Configuration Resilience of Emergency Medical Rescue Networks Under Coupled Disruptions Induced by Extreme Disasters: An Active-Learning-Assisted Optimization Approach
Aug 2026· Systems· Vol 14, pp. 922· 0 citations· 33 references
TL;DR
A two-stage scenario-based mixed-integer programming model is formulated that determines the locations and capacity levels of temporary rescue sites and emergency medical facilities and then optimizes scenario-adaptive casualty transfers among affected areas, rescue sites, emergency facilities, and hospitals.
Abstract
This study investigates configuration resilience in emergency medical rescue networks under disaster-induced compound disruptions. Major disasters can simultaneously intensify casualty severity, disrupt road networks, restrict response access, and increase hospital surge pressure, turning emergency preparedness into a location–capacity–transfer planning problem under uncertainty. We formulate a two-stage scenario-based mixed-integer programming model that determines the locations and capacity levels of temporary rescue sites and emergency medical facilities and then optimizes scenario-adaptive casualty transfers among affected areas, rescue sites, emergency facilities, and hospitals. The model considers 81 compound disruption scenarios defined by casualty severity, road-network disruption, response-access constraints, and hospital surge pressure. To solve large-scale instances, we develop a small-sample active-learning-assisted variable neighborhood search algorithm (AL-VNS), which combines a committee random forest surrogate with a progressive active-verification strategy to prioritize limited exact CPLEX evaluations, while keeping all accepted and reported solutions exactly evaluated. Numerical experiments show that AL-VNS achieves near-optimal performance in small- and medium-scale instances. Based on three independent runs for each large-scale instance, AL-VNS reduces average runtime by 34.71–63.73% relative to baseline variable neighborhood search (BVNS), while maintaining average objective-value differences of only 0.01–0.12% and requiring approximately 217 exact evaluations per instance. In the Ya’an case and the tested sensitivity settings, temporary rescue sites provide a relatively stable spatial triage-and-transfer backbone, whereas emergency medical facilities offer flexible surge capacity for relieving hospital pressure. The findings support resilience-oriented location and capacity planning for emergency medical rescue networks under uncertain compound disruptions.
In major disaster-induced mass casualty events, medical evacuation (MEDEVAC) serves as a key factor in determining casualty survival rates. To overcome the limitations of prior studies—particularly in constraint quantification and practical dispatching — we propose a multi-commodity flow model for MEDEVAC based on a time–space network. Using a state– time expanded network, our model integrates injury severity levels, time windows, asset capacity and type, facility heterogeneity, and route dependencies. We formulate a mixed-integer linear programming (MILP) model and develop a rolling-horizon branch-and-price algorithm. Case study results show an evacuation rate of 81.38% and an average waiting time of 77.81 minutes, demonstrating improved efficiency and resource utilization. This serves as a theoretical foundation for quantitative decision-making and dynamic scheduling of medical evacuation in major disasters.
Kun Dong, Hongming Li, Xiaohui Li et al.· 2026 IEEE 27th China Confere...· 0 citations
A Multi-Agent Disaster Management Simulator that automates the disaster response process using intelligent software agents, machine learning, graph-based routing, and generative artificial intelligence.
Hemanth S, Manoj M, Harisha S, Dr Manjunath B· International Journal of Adv...· 0 citations
Flooding can damage healthcare facilities, interrupt local care capacity, and trigger costly patient evacuations. Addressing these risks requires long-term resilience investments under uncertainty. We develop a bi-objective two-stage stochastic optimization model that jointly determines permanent facility hardening and scenario-dependent evacuation decisions. The first objective minimizes expected evacuation, physical damage, and business-interruption costs. The second minimizes a service-disruption impact index that combines the duration and scale of service loss with a place-based social-vulnerability weight. We develop an exact Benders decomposition and a scalable Lagrangian-dual method with tailored primal recovery, embedding both within an adaptive procedure for constructing informative Pareto frontiers. A case study of 3,752 hospitals and nursing homes in Texas evaluates the framework under climate-informed tropical-cyclone flood scenarios. The results show that minimizing economic losses alone can systematically allocate less protection to facilities located in socially vulnerable areas, while moderate movement along the Pareto frontier can substantially reduce disruption impacts at limited additional expected cost. The exact decomposition solves all tested instances, including larger stress tests for which the extensive formulation becomes memory-limited, while the Lagrangian method provides substantially faster high-quality solutions.
