Underground Logistics System (ULS), as a subterranean urban infrastructure with public utility attributes, can effectively meet urban emergency demands through highly resilient freight networks. However, the operational mechanisms and performance assessment methods for ULS in complex emergency logistics scenarios remain underdeveloped. This study examines ULS emergency service capacity, focusing on the impacts of the operational environment, network structure, and scheduling. A model measuring efficiency, effectiveness, and fairness is developed. Simulations based on freight demand and surface road damage, using the Xianlin case in Nanjing, compare ULS and surface truck delivery. Results show that: ULS exhibits significant advantages in emergency freight performance, particularly under conditions of surface traffic congestion and narrow emergency response time windows. Furthermore, increasing node logistics redundancy, optimizing end-point delivery modes, and ensuring local freight fairness are identified as key factors in enhancing ULS emergency service capacity. This research advances ULS planning theory and offers new insights for urban emergency management.