Pre-movement time

The pre-movement time is the time between the moment occupants receive a first relevant warning or cue of a fire and the moment they begin purposeful movement towards a place of safety. It is part of the required safe egress time, which can be divided roughly into the pre-movement time and the time needed to move. During the pre-movement time, occupants recognise that something is wrong, decide what to do and carry out other actions before they set off, such as investigating, gathering belongings or finding family members.

The pre-movement time is often the largest part of the evacuation time. In unannounced drills in Canadian apartment buildings with an audible alarm, occupants took on average 2 min 49 s to start, and three-quarters of the evacuation time was spent before moving. It is also the part that is most strongly governed by human behaviour in fire, one of the least predictable aspects of fire safety. Design guides therefore give default values that depend on the type of occupants, the alarm system and the management of the building. Such values range from 30 s for alert office workers in the room of fire origin to 600 s for sleeping hotel guests remote from the fire.

Definition and terminology

Several names are used for the same period. The Introduction to Fire Safety Engineering speaks of the pre-evacuation time. C/VM2 (second edition, 2025) uses "pre-travel activity time", the period after an alarm or fire cue is transmitted and before occupants first travel towards an exit. The Australian data sheets for the Fire Safety Verification Method use both "pre-travel time" and "pre-movement time". NIST, in its study of the World Trade Center evacuation, measured the "evacuation initiation delay".

In the SFPE Handbook, Proulx prefers the term "delay time to start". She notes that "premovement" is used particularly in the British literature. Her objection is that occupants often move around a great deal during this period, trying to find out what is happening. Nelson and Mowrer split the same period into two terms of their sum for the required safe egress time. The first runs from notification until occupants decide to act, the second from that decision until evacuation starts.

The start of the period also varies between sources. In C/VM2 the pre-travel activity time begins at notification, after the detection time and a notification time of 30 s for standard evacuation strategies. The International Fire Engineering Guidelines (2005) count from the occurrence of a cue, which can also be a fire cue or a warning by other people. When values from different sources are combined, it must therefore be clear where each period begins.

Components

Most sources divide the pre-movement time into a recognition part and a response part. The Australian data sheets define the recognition time as the period after an alarm or cue is evident but before an occupant begins to respond. During this time, people continue their normal activities, such as working, shopping or sleeping. The recognition time ends when they accept that they need to respond. The response time follows. Occupants stop their normal activities and do other things before they move towards an exit or a safer place.

The pre-movement time runs from the alarm or cue to the start of movement and consists of a recognition time and a response time. Based on Australian Building Codes Board (2020a), Data Sheet D1, pp. 244–245, and National Research Council of Canada et al. (2005), pp. 1.8-2 and 1.8-4.

The International Fire Engineering Guidelines use the same split under other names. The pre-movement phase is the sum of a response period, from the occurrence of a cue to its recognition, and a delay period, from recognition to the start of movement. In the delay period, occupants may carry out a wide variety of actions, including no action at all. Proulx describes three subcomponents: perception of the alarm or cue, interpretation of the information, and action. The Introduction to Fire Safety Engineering uses three phases as well: awareness, decision-making and execution.

The actions in the response part are varied. They include investigating the cue, shutting down equipment, finding friends and family, fighting the fire, alerting others, deciding whether to leave, and waiting for instructions. In residential drills, occupants got dressed and gathered children, pets, purses, wallets and keys before leaving. Office workers finished phone calls, saved data, secured files and collected belongings, and many had to be prompted by their fire warden. In the World Trade Center towers in 2001, such pre-evacuation actions were the strongest predictor of how long people delayed their evacuation.

Interpreting ambiguous cues often takes the form of seeking information. NIST describes this as a typical human response in sudden emergencies, because people need to make sense of an event before they act on it. Its causal model for the towers linked environmental cues, the floor of the occupant, perceived risk and information seeking to the delay. Such step-by-step descriptions of how people perceive cues, judge the risk and decide what to do are treated further under protective action decision model.

Factors that influence pre-movement time

The Introduction to Fire Safety Engineering lists many factors that shape the response to a fire. They include the person's assumed role, earlier experience, cultural background, training, personality and physical characteristics. Also important are the perceived threat, the available escape routes, the actions of others and the signals that are received. Proulx groups the factors into building characteristics and occupant characteristics. Building characteristics include the warning system, the layout and wayfinding, visual access to other people, the presence of a focal point such as a stage, the level of training and the frequency of false alarms. In her experience, three or more false alarms in 6 months already count as a large number.

