Collaborative optimisation algorithm of international trade supply chain driven by artificial intelligence

Against the evolving global trade structure, international trade supply chains have scattered nodes, long cycles and high uncertainty.Traditional management lacks collaboration and dynamic response.This paper proposes an AI-driven supply chain optimisation model using reinforcement learning to integrate cross-border transport, inventory and fulfilment for multi-node coordination.The primary contribution lies in the systemic integration of AI for collaborative decision-making rather than the proposal of a novel algorithm.The results show that under the premise of stable order scale, the collaborative optimisation model has improved in terms of total fulfilment cost, order completion cycle and inventory turnover efficiency.The total cost has decreased by more than 20%, the average fulfilment time has been controlled within 30 days, and the inventory turnover cycle has been shortened to about 32 days.An AI-driven collaborative mechanism reduces multi-node imbalances and provides a feasible technical pathway for international trade supply chain optimisation.