The scarcity of water resources has become a major global issue, exacerbated by the effects of climate change, population growth, and rapid urbanization. This issue is particularly worrying in arid and semi-arid regions, such as Morocco, where renewable water resources per capita are in constant decline, and preserving this existing resource is becoming a strategic priority. One of the essential levers to achieve this lies in reducing losses within drinking water distribution networks. This study is part of this logic and aims to optimize the reduction of technical losses in the drinking water network of the city of Casablanca. The analysis revealed an annual volume of losses exceeding 50 million m³, requiring targeted and prioritized actions. To this end, a two-step methodology was implemented: a water balance according to the standards of the International Water Association (IWA), followed by the development of a decision-making tool based on a multi-criteria risk analysis. The tool was designed in Python in the Visual Studio environment, integrating technical, historical, topographical, and operational criteria to produce a risk index for each network section. Field application helped identify high-risk areas and guide targeted interventions. A concrete case demonstrated the effectiveness of this approach: in an identified critical area, corrective actions resulted in the recovery of 300,000 m³ of water, representing an estimated saving of 1.6 million dirhams. This work confirms the potential of an integrated approach combining technical analysis, modeling, and decision support to improve network performance. The proposed tool is reproducible, scalable, and can be adapted to other urban networks facing the same challenges. It is fully in line with the dynamics of sustainable resource management and smart governance of water services.