This study addresses the challenge of optimizing charging infrastructure for electric vehicles (EVs) in urban areas of Morocco, focusing on the cities of Casablanca and Marrakech. The aim is to develop an effective planning model that integrates various technical factors, such as bus stations, gas stations, hotels, as well as main and secondary roads. Using ArcGIS software, we first calculated the Euclidean distance of each factor across the city of Casablanca. Then, by applying the Analytic Hierarchy Process (AHP) method, we assigned weights to each factor. These weights were used to evaluate the potential of each pixel in our study area, by multiplying the Euclidean distance by the corresponding weight. For predictive modeling, we compared two popular machine learning models: Random Forest (RF) and XGBoost. Our results showed that the RF model outperforms XGBoost in terms of accuracy and reliability for this specific application case. Finally, we applied our optimized model to a new city, Marrakech, thus demonstrating its ability to be adapted and implemented in other urban contexts in Morocco. This research offers an innovative and practical approach to guide the strategic deployment of EV charging infrastructure, thereby supporting the growing adoption of electric vehicles in Moroccan urban areas.