The type of warning has a strong effect. An evacuation alarm system with a plain alarm signal is, according to Proulx, the cue that occupants take longest to respond to when they are untrained or used to false alarms. A voice alarm system with informative live messages is probably the best way to initiate a fast response, and prerecorded messages are rarely as effective. Bryan adds that a directive message loses credibility when it conflicts with what occupants see or smell.

The Introduction to Fire Safety Engineering likewise states that a slow-whoop tone is less effective than the spoken word, and that live messages are followed faster than automatic alarms. In an underground station, passengers never started to evacuate on the fire alarm alone. When staff or precise live voice messages prompted them, they began to move about 15 s after the message.

Occupant characteristics include familiarity with the building, a sense of responsibility, social affiliation and commitment to an activity. Families tend to gather before they leave, and people who are eating, gambling or queuing for a flight are slow to turn their attention to an alarm. Sleeping occupants, and people with perceptual, physical or intellectual limitations or under the influence of drugs or alcohol, need longer. The Australian data sheets note that even blood alcohol levels below 0.05 can leave a significant share of residents unwoken by smoke alarms.

Other people also matter. In experiments by Latané and Darley, students alone in a room reported smoke entering the room in 75 percent of cases. When two passive confederates sat with them, only 10 percent of the groups reported it. The Introduction to Fire Safety Engineering describes the same tendency to conform to the group and to follow authority. People may wait to leave until a manager has given permission. Group behaviour can also help: when enough people make the right decision, as in a well-practised evacuation organisation, the others tend to follow.

Panic is not a major cause of delay. Proulx states that panic is very rare even in fires, and Bryan describes deliberate, purposeful behaviour as the most frequent response. The Introduction to Fire Safety Engineering notes that a lack of awareness of the danger is usually the larger problem, because people find it hard to judge how fast a fire grows.

Measured values

Measured pre-movement times vary widely. In seven residential buildings in Canada, Proulx recorded the start times of more than 500 occupants during drills. In the three buildings with a clearly audible alarm, the mean delay was 2 min 49 s. In a winter drill, with coats and boots to put on, it was 5 min 19 s. In two buildings where more than 20 percent of residents found the alarm too quiet, some started only after 25 min, and the mean was 9 min.

Office workers were much quicker. Video records of more than 1,000 occupants in three Canadian government office buildings gave a mean delay of 50 s. In an unannounced drill in a large retail store, customers started to move on average 25 s after the alarm, with a maximum of 55 s, mainly because staff responded fast. Real fires can be slower. Interviews after a high-rise office fire without a working alarm gave a mean of about 2 min 30 s. After a night-time residential fire, occupants took about 10 min to start.

In the World Trade Center on 11 September 2001, the median time to initiate evacuation in the first tower was 3 minutes below floor 77. Near the impact region, on floors 77 to 91, it was 5 minutes. The floor, environmental cues and seeking additional information best predicted the delay in that tower.

Mean or median times to start evacuation from drills and fires: from under a minute in offices and shops to about 10 minutes at night. Based on Proulx (2002), pp. 3-348 to 3-349, and Averill et al. (2005), p. xxxiii.

Design values

Because there is no validated method to calculate pre-movement times, design guides provide default values. The International Fire Engineering Guidelines (2005) recommend estimates based on published studies, and mention a probabilistic method that fits a Weibull distribution to observed times. The commentary to C/VM2 notes that there are no widely accepted values. It names the British document PD 7974-6 as the most complete collection.

In New Zealand, C/VM2 (second edition, 2025) prescribes pre-travel activity times by occupant type, alarm type and location. Occupants in the enclosure of fire origin get shorter times, because direct cues such as flames and smoke, or indirect cues from others, prompt a faster response. Indirect cues may not be assumed in apartments, hotels or separate office tenancies. The table lists a selection of the values.

Pre-travel activity times in C/VM2 (second edition, 2025) for occupants in and remote from the enclosure of fire origin. Based on Ministry of Business, Innovation and Employment (2025b), p. 30.

Occupants (example)	Alarm signal	In enclosure of origin (s)	Remote from enclosure of origin (s)
Awake, alert and familiar (offices)	Any	30	60
Awake, alert and unfamiliar (retail shops, restaurants)	Standard	60	120
Awake, alert and unfamiliar (retail shops, restaurants)	Voice	30	60
Sleeping and familiar (apartments)	Standard	60	300
Sleeping and unfamiliar (hotels and motels)	Standard	60	600
Sleeping and unfamiliar (hotels and motels)	Voice	60	300
Awake, under care of trained staff (dental office, clinic)	Any	60	120

The commentary explains that these values were chosen from the most optimistic management options. New Zealand rules require evacuation schemes in most public non-residential buildings, which leads to a culture of prompt evacuation. C/VM2 also notes that the incipient phase of a fire is left out of its design fire, which gives an inherent safety factor on these times.

