Production of Potential Blockage Event Map for Urban Sewer Networks Using Neural Network and GIS (Case Study: Region 2 of ABFA of Tehran City)

Wastewater network as one of the most important infrastructure facilities can play an important role in achieving sustainable development by improving public health and environmental protection by preventing pollution of surface and groundwater resources. One of the most common incidents that occurs significantly in this network is blockage of the sewer pipes. Recognizing the factors influencing the occurrence of network blockage has a significant impact on accurately predicting what may happen to the network in the future. In the present study, GIS tools and artificial neural network were used to predict and mapping the potential for blockage in the sewer network. Important factors in the occurrence of blockage including, land use, siphon diameter, siphon depth, depth, materials and age of the pipe were used in neural network analysis. From input data of 70%, 15% and 15%, respectively, were used for training, validation and model testing. The results of the simulation using a neural network with a Performance Indicator of R2=0.9 showed a high fitness between the predicted and observed locations of the blockage. Also, in the blockage potential map, areas with high population density, worn texture and unauthorized constructions (due to the installation of unauthorized branches) were observed blockage potential.