Risk Prediction Of Population In Xidan Commercial District

With the rapid development of urban commercial district, the city’s core business district has become the most active area of the city’s image card and the city’s economy, but behind the prosperity of the city lies a huge risk. In recent years, there have been many people in the country and abroad crowded trample accident for the city business district management has sounded the alarm bell. Crowd gathering is prone to crowded trample, the consequences of the event will have a serious impact on the social economy.Because of its special location, Xidan business district is very easy to appear in a short period of time the crowd gathered. Therefore, the management department needs to monitor the distribution of the population through monitoring, once the region within a certain threshold, it is required to take evacuation measures in time to eliminate security risks. On the one hand, because of the existence of the human resource is limited, the Xidan commercial district management is impossible to keep early presence of management personnel to ease the population; another from the angle of regional population movement, associated with each key point area in the crowd gathered with timeliness and can be predicted in advance to take corresponding measures. Therefore, how to effectively predict the key points in the business district of Xidan crowd gathering risk is a research and practical value of the subject.Based on field research and data analysis, this paper constructs the network topology model of the key nodes in Xidan business district, and gets the relationship matrix between the nodes. Analysis of historical statistics, combined with real-time statistical data, based on the Xidan business district population transfer model, to obtain the population transfer probability matrix. Identify key nodes crowd risk threshold, in building up the simulation based on analysis system, get the next time the node number of people, and to predict the Xidan commercial district in population of the key node aggregation number. By comparing with the crowd risk threshold, the risk points of the possible measures are predicted. Compared with the prediction results of different methods, it is found that this method is effective and practical.