Artificial Intelligence and Machine Learning Applications in the Postharvest Storage of Tomato

Abstract

The integration of artificial intelligence (AI) into postharvest agricultural practices has advanced considerably in recent decades, driven by substantial progress in scientific research and technological development. Tomato softening and quality degradation during postharvest storage present major challenges in minimizing food waste and maintaining market value. This review explores the integration of artificial intelligence (AI) and machine learning (ML) technologies with non-destructive methods such as computer vision, imaging, electrical signal analysis, to monitor and predict tomato ripening stages and shelf life. Recent studies demonstrate high prediction accuracies using advanced models, including artificial neural networks, ensemble learning, and fuzzy inference systems, which analyze features like firmness, color, lycopene content, and texture. These AI-driven approaches enable accurate classification of ripeness stages and optimization of storage conditions, offering significant advantages over traditional destructive techniques. For the future ML, integration of large and complex algorithms and AI-driven systems controlling smart ripening chambers by adjusting ethylene concentration, humidity, and temperature based on real-time sensor feedback, will support uniform ripening, precision postharvest handling.  The potential of mobile applications and/or with the advent of recently developed smart glasses integrated with artificial intelligence, producers will be able to assess fruit ripeness in real time simply by visually inspecting the fruit, enabling rapid and informed harvesting decisions. Overall, the adoption of AI-based solutions in tomato postharvest management holds promise for improving quality monitoring, reducing spoilage, and enhancing sustainability in the agri-food sector.