Urban water demand prediction based on climate change is always challenging for water utilities because of the uncertainty that results from a sudden rise in water demand due to stochastic patterns of climatic factors. For this purpose, a novel combined methodology including, firstly, data pre-processing techniques were employed to decompose the time series of water and climatic factors by using empirical mode decomposition and identifying the best model input via tolerance to avoid multi-collinearity. Second, the artificial neural network (ANN) model was optimised by an up-to-date slime mould algorithm (SMA-ANN) to predict the medium term of the stochastic signal of monthly urban water demand. Ten climatic factors over 16 years were used to simulate the stochastic signal of water demand. The results reveal that SMA outperforms a multi-verse optimiser and backtracking search algorithm based on error scale. The performance of the hybrid model SMA-ANN is better than ANN (stand-alone) based on the range of statistical criteria. Generally, this methodology yields accurate results with a coefficient of determination of 0.9 and a mean absolute relative error of 0.001. This study can assist local water managers to efficiently manage the present water system and plan extensions to accommodate the increasing water demand.

Accurate prediction of short-term water demand plays an important role for water suppliers as well as for government's water plan. This paper aims to predict a municipal water demand for an upcoming year based on previous water consumption in Baghdad city. We have investigated various signal processing approaches to address the noisy time series data of water consumption, while a new methodology for short-term prediction of municipal water consumption has been proposed. This would enable us to forecast the short-term municipal water demand using different windows and multi-stages of hybrid univariate singular spectrum analysis and autoregressive model (SSA-AR model). First, different windows and multi-stages of SSA are utilised to analyse and clean the original water time series from noise. Then, the autoregressive (AR) model is employed to predict water demand based on the treated water consumption time series. In this study, monthly water consumption data from (2006-2015) for Al-Wehda treatment plant in Baghdad city, Iraq is selected to assess the model. The findings show that (SSA-AR model) can predict water demand with high accuracy from high noisy raw data.

Accurate forecasting techniques for a stochastic pattern of water demand are essential for any city that faces high variability in climate factors and a shortage of water resources. This study was the first research to assess the impact of climatic factors on urban water demand in Iraq, which is one of the hottest countries in the world. We developed a novel forecasting methodology that includes data preprocessing and an artificial neural network (ANN) model, which we integrated with a recent nature-inspired metaheuristic algorithm [marine predators algorithm (MPA)]. The MPA-ANN algorithm was compared with four nature-inspired metaheuristic algorithms. Nine climatic factors were examined with different scenarios to simulate the monthly stochastic urban water demand over 11 years for Baghdad City, Iraq. The results revealed that (1) precipitation, solar radiation, and dew point temperature are the most relevant factors; (2) the ANN model becomes more accurate when it is used in combination with the MPA; and (3) this methodology can accurately forecast water demand considering the variability in climatic factors. These findings are of considerable significance to water utilities in planning, reviewing, and comparing the availability of freshwater resources and increasing water requests (i.e., adaptation variability of climatic factors).