During earthquakes, damage and collapse of structures pose physical risks but also social, economic, and environmental challenges. Even buildings that appear not significantly damaged may require demolition and reconstruction to ensure their resilience. The demolition of existing buildings not only results in the loss of property and lives but also contributes to the unsustainability of the building stock. To mitigate these challenges, development of a comprehensive framework is imperative, one that can seamlessly integrate vulnerability and sustainability parameters. This holistic approach aims to achieve a sustainable building stock that not only mitigates physical vulnerabilities but also can be used to address the social and economic aspects. This study presents an interpretable, adaptable, and transparent Rapid Visual Screening (RVS) method combining machine learning, fuzzy logic, and neural networks to assess existing buildings and prioritize them based on intervention requirements and promote building stock sustainability. Buildings with higher scores indicate lower seismic sustainability ratings, emphasizing the need for intervention and improvement to enhance their resilience. The implementation of this framework facilitates the development of a safe, resilient, and environmentally sustainable building stock. In its initial stages, the proposed RVS method achieved an accuracy rate surpassing both conventional methods and the baseline. Ultimately, its application extends beyond existing buildings, as the method can also be used during the design of new buildings in seismic-prone regions.