Cooling demand for indoor thermal comfort is expected to increase as hot days are increasing in temperature and becoming more frequent across the globe. As urban residential buildings and neighbourhoods are increasingly subject to such excessive heat events, urban dwelling can become vulnerable to heat stress if the cooling demand cannot be met for reasons such as power outages or summer fuel poverty. This thesis investigates how data-driven peak cooling energy demand modelling can be developed for assessing heat stress vulnerability (HSV) of a city's residential building (dwelling) stock in the future climate. The hypothesis is twofold: (1) that the predicted future indoor peak cooling demand (PCD, kWh/m2 ) can serve as a heat stress vulnerability (HSV) indicator of a city's dwelling stock on the grounds that higher PCD demands will lead to higher HSV levels, (2) that potential HSV of the current stock composition can be assessed according to the predicted peak cooling loads required to restore the estimated indoor thermal conditions to acceptable thresholds. The purpose of subjecting a city's dwelling stock to the PCD-based HSV assessment is to identify segments of the dwelling stock with higher HSV levels that may require urgent actions of adaptation through renovation or replacement. The thesis presents a modelling framework and then applies it to Seoul's high-rise apartment stock using the multiple data sources available for 2014- 2050 including Seoul's climate projections under RCP4.5 and RCP8.5. The HSV assessment outcome is presented as relative rankings among the six apartment archetypes in the 18 city-district residential neighbourhoods in Seoul. The implications of the findings are discussed as inputs to what, where and how adaptation and mitigation strategies could be developed for the neighbourhoods identified, leading to a significant reduction of peak cooling demands while remaining satisfactory to dwellers' thermal wellbeing as a priority.