Groundwater quality forecasting modelling using artificial intelligence: A review

This review paper closely explores the techniques and significances of the most potent artificial intelligence (AI) approaches in a concise and integrated way, specifically in the groundwater quality modelling and forecasting for its suitability in domestic usage. This paper systematically provides an extensive review of the four most used AI methods: artificial neural network (ANN), adaptive network-based fuzzy inference system (ANFIS), evolutionary algorithm (EA) and support vector machine (SVM), to reflect on the features and abilities while defining the greatest challenges throughout the process of providing desired results. Analysis among the four AI methods found that ANN performed better when handling a large number of data sets and accurately made predictions due to its ability to model complex non-linear and complex relationships, despite some weaknesses. The findings of this review demonstrate that the successful adoption of AI models is impacted by the appropriateness of input consideration, types of individual functions, the efficiency of performance metrics, etc. The outcomes from this study will be beneficial for groundwater development plans and contribute to the improvement of the AI applications in groundwater quality. Recommendations are presented in this study to strengthen the knowledge development towards improving the modelling structure in the mentioned area.