Simulation/optimization modeling for robust pumping strategy design.

Area of Science:

• Environmental Engineering
• Hydrogeology
• Computational Modeling

Background:

• Aquifer parameter uncertainty poses challenges for effective groundwater management.
• Traditional optimization methods may not adequately address system variability.
• Developing robust strategies is crucial for achieving long-term management goals.

Purpose of the Study:

• To present a novel simulation/optimization modeling approach, the Robustness Enhancing Optimizer (REO), for addressing uncertain aquifer parameters.
• To maximize the robustness of pumping strategies while optimizing a primary objective function, such as cost.
• To provide a method for developing reliable management strategies when extensive field data for statistical analysis is unavailable.

Main Methods:

• Coupling genetic algorithm and tabu search as optimizers.
• Incorporating aquifer parameter sensitivity analysis to guide multiple-realization optimization.
• Applying the REO to trinitrotoluene and Royal Demolition Explosive plumes at Umatilla Chemical Depot.

Main Results:

• The REO efficiently develops robust least-cost pumping strategies.
• The approach maintains the optimal value of the primary objective function as strategy robustness increases.
• Demonstrated effectiveness in a case study involving contaminant plumes.

Conclusions:

• The REO offers a valuable tool for groundwater management in data-scarce environments.
• It enhances strategy robustness without significantly degrading the primary objective function.
• REO-developed strategies provide a practical approach to managing complex hydrogeological systems with uncertainty.