The limitation of machine learning methods for water supply and demand forecasting: A case study for Greater Melbourne, Australia
Problem
Cities facing water scarcity need accurate supply and demand forecasts for sustainable urban water management; the study tests ML efficacy for Greater Melbourne.
Why it matters
Poor forecasts can lead to water shortages or costly over-supply in growing cities.
What the AI did
Trained LSTM, stacked LSTM, Bi-LSTM and MLP models on 1990-2019 daily water supply/demand and climate data to forecast supply and demand.
Results
Stacked LSTM: supply R2=0.908, RMSE=335.74 ML, MAPE=23.5%; demand R2=0.791, RMSE=94.88 ML, MAPE=5.3%
Impact
Stacked LSTM performed best, but the title itself highlights limitations of ML for this forecasting task.
Record
| Venue | Water Science & Technology Water Supply |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Maryam Mohammadi, Shirley Gato-Trinidad, Kuok King Kuok |
| Citations | 11 (OpenAlex, at import time) |
| References | 46 |
| Model or method | Stacked LSTM (best); also LSTM, Bi-LSTM, MLP |
| Organization | Melbourne Water; Bureau of Meteorology (data providers) |
| Primary source | https://doi.org/10.2166/ws.2024.225 |
| Open access | hybrid |
| Keywords | LSTM, water demand forecasting, machine learning, Melbourne, water supply |
Importance rating
significant — thousands to millions affected
Abstract states many cities face water scarcity and accurate forecasting is critical for sustainable urban water management; tested on Greater Melbourne, models still limited on extremes.
More in Climate & Environment
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Learning skillful medium-range global weather forecasting | 2023 | 1,416 |
8
|
| 2 | Accurate medium-range global weather forecasting with 3D neural networks | 2023 | 1,631 |
8
|
| 3 | FuXi: a cascade machine learning forecasting system for 15-day global weather forecast | 2023 | 355 |
8
|
| 4 | Skilful nowcasting of extreme precipitation with NowcastNet | 2023 | 451 |
8
|
| 5 | Forecasting Renewable Energy Generation with Machine Learning and Deep Learning: Current Advances and Future Prospects | 2023 | 378 |
8
|