<p>Major pandemic outbreaks have profound global social and economic consequences, underscoring the urgency for cities to adopt adaptive preparedness strategies. During such outbreaks, municipal governments often implement broad containment measures, including mobility restrictions. While these measures are crucial for protecting public health, their application without a nuanced understanding of local pandemic conditions can disrupt socioeconomic systems and disproportionately burden communities. Effective strategies must therefore integrate localized assessments of pandemic vulnerability and precise, neighborhood-level responses to mitigate adverse social and economic impacts. Using data from the COVID-19 pandemic outbreak, this research develops a dual-focused framework to evaluate urban pandemic vulnerability through the intersecting lenses of infection risk and vaccination accessibility. The proposed framework integrates the Spatially-Tuned Poisson Regression machine learning technique to empirically calibrate the distance-decay parameter, allowing the vaccination accessibility model to better reflect the average pull of vaccination sites.​​ Applied to the high-density urban context of Hong Kong, the analysis identifies localized vulnerability patterns at the smallest census unit level, revealing heightened pandemic vulnerabilities in areas of Kowloon, northern Hong Kong Island, and select new towns in the New Territories. These findings demonstrate how spatially explicit assessments prioritize neighborhoods for targeted risk identification and mitigation strategies, offering a transferable framework for precise outbreak response during future pandemic outbreaks.</p>