Using artificial neural network based decision support system for managing water resources.

There was a considerable rise in water consumption by public needs as a result of the increasing demand for water in light of limited water imports, resulting in a shortage in the amount of water resources and storage in the dam. Water import forecasting techniques were utilized to determine the suitable water stock as a result of the scarcity. Data mining techniques will be employed in this research to obtain a smart decision support model that consists of two stages based on natural water imports and for the aim of generating a rational management for the operation of the dam and control of water releases: Assessment and forecasting of the situation, as well as decision-making on the amount of water released In the first phase, time-series models based on Artificial Neural Networks (ANN) theory and the Propagation Algorithm (BP) algorithm were used, and in the second phase, classification techniques represented by the support vector classifier were used to extract predictions of water levels entering the reservoir. MSVM (Multiclass Support Vector Machine) is used to regulate and select the appropriate launch policy. An Artificial Neural Network (ANN) is favored because it has the lowest sum of squared errors (MSE). In the second stage, the researcher applied and used the vector support machine classification method to classify the tank's water discharge output, and discovered that the multiple booster transmission classification method (MSVM) is the most efficient based on the results because it has the highest accuracy in making decisions. The best for waste water and for water consumption needs such as drinking and cooking as needed (few, medium, high) (irrigation, agriculture, industry, electricity). [ABSTRACT FROM AUTHOR]