Uncertainty analysis of passenger evacuation time for ships’ safe return to port in fires using polynomial chaos expansion with Gauss quadrature

Abstract

With the rapid development of water transportation and tourism, safe return to port is required by the International Maritime Organization (IMO) for passenger ships subject to fires. In order to ensure the life safety of passengers, a quantitative analysis method based on polynomial chaos expansion with Gauss quadrature is developed to address the uncertainty of passenger evacuation time for ships’ safe return to port in fires. To illustrate the proposed method, passenger evacuation in a hypothetical main vertical zone of ships is analyzed in this paper. The results indicate that the proposed method can be used to quantify the uncertainty of evacuation time for ships’ safe return to port, significantly reducing the number of evaluation samples. In addition, evacuation times approximately follow normal and lognormal distributions when the initial passenger density is normally distributed. Moreover, for a certain safety level, the safety factor of evacuation time in the day should be larger than that in the night, which depends on its coefficient of variation. For the calculation of evacuation time, the feasibility of the safety factor 1.25 recommended in MSC.1/Circ.1533 is examined as well. Furthermore, the effects of evacuation exit on the expansion degree of the surrogate model, distribution type and the coefficient of variation of travel time are negligible, while it has a significant influence on distribution parameters of travel time. This work will provide a methodology to facilitate the fire safety design of passenger ships.