An artificial intelligence-based model for optimal conjunctive operation of surface and groundwater resources

A hybrid simulation-optimization model is proposed for the optimal conjunctive operation of surface and groundwater resources. This second-level model is created by finding and combining the best aspects of two resilient metaheuristics, the moth swarm algorithm and the symbiotic organization search algorithm, and then connecting the resulting algorithm to an artificial neural network simulator. For assessment of the developed model efficiency, its results are compared with two first-level simulation-optimization models. The comparisons reveal that the operation policies obtained by the developed second-level model can reliably supply more than 99% of the total demands in the study regions, indicating its superior efficiency compared to the two other first-level models. In addition, the highest sustainability index in the study regions belongs to the proposed model. Comparing the results of this research with those of other recent studies confirm the supremacy of the developed second-level model over several previously developed models. Towards optimizing the conjunctive operation of surface and groundwater resources in arid and semi-arid regions, here the authors propose a hybrid method involving moth-swarm and symbiotic organism search algorithms and artificial neural networks and demonstrate it for the HalīlRood basin.