CAREER: Constrained Ranking and Selection and Discrete Optimization via Simulation with Applications to Water Resource Allocation

职业：通过模拟进行约束排序和选择以及离散优化及其在水资源分配中的应用

• 批准号: 0644837
• 负责人: Seong-Hee Kim
• 金额: $ 40万
• 依托单位: Georgia Tech Research Corporation
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-06-15 至 2014-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0644837&HistoricalAwards=false
• 关键词: CAREER; Constrained; Ranking; Selection; Discrete

This grant supports the development of advanced statistical methods'called constrained ranking and selection (R&S) procedures'for selecting the best of a large number of alternative system designs. These competing designs are typically evaluated via computer simulation with respect to the expected value of a primary performance measure, subject to constraints on acceptable expected values of several secondary performance measures. All the relevant expected values of the selected primary and secondary performance measures must be estimated by sufficiently long runs of the corresponding simulation models. Further, this research investigates how to extend the resulting constrained R&S procedures for use within each search step of a simulation optimization algorithm; and how to combine the extended constrained R&S procedures with optimization search algorithms to handle a large search space. Finally, findings from this research will be applied to a real-world problem, namely, groundwater resource allocation in water resource planning and management. If successful, this research will develop generally applicable, statistically valid R&S procedures for discrete optimization under uncertainty both on objectives and constraints; help industry to reduce costs and improve overall decision-making processes; and provide a solid foundation for applying the new procedures to areas in which these techniques are not yet popular'in particular, the filed of environmental and water resources management, whose problems have an enormous impact on public policy/health. More specifically, the groundwater resource allocation problem for the Savannah Georgia region will be investigated from the perspective of environmental water resources planning. Therefore, the completion of the proposed work will have a significant impact not only on the theory and methodology of industrial engineering and operations research but also on environmental and water resources management.

该补助金支持先进的统计方法的发展“所谓的约束排名和选择（RS）程序”选择最好的大量的替代系统设计。&这些竞争设计通常通过计算机模拟相对于主要性能指标的预期值进行评估，并受到几个次要性能指标的可接受预期值的约束。所选主要和次要性能指标的所有相关预期值必须通过相应模拟模型的足够长的运行来估计。此外，本研究探讨如何扩展所得到的约束RS程序内使用的每个搜索步骤的仿真优化算法，以及如何联合收割机扩展约束RS程序与优化搜索算法来处理一个大的搜索空间。&&最后，本研究的结果将应用于一个现实世界的问题，即水资源规划和管理中的地下水资源分配。如果成功的话，这项研究将开发普遍适用的，统计上有效的R-S离散优化程序的不确定性下的目标和约束条件，帮助行业降低成本，提高整体决策过程，并提供了一个坚实的基础，应用新的程序，这些技术还没有流行的领域，特别是环境和水资源管理，其问题有巨大的影响，公共政策/健康。&更具体地说，地下水资源分配问题的萨凡纳格鲁吉亚地区将从环境水资源规划的角度进行调查。因此，拟议工作的完成将不仅对工业工程和运筹学的理论和方法产生重大影响，而且对环境和水资源管理也产生重大影响。