In this thesis, I develop and evaluate a demand-side approach to transition from an intermittent to a continuous water supply using a district metered areas approach in Accra, Ghana, an urban metroplex of 4.0 million persons in western Africa. Intermittent water supply (IWS) is inherently inefficient. These systems foment health hazards and are expensive with an inherent low return on investment. Various causes and effects with positive feedback loops exacerbate intermittency. And building more robust IWS systems does not help because of rapid system degradation and high socio-economic costs render them unsustainable. Continuous water supply (CWS) systems are superior in every respect. Therefore, I view transitioning from IWS to CWS as the optimal choice for urban water districts. The general strategy to make this transition is to increase production and sharply improve the transmission capacity and efficiency of the drinking water system, the so-called supply-side approach. But because the supply side approach does not factor the underlying causes of IWS-- leakage and variable pressure levels, I used a demand-side approach for this study and designed a novel method, based on leakage theory with the application of district metered areas (DMAs). We tested this method on a case study performed in Accra, Ghana. The method is essentially a set of requirements and boundary conditions used to facilitate the transition from IWS to CWS from a demand-side perspective. A decision tree gives insight into the causes of IWS per DMA and the interventions required to reduce intermittency. Based on the foregoing, water engineers can build strategies to roll out a demand-side CWS systems in DMAs for an entire region. In our Accra study, we tested three DMAs using this approach with a focus on supply security. Data from those DMAs was collected and a top down NRW assessment was performed. We selected one DMA to collect and analyse flow and pressure data (minimum night flow, non-revenue water, billing, intermittency level, average zonal pressure). Furthermore, the supply conditions and proposed intervention (pressure adjustments) were hydraulically modelled for the district. Based on the DMA data, a transition strategy was developed for the Greater Accra Metropolitan Area. We also assessed the applicability of the demand-side approach to other areas through a survey among water supply specialists. The study shows that gains in water savings at the DMA level to more than justify the costs of transitioning to CWS. In cases of high pressure and high real loss volumes, pressure management improves supply conditions for customers. And, after attaining CWS, water recovered from loss reductions can be redistributed into neighboring districts, increasing their water availability. Because hydraulic pressure dependent demand (PDD) modelling cannot model leakage accurately, it was not possible to evaluate leakage reduction interventions. And it was not possible to create a working hydraulic model to assess the effect of this DMA approach for the entire Accra metroplex because of limited data availability at a district level, leakage parameters that were difficult to verify with field measurements, and the lack of a proven hydraulic PDD software able to distinguish between domestic demand and real losses. In conclusion, the developed method looks promising. But two factors limited fully exploring its real-world application: (1) the absence of mature modelling software; (2) detailed water district data. Developing a working hydraulic modelling software was beyond the scope of an MSc. thesis. But such hydraulic modelling software should be developed to help water utilities with limited technical and human resources improve services. Further research is required into leakage component modelling and validation from field measurements, after which interventions can be more properly and easily applied.