Evacuation elevators in an underground metro station: A Virtual Reality evacuation experiment

Research output: Contribution to journalArticlepeer-review

8 Citations (SciVal)


The number of underground metro stations around the world is steadily increasing and deep stations are becoming more common. Evacuation elevators can be a way to mitigate many of the evacuation problems of deep stations, e.g., fatigue. However, the elevators need to be used by evacuees to be effective. In order to study the influence of evacuation measures on elevator usage, experiments were performed in a Virtual Reality (VR) model of a deep underground metro station. Four scenarios with different evacuation measures were studied and compared. These scenarios consisted of a base-line scenario, a scenario with enhanced guidance systems e.g., a more informative voice alarm message and flashing green lights to guide people, a scenario with smartphone guidance and a scenario with timers showing a countdown timer above the elevators. The results show that the use of evacuation elevators is affected by the evacuation system design and that a more informative guidance system can increase evacuation elevator usage. Also, smartphone guidance was shown to increase the elevator usage, but results suggest people stop looking at the smartphone after initiating evacuation. The study also suggests that accepted waiting times for elevators can potentially be increased if countdown timers are installed above elevators.

Original languageEnglish
Article number103091
JournalFire Safety Journal
Early online date2020 May 7
Publication statusPublished - 2021 Mar 1

Subject classification (UKÄ)

  • Building Technologies


  • Accepted waiting time
  • Evacuation elevators
  • Exit choice
  • Head mounted display (HMD)
  • Human behaviour
  • Underground metro station
  • Virtual reality (VR) experiments


Dive into the research topics of 'Evacuation elevators in an underground metro station: A Virtual Reality evacuation experiment'. Together they form a unique fingerprint.

Cite this