Prediction of monthly consumption of drinking water in the Sistan region under climate change impact

Water scarcity has become a problem of concern for many cities in the world. Predicting water demand helps policymakers and water suppliers maintain the balance between the supply and demand of urban water resources, thereby preventing water wastage and shortage. Forecasting Urban Water Consumption (UWC) has a significant impress on efficient urban water management in cities in arid regions when considering the implications of climate change.  In this research, five firefly algorithms were used to estimate the consumption of drinking water in the Sistan region for the years 2006-2020 and were compared. The data from 2006 to 2015 were used for training and learning and finding the optimal weight of the model, and the remaining data from 2016 to 2020 were used to test the model. The results of the model showed that 5 different FA models can get possible answers. In the exponential and hybrid models, the relative error in the NDFA algorithm is 0.19, which has the lowest relative error among other algorithms, and in the linear model, the VSSFA algorithm has the lowest relative error with a relative error of 0.196. Therefore, the exponential model and NDFA method have better performance than other models and algorithms. And its prediction accuracy is above 81%. After ensuring the accuracy of the algorithm, the consumption of drinking water was predicted for the years 2023, 2024, and 2025. The results showed that the peak consumption is in July and August and the total consumption in 1402, 1403, and 1404 is equal to 7293, 7558, and 7674 thousand cubic meters, respectively.