BRIGE: Understanding and Managing Humanitarian Logistics Systems through Advances in Optimization and Game Theory

这项扩大参与工程研究启动赠款（bridge）赠款提供资金，用于定量评估在缺乏集中决策权的情况下人道主义和公共卫生后勤系统中发生的效率损失，并通过优化设计系统参数来发现减轻这种影响的机制。在这种情况下，集中决策是不可能的，但分散的系统结果可能很差。这项研究将检查个人选择服务设施的环境，例如，在灾难发生后收集所需物资或接种预防传染病的疫苗。模型将通过整合博弈论和优化的概念来创建，以表示分散代理的决策，这些代理根据不同的目标行事，但其决策会影响整个系统的结果。将开发新的理论性能界限，将分散系统的性能与集中式计划者可以将个人分配到设施的最佳情况系统进行比较。将寻求有效的算法来解决分散的问题。最后，将设计机制来协调分散的系统，或确保其运作更接近集中的系统。如果成功，这项研究的结果将导致创造新的优化和博弈论工具，具有为人道主义和公共卫生物流系统的决策提供信息的巨大潜力。这项研究将强调在这些环境中提高效率和效力的机会，改善对稀缺资源的利用，并为受影响的人口取得成果。对所提出模型的分析也将促进对更一般的去中心化网络的理论理解。拟议的研究活动与一个扩大参与计划相结合，该计划将通过让这些学生参与展示工程师如何改变社会的推广和研究活动，为来自代表性不足群体的大学预科生和本科生创造通往工程职业的途径。

This Broadening Participation Research Initiation Grant in Engineering (BRIGE) grant provides funding to quantitatively assess the loss of efficiency that occurs in humanitarian and public health logistics systems in the absence of a centralized decision-making authority and to discover mechanisms to mitigate this impact by optimally designing system parameters. Centralized decision-making in such settings is unlikely, but decentralized system outcomes can be poor. This research will examine settings in which individuals choose a facility for service, for example, to pick up needed supplies after a disaster or be vaccinated against an infectious disease. Models will be created by integrating concepts from game theory and optimization to represent decisions of decentralized agents, who act according to different objectives but whose decisions impact overall system outcomes. New theoretical performance bounds will be developed to compare the performance of the decentralized system to a best-case system in which a centralized planner can assign individuals to facilities. Efficient algorithms will be sought to solve the decentralized problem. Finally, mechanisms will be designed to coordinate the decentralized system, or ensure that its performance more closely approximates a centralized one.If successful, the results of this research will lead to the creation of new optimization and game theory tools with significant potential to inform decision-making in humanitarian and public health logistics systems. The research will highlight opportunities for greater efficiency and effectiveness in these settings, improving the utilization of scarce resources and outcomes for affected populations. Analysis of the proposed models will also advance theoretical understanding of more general decentralized networks. The proposed research activities are integrated with a broadening participation plan that will create pathways to engineering careers for pre-college and undergraduate students from underrepresented groups by engaging these students in outreach and research activities that illustrate ways that engineers make a difference in society.