In the context of digital intelligence, urban transportation infrastructure is shifting towards a grid-based operational mode. The joint operation of multiple facilities has broken down information and resource barriers, but the problem of response gaps caused by resource congestion also brings new challenges to emergency risk management. This study uses the K-Means clustering algorithm to analyze the road network dependency of accident risks and identify traffic accident hotspots. Then, it analyzes resource reuse and resource congestion in the joint operation of multiple facilities, constructs a resource congestion model to classify hotspots and describe resource coverage requirements, and finally constructs an emergency resource location optimization model based on grid operation. An accident-prone elevated road network in Shanghai is selected for case validation. The results show that the proposed model can optimize emergency resource allocation, improve rescue response coverage, and reduce the loss of road network traffic quality caused by accidents.