Path planning for guided passengers during evacuation in subway station based on multi-objective optimization

Abstract

The reasonable setting of guides can directly improve the evacuation capacity in the subway station. The important role of guide assignment scheme and evacuation path planning of guided passengers is particularly prominent. Considering the distribution of passengers in the station, one contribution is that an assignment scheme of the initial number and location of guides is firstly established based on Gaussian mixture model and cost function method, so as to increase the guidance range and reduce the waste of human resources. Another contribution is that a multi-objective optimization model is developed for guide path planning and NSGA-II method is used to find the optimal solution, which could help avoiding the evacuation bottlenecks in advance. Massmotion embedded with social force model and minimum cost model is adopted to simulate the movement of people, which is verified by the independent sample t-test in this paper. Evacuation experiments under the random distribution of passengers are carried out to demonstrate the significance of the proposed guidance method. The results indicate that the total evacuation efficiency can be improved by about 13% in the scenario with guides, and the passengers’ choice of stairs/escalators is more balanced. Moreover, passengers induced by different guides can avoid reaching the same evacuation node at the same time window, which alleviates the pressure of nodes. The results could provide theoretical support for passenger safety management and even improve the service level of stations.