A Digital Twin–Based Analytical Framework for Early Warning and Preventive Decision-Making in Crop Failure and Food Security

Modern agricultural systems face increasing uncertainties, rare risks, and complex dynamics that make traditional decision-making ineffective. This research presents an innovative digital twin–based framework for proactive decision-making in such systems that simultaneously considers temporal dynamics, structural uncertainty, and conflicts between management objectives. The proposed framework, relying on matrix dynamic modeling and introducing an analytical early warning indicator, allows for the gradual identification of the system’s movement toward critical areas before a crisis occurs. Numerical and behavioral results show that the model is able to produce sustainable, interpretable, and adaptive proactive policies and maintain its structural resilience in the face of rare and severe shocks. This framework is not only theoretically coherent but also provides a practical tool from a managerial perspective for the transition from reactive to proactive and data-driven management in agricultural systems.