This website uses cookies to store information on your computer. Some of these cookies are used for visitor analysis, others are essential to making our site function properly and improve the user experience. By using this site, you consent to the placement of these cookies. Click Accept to consent and dismiss this message or Deny to leave this website. Read our Privacy Statement for more.
An Overview of Research of Occupant Movement in Building Stairwells
Share |

Issue 85: An Overview of Research of Occupant Movement in Building Stairwells

By Chris Campbell

Several research projects have been conducted over the past several decades focusing on the descent of occupants in high-rise stairwells during emergency situations.  A model included in the SFPE Handbook of Fire Protection Engineering1 resulting from that research is in the form of a hydraulic model of people movement and provides an estimate of local occupant speeds as a function of density.  Authors of that Handbook chapter state that the model "tends to an optimistic estimate of evacuation time” due to the simplicity of the model".  


In the hydraulic model, occupants are treated as fluid particles that are not distinguishable from one particle to the next.  Therefore, variable behaviors of different occupants that are seen in live evacuations are not considered.  Key correlations of the hydraulic model, occupant speed and flow rate during egress, are displayed graphically in Figure 1 and Figure 2 .


Figure 1 : Speed as a function of density.1

Figure 2 : Specific flow rate as a function of density1

A three-year research project, funded by the National Institute of Standards and Technology (NIST), was conducted at the University of Maryland involving the analysis of data collected in videos within the stairwells of five U.S. high-rise office buildings during unannounced fire drills. The buildings were located throughout the U.S. geographically, and all were business occupancies.  The intent of the research was to explore the behavior and movement characteristics of occupants as they descend stairwells.


The data collected during the evacuations was captured through video recordings made on alternating stairwell landings. The recordings showed occupants entering the landing from both the stairs above and the present floor, and showed occupants exiting the landing to the stairs below. From the captured videos, analyses involving occupant velocity and flow rate, group movement behavior and occupant merging behavior were performed on approximately 100 observed merging events.


Occupant Flow

Work conducted in the first year of the project focused on the relationship of occupant flow rate and stair width. An initial analysis of the data was conducted using a calculation of density as the number of people between camera views in a given stair divided by the area between those camera views. As shown in Figure 3, the majority of the observed flow rates fell below the flow rates predicted using the correlations in the SFPE Handbook.1 In fact, the SFPE Handbook correlation appears to act as an upper limit on the majority of the flow rate data. Data from four of the stairwells observed is shown in Figure 3, and for each set of data, a quadratic line of best fit is shown.


Figure 3 : Specific Flow as a function of density2. The top curve shows the correlation
from the SFPE Handbook, while the lower four curves show lines of best fit for the four
sets of data in the figure.

Occupant Speed

Work conducted in the second year of the project expanded upon the initial findings, particularly re-examining how to determine the density of occupants on a stair. An analysis of the data was conducted utilizing the average number of people on a single stair landing divided by the area of the stair landing itself. As shown in Figure 4, the majority of occupant speeds on the stair landing fell below the speeds predicted using the correlations in the SFPE Handbook.1 Similar to the flow rate, a best fine line is shown to fall completely beneath the SFPE correlation.


Figure 4 : Speed as a function of density3

Merging of Groups of Occupants

The final year of the project focused on occupant movement in groups, as well as the merging of groups of occupants. An analysis of the merging event that takes place between occupants entering a stair landing from the stairs above and occupants entering the stair from the given floor was conducted.


It was found that the flow ratio, which is the outflow rate from a merging event divided by the combined inflow rate of the two groups entering the merging event, was an average of 75%.  The hydraulic model states that the combined inflow rate should equal the outflow rate of a merging event, which would be equivalent to a flow ratio of 100%. As shown in Figure 5, the flow ratio was found to have a direct relationship to the outflow rate of the merging events.


