Containing groundwater contamination: Planning models using stochastic programming with recourse

Abstract

This paper examines the problem of operating pumping wells in order to contain an area of groundwater contamination when the aquifer properties of the area are uncertain. A stochastic program with simple recourse is formulated, involving a non-convex quadratic objective subject to linear constraints. This problem is solved using an extension to the Finite Generation Algorithm that will find an at least locally optimal solution to a problem involving a non-convex quadratic objective function. A numerical example is presented and analyzed.