ABSTRACT:A groundwater quality modeling advisory system has been developed for the U.S. Air Force for use in investigating remediation alternatives for the cleanup of subsurface contamination. The system is capable of accounting for uncertainty, not only in the prediction of solute transport but also in the optimization of the remediation scheme through chance constraints. The system guides users in the selection of appropriate transport models through an algorithm independently tested with machine learning codes. An application to Hill Air Force Base, Utah, is presented for which different pump‐and‐treat strategies are considered: the results are evaluated in terms of the cumulative distribution of the contaminant concentration for each case and the tradeoff relationship between the cost of remediation and the probability that the remediation strategy exceeds an established maximum allowable contaminant concentrat