The Transition of Postharvest Science Toward Predictive and AI-Driven Systems: A Bibliometric and Technological Review

This study presents a critical historical, bibliometric, and technological overview of the evolution of postharvest science, emphasizing the transition from classical physiology-based approaches to emerging predictive and technology-driven systems. Scientific production related to postharvest research was analyzed using the Scopus and Web of Science databases, while bibliometric mapping and co-occurrence networks were generated using VOSviewer to identify thematic trends, emerging research areas, and structural scientific clusters. In parallel, a technological foresight analysis was conducted through the Lens.org platform to investigate the temporal evolution of patent deposits, the geographical distribution of innovation, the leading institutional applicants, and the predominant technological domains according to the Cooperative Patent Classification (CPC). The results revealed a substantial global expansion of postharvest research over recent decades. This growth was accompanied by increasing technological diversification and stronger integration between scientific knowledge and intellectual property protection. The analysis also highlighted the progressive incorporation of advanced methodologies into postharvest science, including biochemical approaches, non-destructive technologies, artificial intelligence, predictive modeling, and digital tools for quality assessment and shelf-life management. Overall, the study demonstrates that postharvest science is undergoing a paradigmatic transition toward integrated, multidisciplinary, and data-driven systems aligned with current demands for sustainability, food security, innovation, and reduction of postharvest losses.