Identifying technologies in circular economy paradigm through text mining on scientific literature

Abstract

Technological innovation serves as the catalyst for the shift towards circular practices. Technologies not only address technical challenges, facilitating the transition to a more circular economy, but they also enhance business efficiency and profitability. Furthermore, they promote inclusivity and create job opportunities, ultimately yielding positive societal impacts. The research in this area tends to focus on digital technologies, neglecting other technological areas. Moreover, it heavily relies on literature reviews and expert opinions, potentially introducing biases. In this article we investigate the technological landscape of the circular economy through Natural Language Processing (NLP), examining key technologies used in this sector and the primary challenges in managing these technologies. The methodology is applied to more than 45,000 scientific publications and aims to extract technologies in the text of scientific articles with NLP. The findings of our analysis reveal a strong emphasis on emerging digital, life cycle assessment and biomaterials technologies. Furthermore, we identified seven distinct technological domains within the CE field. Finally, we provide advantages and problems arising in the adoption and implementation of these technologies in an industrial context.