<p>Groundwater level contours and observation wells in Qazvin Aquifer, Qazvin Province, Iran.</p> Full article ">Figure 2
<p>Conceptual architecture of employed hidden layer perceptron (NFL = number of neurons in the first layer; NSL = number of neurons in the second layer; NAF = number of activation functions).</p> Full article ">Figure 3
<p>The check point locations (spaced 280 m apart).</p> Full article ">Figure 4
<p>Algorithm for improving existing monitoring networks (NOAW: Number of Additional Observation Well(s); MNAWs: Maximum Number of Added Wells; GA: Genetic Algorithm; SOE: Satisfaction of the Expert; and FIS: Fuzzy Inference System).</p> Full article ">Figure 5
<p>Flowchart of Genetic Algorithm model (BFV: the best fitness value; BPFV: the best previous fitness value).</p> Full article ">Figure 6
<p>Qazvin Aquifer candidate additional observation well locations (search space of Genetic Algorithm).</p> Full article ">Figure 7
<p>Fuzzy Inference System (FIS) process.</p> Full article ">Figure 8
<p>Membership function for NOAWs.</p> Full article ">Figure 9
<p>Membership function for installation cost of one well.</p> Full article ">Figure 10
<p>Membership function for satisfaction of expert.</p> Full article ">Figure 11
<p>Ef and RMSE as functions of NOAWs.</p> Full article ">Figure 12
<p>Fuzzification, inference, and defuzzification processes in determining the Satisfaction of the Expert (SOE) for each NOAW (for NOAWs = 9 and unit well cost = USD 4000, SOE = 61%).</p> Full article ">Figure 13
<p>Normalized Pareto optimum curve of Ef versus SOE for USD 4000 unit well cost (labels show NOAWs).</p> Full article ">Figure 14
<p>The groundwater level contour maps based on bargaining game results.</p> Full article ">