Very limited efforts have been made in the past specifically in India to utilise ANN for water quality modelling
purposes. It is understood that the estimation of water quality of a river at any location is a tedious work due to the
non-linear behaviour of different water quality variables. The ANNs have been successfully used in many
hydrological studies and this was a motivating factor for its application to the present study. The performance of
ANN was tested using RMSE, R and DC. It was found that the ANN approach turned out to be an efficient approach
for water quality modelling. Even more accurate predictions of DO could be obtained with a shorter time base input
data, e.g., 10-daily or daily data. An added advantage of ANN is that it does not require any assumption about the
range of flow discharge, temperature, BOD and DO. However, the input data should be consistent and the
controlling factors should be the same for training and test data.