Li Y P, Huang G H, Nie S L, Liu L
Department of Civil and Resource Engineering, Dalhousie University, Halifax, Nova Scotia, Canada B3J 1Z1.
J Environ Manage. 2008 Jul;88(1):93-107. doi: 10.1016/j.jenvman.2007.01.056. Epub 2007 May 25.
In this study, an inexact multistage stochastic integer programming (IMSIP) method is developed for water resources management under uncertainty. This method incorporates techniques of inexact optimization and multistage stochastic programming within an integer programming framework. It can deal with uncertainties expressed as both probabilities and discrete intervals, and reflect the dynamics in terms of decisions for water allocation through transactions at discrete points of a complete scenario set over a multistage context. Moreover, the IMSIP can facilitate analyses of the multiple policy scenarios that are associated with economic penalties when the promised targets are violated as well as the economies-of-scale in the costs for surplus water diversion. A case study is provided for demonstrating the applicability of the developed methodology. The results indicate that reasonable solutions have been generated for both binary and continuous variables. For all scenarios under consideration, corrective actions can be undertaken dynamically under various pre-regulated policies and can thus help minimize the penalties and costs. The IMSIP can help water resources managers to identify desired system designs against water shortage and for flood control with maximized economic benefit and minimized system-failure risk.
在本研究中，开发了一种不精确多阶段随机整数规划（IMSIP）方法，用于不确定性条件下的水资源管理。该方法在整数规划框架内结合了不精确优化和多阶段随机规划技术。它能够处理以概率和离散区间表示的不确定性，并通过在多阶段背景下完整情景集的离散点处进行交易来反映水资源分配决策中的动态变化。此外，当承诺目标未达成时，IMSIP可以促进对与经济处罚相关的多种政策情景的分析，以及对多余调水成本中的规模经济的分析。提供了一个案例研究以证明所开发方法的适用性。结果表明，已针对二元变量和连续变量生成了合理的解决方案。对于所有考虑的情景，可以在各种预先规定的政策下动态地采取纠正措施，从而有助于将处罚和成本降至最低。IMSIP可以帮助水资源管理者针对水资源短缺和防洪确定理想的系统设计，以实现经济效益最大化和系统故障风险最小化。
