Dealing with uncertainty in water resource planning is problematic because insufficient or underused infrastructure can have social and environmental costs. Multistage stochastic optimisation provides a mechanism to deal with this challenge in water supply capacity expansion planning. However, for real systems it can be mathematically hard and computationally expensive. The ‘Decision-rule’ formulation represents an attempt to remedy this by approximating the multistage problem where decisions at each stage are a function of the uncertainty and the state of the system. We introduce a family of rules to show how they approximate the multistage problem and investigate the implications of the approximation for adaptive water resources planning.

处理水资源规划中的不确定性是有问题的，因为基础设施不足或未充分利用会产生社会和环境成本。多阶段随机优化提供了一种机制来应对供水扩容规划中的这一挑战。然而，对于实际系统来说，它在数学上可能很困难，而且在计算上很昂贵。“决策规则”公式代表了通过近似多阶段问题来解决这一问题的尝试，其中每个阶段的决策都是不确定性和系统状态的函数。我们引入了一系列规则来展示它们如何近似多阶段问题，并研究近似值对自适应水资源规划的影响。