Considerations & challenges for refuge areas

Considerations & challenges for refuge areas

Concentrating all or many occupants of many floors of a tall building onto refuge floors during an emergency has become a concept that has gained increased use. However, the refuge floor concept can pose challenges to efficiency, sustainability and life safety goals. In 2011 the Council for Tall Buildings and Urban Habitats (CTBUH) published a research paper with the intention of fostering a better understanding of the concepts – this paper was revised in 2012. FME is putting Considerations & Challenges for Refuge Areas in Tall Buildings, in the spotlight in this issue as we believe this paper remains relevant today and recognise it as a perennial publication of note.

Tall buildings offer the advantages of housing people and business operations efficiently and vertically in urban areas. One of the great efficiencies is that relatively few independent exit stairs are needed to serve the population of a tall building, housing potentially thousands of occupants.

The provision of automatic sprinkler systems and compartmentation features afforded by fire resistive floor systems are generally accepted measures, which safely allow for evacuation of only a few selected floors during a fire or similar emergency to other floors remote and protected from the fire or similar emergency. 

Although elevators may play a bigger role in total building evacuations (Proulx et al 2009) in the future, there
are a variety of reasons and concerns that stairs (Peacock et al 2009) and elevators alone (Heyes 2009) are not adequate to support occupant needs during a partial or total building evacuation process. 

Major points that have been recognised in the building community are as follows:

  • Tall buildings with large occupant loads are subject to long evacuation times (1-3 hours)
  • Evacuation down long stair routes can be physically demanding
  • or tiring for occupants with low stamina

Stairs are not conducive for movement of people with permanent or temporary disabilities (Proulx and Pineau 1996) that include:

  • – mobility impaired occupants – wheelchairs, walkers, canes
  • – health impaired occupants – those with respiratory or cardiac issues
  • – temporary conditions – pregnancy, broken limbs or injury

Total evacuation may not always be desirable or feasible. A partial evacuation may be appropriate to move only those occupants in or near the affected fire or emergency zone to another area of relative safety. (Lay 2008, O’Connor and Cohn 2008)

Stairs (even pressurised stairs) can become contaminated by smoke when multiple doors are open or doors fail to close. (Bukowski 2009)

Fire department personnel will often rely on exit stairs for staging and hose deployment operations. Such operations can conflict with occupants egress movement and result in smoke entry into the stairway while doors are held open by hose lays.

One or more of these points have been cited as the basis for justification for refuge spaces or refuge floors in tall buildings. The global experience of refuge spaces in tall buildings varies, but includes at least two basic approaches: dispersed protection and consolidated protection. The dispersed approach views all floors as potential refuge areas, while the consolidated approach considers that only several selected floors will serve the refuge function.

The dispersed approach

This most common in Europe, the Middle East and North America where the dispersed approach is reflected in the Federal Law. The USA’s Disabilities Act (ADA), initially implemented in 1991, defines building design standards to implement that law and established two key premises regarding the design of exits in tall buildings:

Where a multi-floor building used stairs for exits, it was recognised that some disabled individuals could not avail themselves of such exits and, therefore, ADAAG required protected “Areas of Rescue Assistance” or the provision of a “horizontal exit” which enables disabled occupants to readily seek refuge. Horizontal exits are constructed using continuous fire barriers (2-hour or greater) and self/automatic-closing fire doors to divide a floor area into independent fire/smoke-protected zones.

Another alternative to providing “Areas of Rescue Assistance” or a “horizontal exit”, and the most commonly practiced alternative, is for the building to have a supervised automatic sprinkler system. 

Effectively, the ADAAG recognised any one of four methods to provide for the safety of occupants during a fire event in the building. The most common and effective solution for tall buildings is the supervised sprinkler system approach. The “areas-of-rescue-assistance” option is generally not an option desired by architects or developers for reasons of cost and loss of useable and/or leasable area.

The consolidated approach 

Refuge floors provide both the option for occupants to pause during the evacuation process, and an assumed safe holding area for occupants. Occupants who are exiting can pause and rest at the refuge floor until they feel they are ready to descend further down the exit stairs (either to the next refuge floor or to the exit discharge). Alternatively, occupants may be directed to the refuge floor and kept there awaiting further instructions.

One interesting design configuration that is frequently incorporated in to the refuge floor concept is the interruption of exit stairs. Typically, the exit stairs above the refuge floor discharge onto the refuge floor, so that stair users must leave the stair enclosure before entering the refuge floor. From the refuge floor it is possible to re-enter the stair if they desire to continue down. The interruption has the dual advantage of making the exiting occupants aware of the availability of the refuge floor; and the interruption or segmentation of the stairs into separate compartments, can mitigate stack effect, and thus improve the performance of stair pressurisation systems over the full height of each segment of the stairs.

Initially, it seems like a very compelling idea to create an entire floor that will function as an area of refuge within a building where a large number of building occupants can be gathered. However, upon further examination, questions arise about whether a refuge floor provides an increased or a decreased level of safety, and whether there are more effective and efficient ways to accomplish the goals that refuge floors attempt to achieve.

Design issues & challenges

The occupant load anticipated to use a given refuge area is based on good-faith design estimates or theoretical worse-case (full code calculated) population loads of the floors being served.

