Short-term water demand forecasting using data-centric machine learning approaches

Accurate water demand forecasting is the key for urban water management and can alleviate system pressure brought by urbanisation, water scarcity and climate change. However, existing research on water demand forecasting using machine learning focused on model-centric approaches, where various forecasting models are tested to improve accuracy. The study undertakes a data-centric machine learning approach by analysing the impact of training data length, temporal resolution and data uncertainty on forecasting model results. The models evaluated are Autoregressive (AR) Integrated Moving Average (ARIMA), Neural Network (NN), Random Forest (RF) and Prophet. The first two are commonly used forecasting models. RF has shown similar forecast accuracy to NN but has received less attention. Prophet is a new model that has not been applied to short-term water demand forecasting, though it has had successful applications in various fields. The results obtained from four case studies show that (1) data-centric machine learning approaches offer promise for improving forecast accuracy of short-term water demands; (2) accurate forecasts are possible with short training data; (3) RF and NN models are superior at forecasting high-temporal resolution data; and (4) data quality improvement can achieve a level of accuracy increase comparable to model-centric machine learning approaches.

Automated summary

Accurate water demand forecasting is crucial for urban water management in the face of urbanization, water scarcity, and climate change. This study evaluates the impact of training data length, temporal resolution, and data uncertainty on forecasting model results using a data-centric machine learning approach. The results show that data-centric machine learning approaches have the potential to improve the accuracy of short-term water demand forecasts, even with limited training data. The Random Forest and Neural Network models outperform other models when it comes to forecasting high-temporal resolution data, and improving data quality can achieve accuracy increase comparable to model-centric machine learning approaches.