<![CDATA[Regional Drinking Water Company (PDAM) is a company engaged in clean water supply services, one of the goals of the establishment of PDAM is to meet the needs of the community for clean water, including the provision, development of facilities and infrastructure services, as well as clean water distribution. PDAM Tirta Sari Binjai as the person in charge of clean water needs in Binjai City installs water connection installations at customers' homes which usually come from the mountains then flows into rivers and is collected first in reservoirs and then distributed to customers' homes . PDAM Tirta Sari Binjai has experienced problems in handling the large number of applications for new connection installations. And to overcome these problems we need a method that can predict the number of new installations. Based on these conditions, PDAM Tirta Sari Binjai needs to create a system that can predict the number of installations of clean water lines that will come in the following days or months. The process of predicting the number of installations of clean water lines that will be carried out with a computerized system, one of the processes that can be done is the application of an Artificial Neural Network (ANN) using the Backpropagation method. The system is designed with the MATLAB R2014a programming application, after carrying out the data training process and data testing on 2016 to 2020 data, the learning rate is 0.2; the maximun epoch is 10000 and the target error is 0.001, the result is that in 2021 the number of installations of clean water channels is 2,238 channels. Keywords: Backpropagation, Artificial_Syaraf_Network, Clean_Air_Channel]]>
