Forecasting water demand is considered essential for managing water resources to achieve sustainable development. This research analyzes 50 research papers focusing on different aspects of water demand forecasting. Various methodologies were used in this review, including traditional statistical models (regression model, time series analysis), Artificial intelligence models are represented by (artificial neural networks (ANN), Support Vector Machines (SVM)), or hybrid models that use a combination of different methods to improve prediction accuracy. In addition to Exploring machine learning algorithms that allow the computer to learn and train from previous data without the need for explicit programming. In general, this review provides a comprehensive overview of previous research trends and methodologies used to forecast water demand and what challenges researchers faced in arriving at the results of their studies. It is considered a valuable resource for those entering the field of water resources management and planning.