Gizem Toplu-Tutay, John J. Hasenbein, M. Kammer-Kerwick et al.· 0 citations
In the critical domain of emergency response, rapid building evacuation and rescue are often hindered by complex layouts and dynamic hazards such as fire spread, toxic smoke, structural instability, and unpredictable occupant behaviour. To address the challenge of optimizing rescue sweep strategies, this paper establishes a comprehensive mathematical framework that transitions from traditional static planning to dynamic, adaptive decision-making, with the objectives of minimizing rescue completion time and maximizing occupant safety under uncertain conditions. First, a three-level evaluation system based on the AHP–Entropy Weight Method is constructed to quantify 32 critical influence factors, ranging from environmental conditions and building layout to occupant behaviour and technical support. The resulting weighted framework provides a data-driven foundation for subsequent optimization models. Three representative building configurations are then examined. For a 1D linear layout, a Divide-and-Conquer strategy reduces completion time by 42% compared with a single-entry approach. For a 2D branching layout, the rescue problem is formulated as a load-balanced Multi-Agent Traveling Salesman Problem (MA-TSP). For a 3D high-rise structure, a hierarchical Pincer Strategy is developed by nesting the MA-TSP within a multi-objective vehicle-routing framework and strategically leveraging vertical assets such as fire-service elevators. Finally, a Dynamic Adaptive Replanning Model is introduced using time-dependent hazard functions, dynamic edge costs, and occupant re-entry events. A Rolling-Horizon Replanning mechanism continuously updates rescue routes as the emergency evolves.
Qihao Chen· Journal of Frontiers in Tech...· 0 citations
Natural disasters and cascading infrastructure failures disrupt transportation networks precisely when reliable logistics routing is most critical for delivering relief supplies, medical equipment, and evacuation support. Static shortest-path and classical vehicle-routing heuristics assume a largely intact network and adapt poorly to rapidly evolving road closures, congestion, and demand surges. This study presents an investigation of deep reinforcement learning (DRL) for adaptive, real-time logistics routing under disaster conditions, culminating in a proposed graph-attention multi-agent proximal policy optimization architecture (GAT-MAPPO).
A synthetic disaster-response environment was constructed as a 60-node transportation network subject to stochastic edge disruption, congestion, and time-varying relief-demand surges representing earthquake, flood, hurricane, and combined multi-hazard scenarios. Six baseline routing strategies - static Dijkstra, A* heuristic, Clarke-Wright savings vehicle routing, deep Q-network (DQN), Double DQN, and asynchronous advantage actor-critic (A3C) - were benchmarked against proximal policy optimization (PPO), soft actor-critic (SAC), and the proposed GAT-MAPPO agent across 500 evaluation episodes per method. GAT-MAPPO achieved the highest demand fulfillment rate (93.7%) and lowest mean delivery delay (11.2 min), outperforming the strongest baseline (SAC, 85.1% fulfillment) and the static-routing baseline (58.2% fulfillment) by wide margins.
The proposed agent retained a demand fulfillment rate above 80% at an edge-failure rate of 30%, generalized across all four disaster-scenario types with less than eight percentage points of performance variation, and exhibited graceful degradation under simulated sensor and communication noise. Multi-objective analysis mapped Pareto-optimal trade-offs between delivery time and fleet fuel consumption, and between responder risk exposure and delivery speed, while an ablation of the reward function confirmed that urgency weighting and failed-delivery penalties contributed most to overall performance. These results, illustrate a coherent workflow for developing, benchmarking, and stress-testing DRL-based disaster-logistics routing systems prior to real-world deployment.
Oyinloluwa Sarah Ogunseye, Marley Brown, Yejide Eniola Dabiri· International Journal of Mul...· 0 citations
Artificial intelligence (AI) stands at the forefront of transforming emergency management, offering unprecedented capabilities in disaster preparedness and response. Recent implementations demonstrate this shift from reactive to proactive approaches, particularly through flood prediction algorithms and maritime search-and-rescue optimization systems that integrate real-time vessel locations and weather data. However, the current landscape reveals a critical challenge: The opacity of AI systems creates a significant trust deficit among emergency responders and communities. Research findings paint a concerning picture of this transparency gap. A comprehensive survey of emergency management AI systems reveals striking statistics: 68 percent lack adequate documentation of their data sources, while 42 percent fail to provide clear justifications for their recommendations. This “black box” phenomenon carries serious implications, particularly when flood prediction models disproportionately affect vulnerable populations or when opaque decision-making processes lead to suboptimal resource allocation during critical rescue operations. Analysis of real-world applications in flood preparedness and search-and-rescue operations exposes systematic communication deficiencies within these essential emergency response frameworks. The research examines how varying levels of AI transparency directly influence emergency responders’ decision-making during crises, exploring the delicate balance between operational openness and security considerations. These findings highlight an urgent need for robust oversight mechanisms and context-specific transparency protocols to ensure ethical AI deployment in emergency management. The evidence points toward a clear solution: developing human-centric approaches that enhance rather than replace human capabilities in emergency response. This strategy requires establishing tailored transparency guidelines and monitoring systems that address current challenges while facilitating effective AI integration. By prioritizing both technological advancement and human oversight, emergency management systems can better serve their critical public safety mission.
Jaideep Visave· Journal of Emergency Managem...· 0 citations
