Location selection for Installation of Surface Water Treatment Plant by Applying a New Sinusoidal Analytical Hierarchy Process: Application of new MCDM in Location Detection

Water treatment plants (WTPs) are responsible for ensuring supply of healthy water to urban and rural consumers for drinking and other related purposes. But the arbitrary selection of a location for installation or relocation of WTPs often fails the purpose of the plant. Presently studies in location selection for water treatment plant are rare. Multi-criteria decision making (MCDM) methods and bagged polynomial neural networks (PNN) were found to be exemplary and easy to use tools for prediction, simulation and optimization of decision-making objectives. The present study tries to apply the advantages of MCDM and bagged PNNs in the identification of an ideal location for a surface water treatment plant. The most significant parameter is found to be WQI which represents the overall quality of water suitable for domestic use. The PNN models were developed with all the selected eight alternatives as input and output. The algorithms like GMDH, SFS, SMS, and QC were used to estimate the weight of connections in between the input and hidden; and hidden and output layers separately for each segment. The application of these two soft computation tools provides an opportunity to the decision maker in the selection of optimal location with the help of an objective and cognitive method.