Emergency Escape Route Design and Application

Abstract

This paper uses k-means clustering algorithm, single objective optimization model and multisource multi-sink problem shortest path algorithm to obtain the optimal path algorithm for emergency evacuation and the shortest time model.Firstly, a code of conduct for the division of terrorist attacks has been established. The number of casualties in various incidents and the scope of destructive power to public areas are counted. A three-dimensional cluster core is selected. All terrorist at- tacks are classified into three levels: level 1-3 by the k-means clustering algorithm, and the degree of harm is gradually increasing.Secondly, the optimal path from each venue to the stairway is calculated. The re- lationship between the speed of flow and the number of evacuees is given in the case of three situations. The multi-source multi-sink problem is transformed into a single- source shortest path problem by adding a virtual node to all venues on the second floor of the Louvre, and the Dijkstra algorithm is called once to obtain the optimal path from the venue to all stairs.The Final step is to establish an emergency evacuation optimization model with the minimum evacuation time as the objective function. The overall evacuation time is divided into the personnel response time, the horizontal road walking time, the time down the stairs, and the waiting time when congestion occurs.  the open number of emergency exits can be judging by different levels of terrorist events, and the optimal evacuation path and the shortest time for different emergency events are given. At the same time, we also analyzed the changing trend of velocity in different cases, which can be adjusted according to the actual conditions.