FPEeXTRA Issue 69

Evacuation Data from an Unannounced Drill at CERN

By: Ruggiero Lovreglio, Anass Rahouti, Charitha Dias, Erica Kuligowski, Giordana Gai, and Saverio La Mendola

View full PDF here

Performance-based and objective-based codes are used in several countries today for the fire safety design of the built environment, such as buildings and infrastructures. These codes are based upon the use of analytical or numerical models that require designers to assess the level of safety provided by the building in case of fire disasters. The level of safety can be assessed by comparing ASET (the Available Safe Egress Time) and RSET (the Required Safe Egress Time). To date, several evacuation modeling tools have been developed to help designers with the estimation of RSET [1]. These evacuation modeling tools require users to supply inputs to simulate occupant behaviors such as pre-evacuation time, exit choice, and evacuation movement speed [2]. Also, several evacuation databases are available to identify reasonable evacuation input values [3,4]. However, existing databases are not comprehensive enough, and new data is necessary to expand them. This work provides an overview of new evacuation data obtained for office buildings by analyzing an unannounced evacuation drill carried out at CERN in Switzerland in 2019 [5]. The data was collected through the observation of the behaviors of 203 evacuees in two office buildings. A summary of the pre-travel time distributions, speed distributions, and fundamental diagrams for the observed evacuees are reported in the following sections.

Fire Drills

The unannounced drill was carried out during normal business hours in the morning on a rainy day in June 2019 in two buildings simultaneously. Building 1 is a five-story office building. Building 2 is a four-story office building, and it is adjacent to Building 1. The drill was started by activating the building fire alarm of both buildings. In both buildings, the data was collected by positioning mini-video cameras at strategic locations, i.e., on corridors and staircases. The geometry of the two buildings and the location of the cameras are illustrated in the original article [5].

Pre-Travel Phase

Pre-travel times for both buildings are illustrated in Figure 1.a. The pre-travel time of an evacuee is defined as the time between the activation of the alarm and the time when the evacuee started to move towards a place of safety. These times were used to estimate pre-travel time distributions which are available in the original article [5]. Figure 1.b shows that the data collected in this evacuation drill represent lower pre-evacuation times compared with values collected from similar building occupancies. This trend could be explained by the relatively quick action of emergency guides and regular training and the fact that some of the literature values were observed during actual fire events instead of drills.

Travel Phase

Unimpeded horizontal walking speeds were collected in both buildings in corridors and on stairs, during which the estimated local density values fell below 1 person/m². Figure 2 illustrates the horizontal travel speeds in Buildings 1 and Building 2 as well as the stairs travel speeds. For Building 1, the mean horizontal speed of occupants was estimated as 1.17 m/s ± 0.32 m/s (N = 249), while, for Building 2, the mean speed was estimated as 1.28 m/s ± 0.31 m/s (N = 87). On the other hand, for Building 1, the mean speed on stairs was estimated as 0.88 m/s ± 0.19 m/s (N = 113), while for Building 2, the mean speed was estimated as 0.88 m/s ± 0.12 m/s (N = 17).

The collected speed and local density data were finally compared with the SFPE fundamental diagrams published in the SFPE Handbook of Fire Protection Engineering [6]. Figure 3 shows this comparison in horizontal (corridors) and vertical (stairs) configurations. These charts show a high scattering of data in both buildings and that the density did not reach values which could have had a major impact on the evacuation speeds.

Figure 1 – (a) Pre evacuation times of the CERN buildings (b) Comparison of the CERN data with the data published in the literature.

Figure 2 – Evacuation travel speed (a) in corridors and (b) on stairs. Note: all the density values were below 1 person/m².

Figure 3 - Fundamental diagrams (a) for horizonal speeds (b) for vertical speeds.

Conclusion

This work summarizes the main findings obtained from an unannounced evacuation drill in two office buildings at CERN. The results show that the pre-evacuation times ranged between 2s to 136s, and these times were substantially lower than the times observed in similar buildings. Further, the horizontal speeds ranged between 0.5 m/s and 2.9 m/s, while the vertical speed ranged between 0.5 m/s and 1.4 m/s. In both buildings, no congestion was observed, and in turn, the density did not affect the evacuation speeds. Horizontal walking speed data collected in corridors was within the range of the values reported in previous studies. Further, walking speeds on descending stairways measured in the present study were significantly higher than those available in the literature (see Section 5 in [5]).  

Note: This paper is a short version of a manuscript that has been published in the Fire Safety Journal. Further information about the evacuation drill and collected data can be found in the full article associated with this work [5].

Ruggiero Lovreglio is is with the School of Built Environment, Massey University

Anass Rahouti, is with Fire Safety Consulting sprl

Charitha Dias is with the Qatar Transportation and Traffic Safety Center, Department of Civil and Architectural Engineering, Qatar University

Erica Kuligowski is with the School of Engineering, RMIT University

Giordana Gai is with the Occupational Health & Safety and Environmental Protection (HSE) Unit

Saverio La Mendola is with the Occupational Health & Safety and Environmental Protection (HSE) Unit

References

[1]      R. Lovreglio, E. Ronchi, M.J. Kinsey, An Online Survey of Pedestrian Evacuation Model Usage and Users, Fire Technol. 56 (2019) 1133–1153. https://doi.org/10.1007/s10694-019-00923-8.

[2]      E.D. Kuligowski, Computer Evacuation Models for Buildings, in: SFPE Handbook of Fire Protection Engineering. 2016: pp. 2152–2180.

[3]      S.M. V. Gwynne, K.E. Boyce, Engineering Data, in: SFPE Handbook of Fire Protection Engineering. Springer New York, New York, NY, 2016: pp. 2429–2551. https://doi.org/10.1007/978-1-4939-2565-0_64.

[4]      R. Lovreglio, E. Kuligowski, S. Gwynne, K. Boyce, A pre-evacuation database for use in egress simulations, Fire Saf. J. 105 (2019) 107–128. https://doi.org/10.1016/J.FIRESAF.2018.12.009.

[5]      A. Rahouti, R. Lovreglio, C. Dias, E. Kuligowski, G. Gai, S. La Mendola, Investigating office buildings evacuations using unannounced fire drills: the case study of CERN, Switzerland, Fire Saf. J. (2021) 103403. https://doi.org/10.1016/j.firesaf.2021.103403.

[6]      S.M. V. Gwynne, E.R. Rosenbaum, Employing the Hydraulic Model in Assessing Emergency Movement, in: SFPE Handbook of Fire Protection Engineering, Springer, 2016: pp. 2115–2151.