Abstract: Aiming at the situation that the injured can't get timely rescue when they suffer from sudden disasters, a location model of emergency medical rescue centers from the perspective of low-carbon is proposed, focusing on the distance and time of transporting medical supplies from hospitals to rescue centers. First, taking Shanghai as an example, the K-means clustering algorithm based on machine learning divides Shanghai into four sub-regions, and the straight-line connection method is used to determine the initial alternative sites of each sub-region. Secondly, the entropy weighting method is used to select several final alternative sites from the initial alternative sites in each sub-region. Finally, by considering the transportation cost, carbon emission cost and late arrival penalty cost, the total cost of all hospitals in each sub-region to the rescue center is calculated, and the site with the minimum total cost is determined as the optimal location site of each sub-region, to ensure that each hospital can provide medical resources to the nearest rescue center. Through the feasibility analysis of the optimal location site, it shows that the research result can be used as a reference for subsequent medical rescue center location in Shanghai.