Cities are dealing with growing pressure from floods, heatwaves, and other hazards as urban areas expand and the climate changes. Disaster planning often examines weather risks, exposed communities, infrastructure, and transport access in separate stages. This paper reviews how artificial intelligence (AI), machine learning (ML), Geographic Information Systems (GIS), graph analysis, and explainable AI have been used in urban disaster preparedness and climate adaptation. It considers work on disaster classification, mapping of vulnerable locations, infrastructure dependencies, and emergency route selection. The review points to a practical gap: decision-makers need a shared view of hazard conditions, population exposure, essential services, and road access rather than separate outputs from disconnected tools. In response, a Smart Urban Resilience Intelligence Platform is outlined. Its proposed modules cover risk prediction, vulnerability assessment, resilience scoring, emergency routing, and preparedness recommendations. The paper also notes issues that would need attention before deployment, including data quality, the clarity of model explanations, links between infrastructure systems, and testing with real urban data. The framework is a proposal; its effectiveness would need to be established through implementation and validation.