The prediction of the blue water footprint in water services such as in water treatment plants (WTPs) is non-trivial to water resource management. Currently, the sustainability of water resources is of great concern globally, particularly in addressing the 6th goal of the United Nation's Sustainable Development Goals (UN SDGs). This study focuses on the blue water footprint (WF_blue) assessment and prediction of WTP located at the Kuantan River Basin, Malaysia. The intake water of WTP is directly obtained from the mainstream river within the basin known as the Kuantan River. The predictability of the WF_blue was evaluated by means of Artificial Neural Networks (ANN) and Support Vector Machine (SVM). Different hyperparameters of both the ANN and SVM models were investigated to ascertain the best prediction models attainable by evaluating both the mean squared error (MSE) as well as the coefficient of determination, R. It was demonstrated from the study that the optimised ANN model is able to yield a better prediction performance in comparison to the optimised SVM model. Therefore, it could be concluded that the application of ANN to predict the future trend is pertinent and should be incorporated in water footprint studies as it is vital for water resources regulators to anticipate the condition of WF_blue in the future and to line up the appropriate actions especially in controlling the influencing parameters namely, water intake, rainfall and evaporation.

水处理厂 (WTP) 等供水服务中蓝水足迹的预测对于水资源管理来说意义重大。当前，水资源的可持续性受到全球关注，特别是在解决联合国可持续发展目标（UN SDGs）的第六个目标方面。本研究的重点是位于马来西亚关丹河流域的 WTP的蓝水足迹（WF _blue）评估和预测。水厂取水直接取自流域内的主流河流，即关丹河。WF_蓝的可预测性通过人工神经网络 (ANN) 和支持向量机 (SVM) 进行评估。研究了 ANN 和 SVM 模型的不同超参数，以通过评估均方误差 (MSE) 以及决定系数 R 来确定可获得的最佳预测模型。研究表明，优化的 ANN 模型是与优化的 SVM 模型相比，能够产生更好的预测性能。因此，可以得出结论，应用 ANN 来预测未来趋势是相关的，应纳入水足迹研究，因为水资源监管机构预测 WF_蓝的状况至关重要。 并采取适当的行动，特别是在控制影响参数方面，即取水量、降雨量和蒸发量。