Disturbance-aware collaborative routing for concurrent emergency response

Concurrent emergency response (CER) scenarios pose significant challenges to emergency vehicle (EV) routing and scheduling due to dynamic traffic disturbances and resource competition. Existing methods typically decouple routing and scheduling and fail to capture disturbance-aware traffic dynamics, leading to suboptimal performance. This paper proposes a disturbance-aware hierarchical framework that jointly optimizes route generation and collaborative scheduling. A Disturbance-Aware Graph WaveNet (DAGWN) is developed to model spatiotemporal traffic evolution and predict traffic states and disturbance intensity. Building on this, a unified scheduling cost model incorporating task urgency, travel time, route conflict, and fairness is formulated and solved via a minimum cost flow (MCF) approach. Experiments on a SUMO-based urban traffic simulation demonstrate that the proposed method improves prediction accuracy and reduces total completion time, average response time, and maximum response time compared with state-of-the-art baselines. The results validate the effectiveness of integrating disturbance-aware modeling with joint routing and scheduling for efficient emergency response.