A simulation study of tenability for passengers in a railway tunnel with arson fire

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

8 Scopus Citations
View graph of relations



Original languageEnglish
Article number103679
Journal / PublicationTunnelling and Underground Space Technology
Online published5 Nov 2020
Publication statusPublished - Feb 2021


Many arson fires occurred recently in railway systems in different countries. Arson fire occurring inside railway tunnel with a crowded train could give rise to serious consequences. The severity of tunnel fire increases with the length of tunnel, the fire size and the presence of obstructions inside the tunnel. Thus investigation of different aspects of tunnel fire is important in tunnel design, fire safety, and rescue. While real-scale experimental study of tunnel fire is in general too demanding in terms of resources and feasibility, simulation has an important role to play in such studies. The present study employs the Fire Dynamic Simulator (FDS version 6.4.0) to perform simulation of arson fire in front of a train in an underground tunnel. The arson fire occurs in an underground tunnel of length 500 m, which resembles some subway tunnels in Hong Kong. The scenario consists of a metro train stopping inside the tunnel when an arson fire is set in front of the train. In addition, the worst scenario in which the mechanical ventilation system inside the tunnel is not functioning, either due to malicious damage or power failure, is assumed. Fire sizes of 6 MW to 16 MW for durations of 600 s, which are based on the amount of fuel that is likely to be brought to the spot by arsonists, are adopted for simulation. In the simulation, temperature inside the train car and smoke layer height, temperature, visibility and concentration of carbon monoxide at important locations outside the train are investigated. Most importantly, analysis of tenability in such a tunnel fire scenario is performed. The simulation results show that inside the train and near to the train front temperature could be a threatening factor while outside the train visibility and carbon monoxide concentration are threatening factors. Thus no general rule can be formulated in regards to whether passengers should stay inside the train or escape into the tunnel under the scenario studied. Provided that the train compartment is not too crowded and that the passengers are not in great panic, it would be safer to move inside the train to locations further away from the train front than to escape from the train and move in the tunnel where the geometrical environment is not favorable. In addition, the effect of longitudinal ventilation is investigated by comparing the results of simulation for an unventilated tunnel and a tunnel under a ventilation velocity of 2 m/s. The results of the present study could be of importance in drafting plans of rescue and evacuation in tunnel fires.

Research Area(s)

  • Arson fire, Smoke, Temperature, Tenability, Tunnel fire, Unventilated tunnel