Nihonbashi in Tokyo is a high-density urban environment that faces increasing risks of disaster such as heatwaves, yet current disaster management plans lack dynamic, real-time response mechanisms to address uneven vulnerability across neighborhoods. This study enhances Nihonbashi’s urban heat resilience using urban digital twins, a dynamic virtual replica integrating real-time data to simulate and optimize urban systems. Employing a 4-step scenario planning model (descriptive, evaluative, predictive, prescriptive), it integrates real-time heat risk, resilience hub occupancy, and social demographic data to optimize access to cool spaces and transportation routes. Leveraging open-sources tools like Network X, OSMnx, and Getis-Ord Gi*, the framework identifies high-risk zones such as office headquarters and subway stations to identify vulnerability hotspots and simulates urban network performance during heatwaves. A heat scenario classifier achieves 96.7% accuracy in predicting heat risk levels. Built on ArcGIS Experience Builder, the platform enables dynamic rerouting to less occupied shelters and shaded pedestrian pathways, prioritizing vulnerable populations, particularly the elderly. Unique contributions include real-time data integration, high-accuracy heat prediction, and an equity-focused approach, distinguishing it from static GIS-based simulations. Data from OpenStreetMap, PLATEAU, and e-Stat ensure reliability, although real-time data access poses challenges. Stakeholders, including planners, emergency responders, and residents, can engage via the interactive platform to simulate scenarios and enhance resilience. This scalable, open-source framework offers a transformative model for urban heat management adaptable to other cities.