The rapid growth of waste electrical and electronic equipment (WEEE) is one of the most pressing environmental issues and Electronic Equipment (WEEE) is one of the most pressing environmental challenges facing smart cities, requiring predictive and sustainable management strategies. This study validates the development of an intelligent predictive system for WEEE management in Guayaquil (Ecuador). The model integrates ambient intelligence (AmI) with artificial neural networks (ANN), which are trained using a representative dataset of sociodemographic and technological consumption variables in order to estimate short- and medium-term WEEE generation. A rigorous methodological framework was adopted. This framework combines hold-out, K-fold cross-validation and bootstrap sampling. The results were evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and the coefficient of determination (R2) was also used for evaluation. The results demonstrate high predictive accuracy (R2 ≈ 0.97) and a strong capacity for generalisation across the validation techniques, confirming the robustness of the ANN approach. Additionally, an interactive web interface was also developed to allow visualisation of monthly and yearly projections, as well as the identification of the most polluting zones and sectors. This supports evidence-based decision-making by municipal authorities. The findings highlight the potential of AmI-based predictive systems as a scalable tool to enhance urban sustainability and optimise waste management logistics. They also promote citizen participation. Future extensions will focus on incorporating real-time IoT data, advanced deep learning techniques and system replication in other Latin American cities.