The Australian data sheets give distributions rather than single values. For staff in a simple workplace with a well-trained emergency organisation and an automatic alarm system, 10 percent start after 30 s, 80 percent after 60 s and 10 percent after 90 s. With a minimal emergency organisation, much longer times are expected. A check of the design against failure of that organisation may assume that 10 percent do not respond and that the others take 15 to 30 minutes. The data sheets warn that a split in which 80 percent of occupants start at the same moment may be too crude.

Suggested default distributions of pre-movement time for staff in simple workplaces with two levels of emergency management. Based on Australian Building Codes Board (2020a), Data Sheet D1, p. 270.

Proulx reproduces values from the British draft DD240 that depend on the warning system. For offices and schools, the estimated delay is under 1 min with live directives, 3 min with prerecorded messages and trained staff, and more than 4 min with a plain alarm and untrained staff. For hotels the three values are under 2, 4 and more than 6 min, and for hospitals under 3, 5 and more than 8 min. For one and the same office building, the choice of warning system and staff training thus shifts the estimate by several minutes.

Use in fire safety design

In an analysis that compares the required with the available safe egress time, the pre-movement time is added after the detection and notification times. Older calculations ignored it. Proulx notes that this neglect usually explains why calculated evacuation times were shorter than those observed in drills. Earlier editions of the SFPE Handbook chapter suggested doubling the calculated movement time. Many engineers found this too demanding and added only 15 to 30 s instead.

Because occupants do not all start at once, the Australian data sheets recommend representing the population by a distribution or by several design occupant profiles. NIST simulated a phased evacuation of three floors in a World Trade Center tower. Random start delays of 0 to 10 minutes raised the time to clear the floors from about 4 to about 11 minutes. The pre-movement time can also be shortened by design and management. The International Fire Engineering Guidelines list additional and less ambiguous cues, more trained personnel, training programmes and better information as measures.

Limits and common misconceptions

A common assumption is that people start to evacuate as soon as they hear an alarm. Field studies show instead that people tend to ignore or play down initial cues, investigate and only then decide. Another assumption is that real fires produce faster responses than drills. Interviews with fire victims show the opposite: the first signs of a fire are usually ambiguous, which tends to lengthen the delay.

Default values must be used with care. The Australian data sheets state that their suggested times must be checked against literature, drills and incident reports, and adjusted to the building. The International Fire Engineering Guidelines describe only "non-validated quasi-methodologies" for this phase. Any value based on engineering judgement must be justified in detail. For factories, time to shut down equipment may have to be added to the published values.

See also

• Unwanted fire alarms
• Evacuation
• Affiliative behaviour

References

• Australian Building Codes Board. (2020a). Data sheets: Fire Safety Verification Method (Handbook annex, Print version 1.1).
• Averill, J. D., Mileti, D. S., Peacock, R. D., Kuligowski, E. D., Groner, N., Proulx, G., Reneke, P. A., & Nelson, H. E. (2005). Occupant behavior, egress, and emergency communications (NIST NCSTAR 1-7). National Institute of Standards and Technology.
• Bryan, J. L. (2002). Behavioral response to fire and smoke. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-315–3-341). National Fire Protection Association; Society of Fire Protection Engineers.
• Ministry of Business, Innovation and Employment. (2013). Commentary for Building Code clauses C1–C6 and Verification Method C/VM2.
• Ministry of Business, Innovation and Employment. (2025b). Verification Method C/VM2: Framework for fire safety design (2nd ed.).
• National Research Council of Canada, International Code Council, Department of Building and Housing, & Australian Building Codes Board. (2005). International fire engineering guidelines (2005 ed.). Australian Building Codes Board.
• Nelson, H. E., & Mowrer, F. W. (2002). Emergency movement. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-367–3-380). National Fire Protection Association; Society of Fire Protection Engineers.
• Proulx, G. (2002). Movement of people: The evacuation timing. In P. J. DiNenno, D. Drysdale, C. L. Beyler, W. D. Walton, R. L. P. Custer, J. R. Hall Jr., & J. M. Watts Jr. (Eds.), SFPE handbook of fire protection engineering (3rd ed., pp. 3-342–3-366). National Fire Protection Association; Society of Fire Protection Engineers.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.