Machine Learning for Predicting and Optimizing Global Trade Flows: Enhancing Efficiency in Cross-Border Transactions

This study explores the application of machine learning in predicting and optimizing global trade flows, improving trade forecasting, risk assessment, and regulatory compliance. A quantitative approach was utilized, where machine learning techniques such as time series forecasting, clustering, and anomaly detection were applied to analyze trade data from Indonesia. The case study of Indonesia's trade network was chosen to explore the practical application of these techniques in optimizing trade flows and assessing risks. Machine learning enhances trade flow predictions, detects undervaluation risks, and confirms that corporate affiliations significantly influence trade relationships. AI-driven risk assessment improves trade efficiency and regulatory oversight. Data quality and real-time integration remain challenges, and findings are specific to Indonesia’s trade network, requiring broader validation. AI-driven trade analytics can optimize supply chains, reduce disruptions, and enhance regulatory compliance by detecting trade mispricing and improving risk management. This research highlights the role of machine learning in trade forecasting and risk assessment, providing empirical insights for improving global trade efficiency.