Abstract: In response to the problems of low decision-making credibility and difficulty in cross-subject collaboration faced by the scheduling of each link in the industrial chain in a complex and uncertain environment, this paper proposes an intelligent scheduling framework that integrates cross-chain collaboration and causal inference, and constructs an integrated industrial chain scheduling decision optimization model. Firstly, through operations research and optimization of artificial intelligence and other technologies, the capacity constraints of each link are comprehensively considered. Secondly, based on the improved sparrow algorithm, causal weights between events are mined, and then linear or nonlinear optimization models are established. Finally, a decision-making optimization model for complex and uncertain industrial chains is designed and constructed to optimize decision-making schemes in specific abnormal scenarios, providing auxiliary decision-making solutions for quickly and efficiently solving supply chain problems. The problems of insufficient coal loading and the decline in railway section transportation capacity are taked as the objects to complete the optimization of the decision-making model, achieving the maintenance of safety stock and fulfillment status of work plans and supply plans as well as the guarantee of benefits, significantly improving the efficiency and accuracy of dispatching decisions in complex and uncertain industrial chains.