Treffer: Exploring the applications of natural language processing and language models for production, planning, and control activities of SMEs in industry 4.0: a systematic literature review.

In the wake of the prominence of language models such as ChatGPT/GPT4 and the emergence of various Natural Language Processing (NLP) approaches, there has been growing interest in their applications. However, a gap exists in scientific documentation regarding Small and Medium Enterprises (SMEs) within the industrial sector. This paper addresses this gap. This is the first systematic review of the literature associated with the context of NLP in industry. Through five research questions, it provides an overview of NLP applications, goals, technical solutions, obstacles, and applicability to SMEs, which is useful for both researchers and manufacturers. Following the PRISMA 2020 methodology, this study reveals a lack of literature addressing the use of NLP in industrial SMEs. The findings suggest that NLP is predominantly applied in specific industrial domains, including design, process monitoring, and maintenance. NLP applications mainly aim to enhance operational performance, notably in support functions like maintenance, safety, and continuous improvement. Practical implementations include automatic data analysis, similarity searches, information retrieval, and the conversion of raw text into standardized data. When looking at the technical solutions implemented, the paper demonstrates a strong diversity in the encountered algorithmic approaches. Challenges include remaining up-to-date, scaling, addressing low-quality or insufficient data issues, and navigating domain- or operator-specific vocabulary. In particular, maintaining up-to-date data presents a critical challenge for NLP applications but with limited identified solutions. Finally, the study indicates that only a fraction of the proposed NLP algorithmic solutions may apply to SMEs because of a lack of resources, expertise, and standardized procedures. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Intelligent Manufacturing is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)