This thesis develops SunlightCity, a digital twin-enabled urban solar exposure analytics framework for pedestrian route planning and urban comfort assessment in Manhattan. The study integrates a Unity-based 3D city model, annual solar trajectory simulation, time-indexed shadow generation, and a structured micro-environmental observation representation to quantify dynamic sunlight exposure at the road-network level. On this basis, a time-dependent multi-objective routing model is formulated to evaluate trade-offs between travel effort and cumulative solar exposure under both shade-seeking and sun-seeking preferences. A label-correcting Pareto search algorithm is implemented to generate interpretable route alternatives, and the full workflow is integrated into an operational prototype with web-based and digital twin interfaces. The Manhattan case study shows that pedestrian solar exposure varies substantially across space and time, and that environmentally informed routing can reveal meaningful trade-offs and planning-relevant corridor patterns for healthier and more climate-responsive cities.