Fuzzy optimization model and fuzzy inference system for conjunctive use of surface and groundwater resources

Sustainable water resources management in most river basins needs appropriate techniques to implement the conjunctive use of surface and groundwater resources with consideration of uncertainties. In this study, first a linear fuzzy optimization model was used to find the optimal surface and groundwater withdrawal. Then, by using the results of this model, a Fuzzy Inference System (FIS) was developed to determine the groundwater withdrawal, automatically. The capability of this approach is studied in the Astaneh-Kouchesfahan Plain in north of Iran. To do this, the groundwater of this Plain was simulated using MODFLOW code. Then, two fuzzy optimization methods were developed to minimize the water shortage. Results showed that the average water shortage was equal to 215 and 138 MCM according to current water withdrawal and the best solution for the fuzzy optimization model which was 22.0% and 14.6% of the water demand, respectively. The results of the optimization model were used to predict the optimal water withdrawal of the aquifer using the FIS, considering the predictor variables. These variables were applied to the developed FIS model under different scenarios in order to determine the best scenario with the highest prediction performance for water withdrawal from the aquifer. The evaluation criteria were coefficient of determination, normalized root mean square error, and the mean absolute error which were equal to 0.94, 0.26 MCM and 3.8 MCM, respectively for the best scenarios. These scenarios were able to predict the optimal amount of groundwater withdrawal and can replace the numerical optimization methods.