Dynamic priority-driven dispatch in gas pipeline networks under catastrophic disruptions

In emergencies, the quick response capability of gas pipeline network dispatch is critical for maintaining flow equilibrium. For moderate disruptions, operators can sustain system balance through flow redistribution or reserve utilisation. However, in catastrophic events—marked by high uncertainty in supply restoration timelines and magnitudes—demand prioritisation and curtailment based on user heterogeneity become imperative, posing new challenges to conventional time-window constrained hierarchical dispatch. This paper proposes a user classification mechanism based on demand elasticity. Under tight resource limits, we study hierarchical scheduling to unlock demand flexibility, fill emergency flow allocation gaps, and boost system resilience. Thus, the emergency scheduling process is modelled as a priority-driven Markov Decision Process. Departing from proximity-based allocation, the model prioritises differential urgency in user demand profiles instead. A case study of China’s gas pipelines validates the model, showing interruptible users buffer shocks via demand elasticity. Moreover, gas reserve effectiveness is limited by pipeline capacity—releasing stored gas in severe disruptions may create bottlenecks, hindering efficient delivery. This approach provides an integrated emergency allocation system and theoretical foundations for priority-based flow management.