A Bayesian decision rule for remediation actions at toxic waste sites

The US Environmental Protection Agency (EPA) recommends the use of an upper confidence limit in making a decision of whether a possibly polluted environment needs clean-up. This decision rule, however, is frequently too conservative and does not take into account the costs and/or benefits from making a correct or a wrong decision. In this paper we propose an asymmetric loss function and a Bayesian decision rule for remediation actions. The new loss function accounts for both false-positive and false-negative errors possibly involved in a decision, and accommodates the needs of both the EPA and other parties involved.