The specific criteria for design of systems for the remediation of con
taminated ground water can affect the cost, configuration, and feasibi
lity of the remedial system. Two important design criteria are the deg
ree of remediation required, as measured by water concentration, and t
he time by which compliance must be achieved, In this paper the impact
of these criteria on remedial design is studied for a hypothetical pu
mp-and-treat remedial system applied to a simplified model of a large
organic solvents plume in Tucson, Arizona. The analysis is performed b
y combining two-dimensional numerical flow and transport modeling with
nonlinear optimization, The use of optimization provides a means of s
ystematically varying design criteria in an experimental framework, Ba
sed on the data for this site? variation of the concentration standard
over one order of magnitude can affect the optimal cost by as much as
220%. Change in the time to achieve compliance over less than one ord
er of magnitude can affect optimal cost by as much as 175% and in some
cases can produce problems which are infeasible due to physical or te
chnical limitations, Under some conditions, a time to achieve complian
ce can be identified which has minimum cost, Changes in the design cri
teria are also shown to have significant impact on the distribution of
steady pumping rates among the individual wells.