Urban management is one of the most prominent problems in modern and contemporary social governance. With the scale of the city, the flow of industries and population becomes larger and larger, and the city takes on more and more risks and emergencies. In order to minimize losses and prevent in advance, early drills are often the best way to prevent emergencies. This paper aims to use data mining technology to design a smart city emergency management system to achieve the role of early warning of emergency time and emergency response to emergency events. In response to this, this paper proposes the EIM-DS algorithm, an optimized data mining algorithm based on high-utility itemset mining. This paper improves the threshold setting and data classification method, which greatly develops the execution efficiency of the algorithm. In addition, an intelligent emergency management system is also designed based on the optimization algorithm. The system focuses on analyzing enterprise data, emergency data and emergency command data. The combination of the three data can evaluate the order of the entire city. The experimental results of this paper prove that the calculation time of the algorithm in this paper is about 50% lower than that of other algorithms under different datasets, which has better efficiency, and the designed system can run more smoothly through black-box and white-box experiments.