Cross-border supply chain full link state visualization and abnormity intelligent interception recovery method

Abstract

The invention discloses a cross-border supply chain full link state visualization and abnormity intelligent interception recovery method, which comprises the steps of collecting and fusing order state information of a multi-source heterogeneous system, and constructing a unified and traceable state version tree; generating a unique credible consensus state sequence based on a multi-stage dynamic arbitration rule; performing anomaly identification based on a consensus state, and performing probabilistic prediction and root cause positioning on potential risks by using an anomaly conduction network and Monte Carlo simulation; modeling the recovery decision into a multi-objective optimization problem, and generating a Pareto optimal recovery strategy set by adopting an algorithm; automatic strategy execution is realized through a man-machine collaborative decision and strategy execution adapter, and arbitration, prediction and strategy generation models are continuously optimized based on reinforcement learning feedback. According to the method, the three core problems of state multi-version conflict, difficult abnormal conduction prediction and recovery strategy combination explosion in the cross-border supply chain are solved, and full-link state credible visualization and abnormal intelligent self-recovery are realized.

Images