Figure 5 : Flow Ratio as a function of out flow rate4

It is not surprising that the flow ratio, which is essentially the outflow divided by the inflow, is shown to have a direct relationship to the outflow itself. However, the trend line fitted to the data shows that as the flow ratios vary from approximately 60% to 100%, the outflow rate changes by only 0.5 person per second. This suggests that even a small change in the outflow rate from a merging event can have a significant effect on the flow ratio.



The importance of this research is seen in a range of applications, from strengthening computer-based egress models to improving applicable codes and standards, such as NFPA 101.5 Egress models are an attempt to accurately replicate human behavior during egress from a building, thus a better understanding of how people behave in such situations can result in a stronger model of that behavior. Several of the theses resulting from this research compared the results of the empirically collected data to the results of various computer-based egress models. While the accuracy of these models varied greatly depending on the model and the situation, it was clear that egress models can continue to be improved.


Perhaps more important is the potential to improve codes and standards, which are used much more frequently than egress models. The research summarized in the SFPE Handbook is used as a foundation for many life safety related code requirements enforced today. For example, as cited in the annex of NFPA 101,5 "the effective capacity of stairways has been shown by research to be proportional to the effective width of the stairway…This phenomena, and the supporting research, were described in the chapter, "Movement of People,” in the SFPE Handbook of Fire Protection Engineering…” With more recent and detailed data that considers the variable behaviors of different occupants, codes and standards should be updated and improved to more accurately address the nature of occupant movement.


Chris Campbell is with the Protection Engineering Group.

  1. Gwynne, S. and Rosenbaum, E. "Employing the Hydraulic Model in Assessing Emergency Movement,” in The SFPE Handbook of Fire Protection Engineerng, Fourth Edition, Quincy, Massachussetts: National Fire Protection Association, 2008.
  2. A. Blair: The Effect Of Stair Width On Occupant Speed And Flow Of High Rise Buildings, University of Maryland, 2010.
  3. A.  Leahy: Observed Trends In Human Behavior Phenomena Within High- Rise Stairwells, University of Maryland, 2011.
  4. C. Campbell: Occupant Merging Behavior During Egress From High Rise Buildings, University of Maryland, 2012.
  5. NFPA 101, Life Safety Code, National Fire Protection Association, Quincy, MA, 2012.

Related Articles:

2nd Quarter 2013 – Smoke Control in Very Tall Buildings – Past, Present and Future -- Erik Anderson, P.E., Koffel Associates, Inc.
An overview of engineered smoke control systems in very tall buildings, which use mechanical means to produce pressure differentials across barriers to inhibit smoke spread. The author discusses code trends of the past and some of the relevant design considerations for smoke control in very tall buildings of the future. READ MORE


2nd Quarter 2013 – Elevator Pressurization -- John H. Klote, Ph.D., P.E., FSFPE (John H. Klote, Inc.), Michael J. Ferreira, P.E. (Hughes Associates, Inc.), and James A. Milke, Ph.D., P.E., FSFPE (University of Maryland)
Elevator pressurization systems are intended to prevent smoke from flowing through an elevator shaft and threatening life on floors remote from a fire. Elevator pressurization is an alternative to enclosed elevator lobbies. Many pressurized elevators are in buildings that have pressurized stairwells, and the focus of this article is on both of these pressurization systems operating together. READ MORE


2nd Quarter 2013 – Evacuation of Tall Buildings -- Bryan Hoskins, Ph.D., Oklahoma State University
This article looks at components of the evacuation time of occupants in tall buildings and the assumptions that are made by egress system designers. The focus is on the movement to and within the stairs as well as the data used to develop an estimate of the descent rate. Various egress options are discussed. READ MORE

For questions concerning delivery of this newsletter, please contact our Customer Service Department at (216) 931-9934 or
Copyright 2014, SFPE and Penton. All rights reserved. This newsletter is protected by United States copyright and other intellectual property laws and may not be reproduced, rewritten, distributed, re-disseminated, transmitted, displayed, published or broadcast, directly or indirectly, in any medium without the prior written permission of SFPE and Penton.

Corporate 100 Visionaries

About Us

© 2019 SFPE | All Rights Reserved
Privacy Policy