Multi-objective optimization for enhancing hospital network resilience under earthquakes

The catastrophic earthquakes will cause numerous casualties requiring medical treatment, resulting in overcrowding in hospitals. To maintain the functionality of the hospital network, ambulance diversion (AD) between hospitals is required to relieve the stress of crowded hospitals. This research proposes a multi-objective optimization (MOO)-based emergency response model to enhance the resilience of the regional-level hospital network following earthquakes. The proposed model determines the optimal AD plan between hospitals for a given earthquake scenario. The average waiting time in the emergency department (ED) and the shortage of inpatient beds are selected as the objectives to measure the functionality of the hospital network, where the optimal solution is obtained by the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The hospital network of Chengdu, China with the earthquake scenario of the Wenchuan earthquake is selected as the case study to demonstrate the validity of the proposed model. Important findings are: (1) in the optimal response plan, 66.2% of medical demands is transported to other hospitals by AD, (2) compared with the baseline situation, the average waiting time and the shortage of inpatient beds improve by 99.4% and 94.2% with the optimal emergency plan, (3) the high medical demand, low transportation efficiency, and the lack of ambulances have significant adverse impacts on the functionality of the hospital network, and (4) the proposed model is composed of high-effective coordination and a comprehensive hospital network, which avoids the waste of medical resources.