Because of the global warming, urban planning strategies must be investigated to reduce the building energy consumption and increase the thermal comfort in cities. In the framework of Energy Strategy 2050 of Switzerland, this research aim to highlight the impact of future climate change on urban planning and proposes strategies to help urban planners and policymakers face this new challenge particularly in a future where heat waves are going to become common at mid-latitudes. However, to do so in the best possible way, the models currently used have to be robust enough in complex regions (with lakes and mountains) to evaluate future planning scenarios. First, this study proposes a methodology to cluster buildings in urban areas and use it in a mesoscale numerical weather prediction system to evaluate the urban heat island and its impact on the energy demand for cooling at the city scale. Second, this research proposes urban planning strategies to reduce building energy consumption due to the urban heat island phenomenon in the city of Lausanne. The model has been validated with six different weather stations. It emphasizes the presence of the urban heat island in the Lemanic region which lead to the increase of air temperatures in Geneva and Lausanne by 2°C during the day and by up 4°C during the night compared to the surrounding areas. The presence of multiple cities around the lake lead to the creation of the urban heat archipelago. We demonstrate with these simulations that the performance of the WRF model is very heterogeneous in this complex setting. In the evaluation of the future urban densification scenario, we noted that there will be an increase in temperature caused by the UHI phenomenon by up 0.6°C. The increase of albedo of roofs and walls and the use of greenings are good mitigation solutions, with a most significant impact for the albedo. In addition to the climatic impact, the model highlights that the densification scenario has a significant impact on the peak cooling demand.