ABSTRACT
The choice of a ground water pollution remediation strategy is a complicated decision in a realm of uncertainty. In recent years a large amount of work has been ongoing to develop methods to aide in such decision making. Most of this work concentrated on optimizing a specific technology choice for a single realization of the parameter set. Some work has looked at optimizing a specific technology choice given uncertainty in the parameter set. However, there has been little work looking at the choice between different technologies, especially considering parameter uncertainty and reliability in an optimization context. That in part has been due to the computational burden associated with such problem solving. The work here is a preliminary investigation this problem: optimization between remediation strategies, under parameter uncertainty and considering strategy reliability for ground water pollution. Reliability and cost are the factors in the objective function used in a genetic algorithm optimization. Risk in terms of supplementary remediation costs is also included in the objective function. The results of the optimization are placed into a decision framework context to improve the transparency of the results of the optimization and the implications of low reliability. Risk taking and hedging paradigms as well as other decision making paradigms can be illustrated using decision analysis methods. Furthermore, the risk avoided or assumed by various positions is clarified by applying such methodologies.
