data privacy and reliability of use, and proposes targeted solutions. Through the analysis of actual cases, this paper verifies the significant advantages of LLMs in improving productivity, optimizing design processes and reducing costs, indicating that its application potential in offshore engineering consulting and design companies is significant.
Authors
- Stian Laland Rasmussen
- Per R. Nystrøm
- Yong He (ORCID: https://orcid.org/0000-0001-6752-1757)
- Anders Jakobsson
Institutions
- Marine design (Norway) (NO)
Publication Details
- Published
- 2025-06-22
- DOI
- https://doi.org/10.1115/omae2025-158330
- Primary Topic
- Manufacturing Process and Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00