Real-time evacuation route optimization in the fire scenarios of cruise ships

Abstract

The optimization of evacuation routes for evacuees in fire scenarios on cruise ships directly affects evacuation efficiency. A time-varying equivalent weight Dijkstra algorithm (TVEWDA) is proposed, combining a fire dynamics simulator (FDS) to find optimal evacuation routes in case of a sudden cruise ship fire. First, the FDS software is used to simulate a cruise ship's single and multi-source fire scenarios. Second, the Dijkstra algorithm using only distance between nodes is improved by considering the effects of CO concentration, visibility, and temperature. The evacuees’ speed is taken into consideration within the TVEWDA. The effects of congestion are considered in the evacuees’ moving speed. Finally, the TVEWDA and a cruise ship evacuation network model are integrated to develop a real-time evacuation route optimization model. Simulation results show that the evacuation time achieved by the TVEWDA is 22% shorter than that of the Dijkstra algorithm using only distance between nodes and 11% shorter than that of an improved social force model in the multi-source fire scenario. This approach is well-suited to guide the evacuation of cruise ship evacuees in the event of single-source or multi-source fires, effectively ensuring the safety of passengers and the preservation of their property.