Abstract

Water resource allocation problems have challenged water managers for decades. Allocation schemes often prove controversial and laden with conflict as competition among multiple water-users such as municipalities, industries and agriculture intensifies. Recently, increased population shifts, and shrinking water supplies have magnified the user competition. This competition will become even more aggravated if natural conditions become more changeable and as concern for water quantity and quality grows. A poorly-planned system for allocating water can become a serious problem under disadvantageous climate and river-flow conditions. For a long time, increasing water demand has been met by developing new sources of water. However, significant economic and environmental costs associated with developing new water sources have made this approach unsustainable. This has led decision-makers to the point that unlimited expansion can no longer be the primary objective in water resources management. Instead, for optimum water resource allocation, it is desired to improve the existing water allocation and management in a more efficient, equitable, and environmentally-benign manner by developing innovative environmental policy formulation techniques for water allocation under various complexities. Such environmental policy formulation can prove to be extremely complicated, since water systems generally contain considerable degrees of uncertainty. The abundance of stochastic uncertainty renders most deterministic optimization methods relatively unsuitable for practical implementation. While optimization-based techniques generally create single best solutions to problems, it is often preferable to generate several alternatives that provide multiple different perspectives to the same problem from an environmental policy formulation perspective. Preferably these alternatives would possess near-optimal objective measures, but would differ from each other in terms of the system structures characterized by their decision variables. In response to this option-generation requirement, several approaches collectively referred to as modelling-to-generate-alternatives (MGA) have been developed. Policy-makers can then perform a subsequent comparison of these alternatives to determine which option most closely satisfies their specific circumstances. Simulation-optimization (SO) is a family of information technology techniques which incorporate system uncertainties using probability distributions that have recently been used for optimal environmental planning. While SO holds considerable potential for application to a wide range of difficult stochastic problems, its solution search times are stochastic and vary considerably from one implementation to the next. In this study, techniques are introduced to reduce SO search times and it will be shown that this approach can simultaneously be used to create multiple policy alternatives meeting required system criteria. The efficacy of this MGA approach for policy formulation is illustrated using a water resource management case study. Since SO techniques can easily be adapted to many different stochastic problems, the practicality of this MGA approach can clearly be extended into many other environmental planning applications containing significant sources of uncertainty.