Use of Artificial Intelligence and Predictive Analytics in Forecasting Logistics Risks: Contemporary Approaches to Supply Chain Resilience Management in the United States

The purpose of the article is to examine contemporary approaches to the application of artificial intelligence (AI) and predictive analytics in logistics risk forecasting, analyze their implementation in supply chain management systems of U.S. enterprises, and develop the author's conceptual framework for integrating AI and predictive analytics into logistics risk management in order to enhance supply chain resilience in the context of the digital transformation of the U.S. economy. Research methodology. The methodological framework of the study is based on a combination of general scientific and specialized research methods. Methods of analysis, synthesis, and literature review were employed to systematize contemporary theoretical approaches to the application of AI and predictive analytics in logistics risk forecasting. A systems approach was used to examine the interrelationships among supply chain components and logistics risk management processes. Comparative analysis was applied to investigate AI implementation practices adopted by leading U.S. companies. The method of generalization was used to identify major trends in the digital transformation of logistics and to formulate the study's conclusions. A graphical method was employed to develop the author's conceptual framework for integrating AI and predictive analytics into a logistics risk forecasting system. Finally, the logical-analytical method was applied to substantiate directions for improving supply chain resilience management under conditions of economic digital transformation. Results. The findings indicate that the implementation of artificial intelligence and predictive analytics is fundamentally transforming logistics risk management by enabling a shift toward data-driven models based on continuous monitoring and forecasting of potential supply chain disruptions. The study demonstrates that the effectiveness of such systems depends on the integration of heterogeneous data sources, the application of machine learning algorithms, and their interaction with enterprise information platforms. The analysis of best practices adopted by leading U.S. enterprises made it possible to systematize the principal areas of AI application in logistics, including demand forecasting, transportation route optimization, inventory management, supplier reliability assessment, real-time monitoring of logistics operations, and decision support. The study found that the integrated use of these technologies significantly enhances the adaptability of supply chains to external risks. Practical implications. A major practical outcome of the research is the development of the proposed conceptual framework for integrating AI and predictive analytics into a logistics risk forecasting system. Unlike existing approaches, the proposed framework combines data collection and processing, intelligent forecasting, risk assessment, decision support, and a continuous model improvement mechanism based on newly generated data. This integrated approach provides a foundation for improving logistics management efficiency and strengthening supply chain resilience. The proposed framework may be applied by enterprises, logistics service providers, and other organizations in developing digital risk management strategies, as well as in future research on intelligent decision support systems for logistics. Value/Originality. The principal theoretical contribution of this study is the development of an original conceptual framework for integrating Artificial Intelligence and Predictive Analytics into logistics risk forecasting systems. The framework combines data acquisition, intelligent data analytics, risk forecasting, decision support, and the continuous improvement of predictive models within a unified digital architecture. The proposed approach may serve as a methodological foundation for the development of digital logistics management systems across enterprises operating in various industries.