RESilient Emergency Preparedness for Natural Disaster Response through Operational Research(RESPOND-OR)

自然灾害对人类、社会和经济环境造成严重后果。虽然大规模自然灾害发生在世界各地，但统计证据表明，其负面影响在较不发达国家更为明显。2003年至2013年期间，发展中国家的自然灾害造成约5 500亿美元的损失，影响到20亿人。印度尼西亚和苏丹是受自然灾害经济和社会后果严重影响的国家之一。在印度尼西亚，一场自然灾害可能同时或相继引发另一场自然灾害。此外，印度尼西亚的救灾工作需要考虑到该国的群岛结构。在苏丹，洪水这一普遍灾害可能引发需要同时考虑的卫生紧急情况。此外，苏丹的救灾行动因国内冲突而具有高风险的特点，优化备灾和救灾干预措施为减少灾害的负面影响提供了充分的潜力，对受影响社区的可持续性具有重大的经济和社会效益。然而，现有的办法大多基于一般性假设，往往过于简化灾害管理机构的决策需要。具体而言，现有的备灾和救灾模式不能充分应对以下挑战：1.为同时和（或）相继发生的大规模自然灾害（如地震引发海啸或灾后疾病爆发）的救灾资源分配建立模型。战略备灾和业务救灾决策的综合建模。3.在非暴力反抗和社会冲突的情况下模拟人道主义支助资源的路线和时间安排。4.将公平标准纳入备灾和救灾决策模型。由于缺乏捕捉真实的世界复杂性的模型，导致稀缺的备灾和救灾资源分配和使用效率低下。因此，迫切需要解决灾害管理机构真实的世界决策环境复杂性所产生的数学建模和相关的计算和数据管理挑战。RESPOND-OR项目将开发下一代模型，其中将纳入所有相关利益攸关方的要求。所提出的模型的复杂性将需要开发新的超级算法，这将在非常短的计算时间内提供高质量的解决方案。数学模型、求解算法以及数据管理和可视化工具将支持决策支持系统的开发，该系统将提高印度尼西亚和苏丹备灾和救灾组织的决策能力。该研究团队在数学建模，启发式开发，随机优化，数据管理和可视化以及灾害准备和响应管理等领域具有国际领先的地位。研究团队在利益相关者参与方面有着出色的记录。我们将与我们的利益相关者合作伙伴密切合作，以确保RESPOND-OR的结果在科学上是合理的，并完全符合他们的需求。

Natural disasters have grave consequences for human, social and economic environment. Although large-scale natural disasters occur worldwide, statistical evidence suggests that their negative impacts are much more pronounced in less developed countries. Between 2003-2013, natural disasters in developing countries cost about $550 billion and affected 2 billion people. Indonesia and Sudan are among the countries enormously affected by the economic and societal consequences of natural disasters. In Indonesia, a natural disaster may trigger another natural disaster, either simultaneously or in a sequential order. In addition, the disaster response in Indonesia needs to take into consideration the archipelago structure of the country. In Sudan, the prevalent disaster, which is flooding may trigger a health emergency that requires simultaneous consideration. Furthermore, the disaster response operations in Sudan are characterized by high risk due to civil conflicts.The optimization of disaster preparedness and response interventions provides ample potential to decrease the magnitude of the negative impacts of the disasters with significant economic and societal benefits for the sustainability of the impacted communities. However, available approaches are mostly based on generic assumptions that tend to oversimplify the decision-making needs of disaster management agencies. Specifically, available disaster preparedness and response models do not adequately address the following challenges:1. Modelling of the allocation of disaster response resources for combined large-scale natural disasters that happen simultaneously and/or sequentially, i.e. earthquakes triggering tsunamis, or outbreak of diseases following floods.2. Integrated modelling of strategic disaster preparedness and operational disaster relief decisions. 3. Modelling the routing and scheduling of humanitarian support resources in the presence of civil disobedience and social conflict. 4. The incorporation of fairness criteria in modelling disaster preparedness and response decisions. The lack of models capturing the real world complexities leads to inefficient allocation and use of scarce disaster preparedness and response resources. Therefore, there is an urgent need to address the mathematical modelling and associated computational and data management challenges stemming from the complexity of the real world decision-making environment of disaster management agencies. The RESPOND-OR project will develop the next generation of models which will incorporate the requirements of all relevant stakeholders. The complexity of the proposed models will necessitate the development of new hyper heuristics that will provide good quality solutions in very short computational times. The mathematical models, the solution algorithms, and the data management and visualization tools will underpin the development of a Decision Support System (DSS) that will enhance the decision-making capabilities of disaster preparedness and response organizations in Indonesia and Sudan. The research team has an internationally leading profile in the areas of mathematical modelling, heuristic development, stochastic optimization, data management and visualization, and disaster preparedness and response management. The research team has an excellent record in stakeholder engagement. We will work very closely with our stakeholder partners to ensure that the outcome of RESPOND-OR will be scientifically sound and fully aligned with their needs.