Genetic algorithm solution of groundwater management models

• Published in: Water resources research 1994-06, Vol.30 (6), p.1897-1906
• Main Authors: McKinney, Daene C., Lin, Min‐Der
• Format: Article
• Language: English
• ISSN: 0043-1397; 1944-7973
• DOI: 10.1029/94WR00554

Groundwater simulation models have been incorporated into a genetic algorithm to solve three groundwater management problems: maximum pumping from an aquifer; minimum cost water supply development; and minimum cost aquifer remediation. The results show that genetic algorithms can effectively and efficiently be used to obtain globally (or, at least near globally) optimal solutions to these groundwater management problems. The formulation of the method is straightforward and provides solutions which are as good as or better than those obtained by linear and nonlinear programming. Constraints can be incorporated into the formulation and do not require derivatives with respect to decision variables as in nonlinear programming. More complicated problems, such as transient pumping and multiphase remediation, can be formulated and solved using this method. The computational time required for the solution of genetic algorithm groundwater management models increases with the complexity of the problem. The speedup attainable by solving genetic algorithm problems on massively parallel computers is significant for problems where the simulation time required to complete each generation is high.