In many places in Uganda, people do not have a connection to the drinking water supply system and there is a lack of treated water supply, meaning that people only have access to water a certain part of the day. As a result many people rely on springs, handpumps, rivers or lakes, of which the quality cannot be monitored or controlled.

During this multi-disciplinary project, we worked together with the National Water & Sewerage Corporation (NWSC) and the Ministry of Water and Environment (MWE) to research the possibilities of extending the water supply system of two project areas, Bugiri District and Hoima City. The current water supply in both areas use groundwater as a source and the possibilities for the extension also consider using surface water besides groundwater.

The different alternatives for the extension of the water supply in Hoima City and Bugiri District are evaluated using a multi-criteria analysis (MCA), consisting of a financial analysis, a performance analysis and a risk analysis. By evaluating the different options using an MCA, the decision-making process can become less complicated.

The MCA-tool that is set up in this research can be used by engineers to study different areas in Uganda and make it easier to compare different options for the extension of a drinking water supply system in an early design stage. The tool is for the two project areas as examples, after which it is also tested during a case study with engineers from both NWSC and MWE. Useful feedback came out of this session which will be used to finalize the tool and elaborate on it.

To design the different alternatives for the project areas and to get insight into the drinking water supply of Uganda, Hoima and Bugiri are visited at the beginning of the project.

For both project areas, it is recommended to improve the operational performance of the already existing groundwater supply system as a short-term (5 years) solution. The long-term (25 years) solutions consider groundwater options as well as surface water options, using for example Lake Victoria, Lake Albert and River Nile as water sources.

In this research the possibilities of the application of machine learning models at ‘Hoogheemraadschap van Delfland’ are studied. A random forest (RF) and an LSTM model are used for the prediction of the sum of the discharge in the next 2, 8 and 12 hours from the polders to the boezem canals. This research has showed the potential of machine learning models for the prediction of discharge for the considered pumping stations in the case area. This case area is clustered in the Sobek RR model as node 49. The RF and LSTM model are compared to the current Sobek RR model, the machine learning model of Delfland (ReRengAI) and a naïve model by calculating the root mean squared error (RMSE) for the last year of the dataset. For the prediction of the 2 hourly sum of Node 49 the RF model performs the best. Additionally, the performance of the RF model for the 12 hourly sum is satisfactory with a RMSE of 11,071 m3, though using a deep learning model (LSTM) the performance improved to a value of 10,181 m3 for the RMSE. Machine Learning models are known as black-box models and are hard to explain and interpret, which makes the practical implementation of these new models, despite good model results, challenging. Technical recommendations for implementation ML models are improving the quality and availability of the data, increasing the interpretability and explainability of the model, combining multiple objectives in the new model or combining a ML model with a physical model. Organizational recommendations are improving the knowledge about these models within the organization, studying the advantages of these models in comparison to the current model and involving different departments of the water authority in the development of these new models.