SIFLM: A Spatio-Relational Intelligence Framework for Enhanced Food Supply Chain Risk Assessment

Despite the proliferation of food supply chain process data due to digitization strategies, inadequate understanding of the complex spatial relationships within these chains has led to a scarcity of effective assessment models. This deficiency undermines the transition from reactive to proactive management of food safety, resulting in underutilized data resources. Addressing this gap, this paper introduces Spatial Integrated Food Logistics Model(SIFLM), a novel spatio-relational intelligence framework. By incorporating geographic relationships and deploying a relatedness analysis module, SIFLM optimizes the GRU model’s forecasting prowess, enabling the precise identification of critical nodes and prediction of safety vulnerabilities. This advancement equips regulators and industry players with the means to implement proactive strategies, shifting from post-event remediation to anticipatory action. By simulating the dynamic, spatiotemporal characteristics of food supply networks, SIFLM optimizes resource allocation and bolsters food security and sustainability in the face of supply chain informatization and environmental uncertainties.