Workshop on Building Occupant Movement During Fire Emergencies, June 10-11, 2004, National Institute of Standards and Technology

Richard D. Peacock (richard.peacock@nist.gov) and Erica D. Kuligowski (erica.kuligowski@nist.gov), Editors

Both before and since the World Trade Center tower collapses, there have been far too frequent events in which there was extensive life loss because the time needed for safe evacuation from a threatened building was not available – it was less than the time available for escape. There is a broad range of emergency scenarios for which there is an alarming gap between the public expectation of safety and the ability to provide it. These include man-made threats, natural disasters, and the more common system failures (e.g., gas leaks and power outages). The urgency of response to knowing something is very wrong within a building is now being accentuated and perhaps even changed, as the old paradigms of "orderly movement will get you out in time" and "find a safe part of the building and wait for rescue" are open to question. Thus, the need for accurate, quantitative assessment of people movement in emergencies has never been greater than it is today.

To this end, the Building and Fire Research Laboratory at the National Institute of Standards and Technology, in cooperation with the United Technologies Research Center, hosted a two-day workshop focusing on needed research on occupant behavior and movement during building emergencies. This workshop was motivated by a renewed interest in how buildings should be evacuated during fire emergencies and by the desire to provide a forum for the exchange of experiences among the fire and non-fire communities working on emergency egress. Organized into several sessions with specific topics areas, several presentations were included in each session, with an extended period for discussion at the end of each session.  For each presentation, visuals used for the presentation are included, along with any additional information provided by the author on the topic.  For each workshop session, the session moderator prepared a summary of key points of research interest from the presentations and discussion.

A document of the proceedings is available for download here. All the documents, including the full proceedings are in PDF format, requiring Adobe Acrobat Reader to access the files. To install Adobe Acrobat Reader, click here.

Workshop Sessions

Codes and Standards Requirements for Building Evacuation, Session Chair: Richard Bukowski, NIST. Several recent events have motivated discussions on how to best protect and safely evacuate building occupants during fire emergencies. As a result, modifications to current building and fire codes, such as stairwell capacity or the use of elevators, are being considered. This session aims to provide insight on egress code characteristics in countries around the world as well as efforts to change codes involving the life safety of buildings by industry, code developers, and first responders. It is also of interest to discuss how to guarantee certain levels of traffic performance during evacuations. To view details of the talks in this session, click here.

Building Egress Strategies, Chairs: Dennis Mileti, University of Colorado and Mauro Atalla, United Technologies Research Center. Investigating and revising the evacuation strategies of buildings have become a primary focus for safety officials in the U.S. and around the world. Large-scale evacuations during natural disasters such as floods and earthquakes have been the subject of extensive research resulting in an understanding of information exchange and evacuee behavior consistent with emergency planning needs for such events. The purpose of this session is to discuss the use of predictive models in developing new evacuation strategies and current procedures for training and information exchange, especially in tall buildings. Although NIST is interested in “egress strategies,” the body of knowledge we would like to access is how human beings behave in response to emergency information and/or life threatening events such that they take protective actions (and only ONE of the appropriate protective actions during a fire may be egress). Also, it is a goal to provide a forum for researchers working in relevant fields (social science, fire protection, human factors, traffic planning, numerical analysis, controls, etc.) to discuss similarities between evacuation from buildings and other systems, for instance the evacuation from cities. Desirable outcomes of this exchange of information are: (a) a discussion of the issues related to egress at a basic level to propose possible near-term solutions for fire officials, and (b) identification of knowledge gaps that need to be addressed in the mid-term. To view details of the talks in this session, click here.

Data Needs for Predictive Building Movement Models, Chairs: Erica Kuligowski, NIST and Guylène Proulx, National Research Council. The usefulness of a model-based predictive capability is heavily dependent on an understanding of its accuracy and limits of applicability. However, the current situation is such that small amounts of data exist and that they are not generally available to groups studying this topic. Additional experimental data gathering efforts are needed in order to quantify current and future egress predictive capabilities. Areas of interest may include pre-evacuation timing and behavior, disabled occupant responses to fires, exhaustion on stairs and interaction of occupants and building systems with the environment of the event. Several efforts have been pursued in recent years in the areas of evacuation and egress modeling in the context of fire protection engineering. Outside of the fire-protection engineering field, models are being created to simulate evacuation from cities and/or human behavior in response to terrorist events. This session aims to discuss new methods in modeling, especially by providing an interdisciplinary information exchange on the prediction of human behavior and threat evolution in emergencies. Also of interest are new models, results and experimental data gathering related to the use of models to investigate the building system behavior during emergencies. To view details of the talks in this session, click here.

Workshop Summary

Participants at the workshop representing varied disciplines – psychology, human factors, sociology, engineering, computer science, government agencies, and toxicology – discussed common efforts towards more accurate prediction methods and information on human behavior in fires and other emergencies. There is a great deal of work already completed in many disciplines (information flow, building technology sensors, elevator use, community evacuation planning, group dynamics, etc.) that can be used to provide better prediction tools.

Overall

Participants of the workshop were introduced to research in many different disciplines with common links between their research and what is going on in the fire field. The collaboration of the disciplines can provide guidance on the several aspects of evacuation:

Several research needs were expressed during the workshop. These include:

Specific needs obtained from workshop participants

There is a need to understand how the people, the building, and the environment react together. This involves an integrated “model” and more of a systems view of the evacuation. Suggestions were made to use technology in buildings (sensors) to help people during their evacuation. For instance, giving them specific information on which route to take or providing them with a sensor at each door to let them know if there is fire or smoke behind the door.

There is a need to better understand the behavioral aspect of evacuation for better prediction methods and more effective training techniques. However, it is not clear how specific this understanding should be to accurately provide safety for building occupants. For instance, do we need to outline each decision made by every occupant during the pre-evacuation stage or is it enough to simply assign a distribution of pre-evacuation time delays to represent time spent before beginning movement toward an exit? In either case, data are needed.

There is a need to collaborate with other disciplines on providing more effective emergency planning. On one hand, we need to know what to expect from occupants and base the emergency plan on that. For instance, people tend to leave the way they come into a building. Because of this, we could possibly widen main doors and/or plan for elevator use in certain emergencies. On the other hand, behavior is pliable and we need to impact occupant behavior in our building design and the information given to occupants. Emergency plans should involve input from the actual occupants and involve extensive practice (even including motivational rewards).

There is a need to include the impacts of human behavior in predictive models. Currently used evacuation models lack certain behavioral aspects of an evacuation, including group behavior and accurate representation of the disabled population. Projects are in the works to help identify gaps in the evacuation models and eventually update current models with needed data. NIST is working to provide a central repository for such data on human behavior and movement during evacuation to make data widely available to researchers. The key is to ensure that available data are sufficiently documented to make it useful to researchers who were not involved in the original data collection or those in disciplines different from the original researchers.

There is a need for building codes and regulations to better reflect the impact of human behavior during emergencies. Much work is needed in the code area to make sufficient changes to current codes in response to recent events. Workshop participants expressed a desire that codes and standards be based on appropriate scientific study rather than reaction to specific events. This would include study not only of changes to specific code requirements but also the overall scope and the balance of cost with benefits provided by major revisions to existing codes and standards.


Graphic Rule

Privacy Statement/Security Notice  |  Disclaimer  |  FOIA

NIST is an agency of the U.S. Commerce Department's Technology Administration.

 BFRL Logo