Our research in this area is intended to develop a comprehensive simulation-optimization framework and computer codes for groundwater resource management and remediation design optimization. In our work, we have focused on the modern global optimization methods, such as genetic algorithms, simulated annealing and tabu search, because of their ability to identify global optimum, their efficiency in handling discrete decision variables such as well locations, and the ease and generality with which they can be linked to any flow and transport simulation models. We have also used traditional gradient based local optimization methods, such as differential dynamic programming, because of their advantage in dealing with a large number of decision variables and management periods. We have implemented the various optimized techniques on the commonly used flow and transport simulation codes MODFLOW and MT3D, and have successfully applied them to sites with complex hydrogeologic conditions.
Return to Research Projects Main Page
Last modified: January 27, 2002