Digital twins as a tool to increase the efficiency and sustainability of the agro-industrial complex

This article examines the applicability and impact of digital twin technologies across the agri-food sector, from field operations to postharvest logistics. Building on a structured review of recent literature and industry cases, we synthesize how physics-based and data-driven twins integrate sensing, IoT, and AI to enable real-time monitoring, scenario analysis, and decision support. We classify agricultural digital twins into four functional groups – crop and livestock production, machinery and maintenance, postharvest handling, and supply-chain coordination – and map them to key performance indicators such as yield stability, resource efficiency, quality preservation, and risk reduction. The analysis identifies consistent benefits, including improved input use, earlier anomaly detection, and more resilient planning under weather and market uncertainty. However, adoption is constrained by data quality and interoperability issues, high initial costs, skills gaps, and unresolved concerns around cybersecurity and governance. We propose a staged implementation framework that prioritizes high-value use cases, lightweight edge analytics, and open data standards, accompanied by a governance model for data sharing. The article contributes a practical roadmap and a set of evaluation metrics that link digital-twin capabilities to farm-level and supply-chain decisions. Limitations relate to heterogeneous evidence and context specificity. Future work should explore integration with generative AI, privacy-preserving learning, and cross-farm platforms.