State-of-art implementation of computing intelligent models for water demand modelling: A decade review and future direction

The main insight from this research is that there has been significant progress in the development of computer-aided models for water demand modelling over the past two decades. These models are used to balance water demand and supply, which is critical for effective water supply management systems. The equilibrium is achieved through various measures, many of which involve the use of forecasting tools. Recent research on urban water demand forecasting using artificial intelligence (AI) models is discussed in this article, to present the ‘state of the art’ on the issue and provide some insights and suggestions for future research on methodologies and models. The review examines models developed using traditional statistical methods, including artificial neural networks, linear regression, and time-series analysis, as well as soft computing techniques. This paper demonstrates that the study is focused on a decade-long evaluation of operating system management, indicating an opportunity for long-term projections. It goes without saying that no single model outperforms all the others; however, it is vital to assess the strengths of each model or combination of models for each country or region to determine which model works best in that location. Although the usage of AI and machine learning (ML) has increased significantly in recent decades, there is still potential for development in the field of water demand forecasting.