LLM-empowered remanufacturing: A novel human-AI interaction under the lens of a scoping review
TL;DR
Gen AI systems text-based, vision-based, and multimodal, deployed both locally and in cloud-based environments, were found to be utilized in remanufacturing processes, with the majority of studies focusing on LLM models developed by OpenAI.
Abstract
In recent years, due to the high importance of raw materials, cost reduction pressures, and environmental concerns, remanufacturing processes have gained significant importance. In parallel, different Artificial Intelligence (AI) models especially, Generative AIs (Gen AIs) have been penetrating into work environments to improve efficiency and effectiveness of the tasks. Aim of this article is to provide a scoping review of studies addressing the use of Gen AI in remanufacturing processes. The literature search was conducted across four major databases, and the results were complemented using Google Scholar. In total, 23 articles were included in the final stage of analysis. Gen AI systems text-based, vision-based, and multimodal, deployed both locally and in cloud-based environments, were found to be utilized in these processes. The majority of studies focusing on LLM models developed by OpenAI. The most prominent contributions of Gen AI were observed in the following areas: Code Generation and Robotic Control, Safety and Risk, Supporting system performance as a powerful “eye”, and Intelligent Planning and Reasoning (as the strategist of the system). It appears that the initial phase of introducing diverse applications of generative AI models in remanufacturing processes is approaching its conclusion. In the coming years, with the rapid advancement of generative AI, a substantial portion of these processes is expected to be delegated to such systems, and it is even conceivable that they may partially replace robots in certain tasks.
