A hypothetical contaminated site was taken as basis for a 2D-flow and transport model, which is coupled with the heuristic optimization algorithm Derandomized Evolutionary Strategies (DES) with Covariance Matrix Adaptation (CMA). The focus is to test different alternative combinations of the optimization algorithm, fitness function and model to identify the best configuration of one or two pumping well(s). Three different contaminant degradation constants that exhibit three contaminant plumes were simulated. For each of these three degradation cases, one-well and two-well scenarios were examined. The optimization goals were modified by using a variable set of constraints, realized as penalty terms in the fitness functions of both scenarios. The obtained results demonstrate that DES is a robust optimization algorithm, capable of finding feasible and optimal or near-optimal solutions with appropriately imposed fitness function penalties for all of the three degradation cases in both scenarios.