The actual utilisation of a refuge floor is, however, not easily estimated or predictable and is dependent on highly variable factors at the time of an emergency incident such as the fire incident location, extent of fire spread, nature of the occupants and nature of the emergency communications/alarms in the building.

Assuming that the estimated hundreds or thousands of occupants would assemble on a given refuge floor, this approach then poses a number of issues related to the comfort and safety of those occupants for the duration of the stay including :

Toilet and drinking water provisions

  • Seating facilities or standing room accommodations only
  • Emergency power for lighting and refuge floor amenities
  • Protection of floor from increasing /spreading fire effects
  • Ventilation/HVAC design and reliability for duration of the event

The latter is perhaps the most significant cost and implementation concern. The HVAC system needs to perform three functions: 1) provide adequate fresh air to occupants; 2) maintain any pressurisation necessary to keep the building occupants and the exits clear of smoke; and 3) maintain the temperature within the refuge floor at a range that does not negatively impact human survival or the ability those occupants to be mobile. This third function is of a greater concern for buildings in high temperature climates. If the HVAC system cannot perform any one of those three functions, it will become necessary for the occupants on the refuge floor, to relocate.The costs and energy needed to provide for the large population on a refuge floor for the occasionally (if not rare) event are counter to sustainable building design objectives. Other potential concerns are related to HVAC system failure. 

Challenges of life safety

Human behaviour is an area of concern and uncertainty when refuge floors are implemented. Although training and evacuation drills are important factors to encourage proper evacuation actions there are potential issues that may not simply be addressed by evacuation drills (Bukowski 2008).

Large crowd behaviour under such conditions can be difficult to predict and difficult to manage.

The provision of a refuge floor does not mean that occupants will utilise the floor as intended. Overcrowding or non-use are both potential outcomes. However, assuming a refuge floor is “appropriately” utilised, it should be recognised that the refuge floor becomes an assembly occupancy. 

The increased density poses crowd management risks and subjects a large percent of the building population to fire/smoke exposure risks in a single location. Should the refuge floor need to be evacuated there can be crowd management issues due to the large number of occupants competing for access to a limited number of evacuation routes (stairs and elevators).

Challenges of cost effectiveness

Refuge floors result in a reduction in the usable floor area and therefore the efficiency within a building. The dedicated refuge floor concept, where an entire floor or large portion of a floor is given over to the exclusive (single purpose) use as an area of refuge, is an exceptionally expensive requirement. It will result in an increase in construction cost, and will add to ownership and operations costs over the life of a building after construction is completed.

This applies to buildings that are leased or rented, (including offices, apartments and hotels) and results in a continuing loss of revenue for building owners over the life of a building. It also applies to privately owned, residential buildings. 

Of course, in high-rise construction the cost of additional floors is not linear and the cost of the next higher floor is likely to cost more than the previous floor.

There are a number of other added costs which are not estimated at this time, but collectively are significant. These differ from city to city and market to market, and therefore the extent will vary from building to building. These costs are acknowledged and itemised here as follows:

  • Added costs from additional construction time to add floors
  • Increased maintenance cost – cleaning, HVAC for normally empty space
  • Loss of efficiency/impact on leasing, usability for multi-floor tenants
  • Property Tax issues – floor area that fails to generate revenue, yet taxes are to be paid
  • Loss of zoning potential – reduced building height or area allowed in a given zoning space sustainability?

In brief, based on the information noted above, a refuge floor is not consistent with sustainability. They require a significant increase in use of material and energy resources, both during construction and operation during the life of a building, but they do not provide a commensurate increase in use or value.

Alternatives & solutions

For many buildings, the thoughtful implementation of a combination of “defend-in-place” and evacuation of floors in a selective and strategic sequence would provide the same, or better level of life safety, and at a less financial cost than a refuge floor system. This approach has been effective for many of the world’s tall buildings. There are always methods that can be implemented to improve that approach, and continued effort to do so, should be encouraged. Some key ideas to consider in lieu of the dedicated refuge floor or to minimise the impact of the concept are as follows:

  • Most, if not all floors, can serve as refuge space.
  • Refuge floors can double as useable occupancy.
  • Buildings with bridge connections need not rely on refuge floors.
  • Cost savings of eliminated refuge floors can be better spent on highly-reliable fire safety systems and “defend-in-place” strategies.
  • Property tax and zoning regulation relief for buildings using a refuge floors concept

Most, if not all, floors of a building can effectively be viewed as a refuge floor. 

Given appropriate levels of floor-to floor compartmentation, highly reliable sprinkler/water supply systems, and effective facilities for prompt and capable fire department response it is believed large refuge areas or floors can be effectively eliminated. If refuge floors are mandatory there are several potential opportunities for mitigating the financial impact of refuge floors. One of the most significant changes would be to permit refuge floors to have uses during non-emergency times. The zoning regulations could be modified to “recapture” floor area lost to refuge floors; tax regulations could be modified to eliminate costs for the areas of refuge floors. Such solutions pose both technical and political complications.


In conclusion, the opportunities for mitigation of the negative financial impact of refuge floors seem minimal compared to the costs of construction and ongoing costs of operation and maintenance. And so the question remains, is there a more cost effective, sustainable method of accomplishing what refuge floors are endeavoring to accomplish?

The CTBUH Fire Safety Working Group believes the concept of a refuge floor needs to be reconsidered. It may make sense for some areas of buildings of exceptional height or special occupancy conditions, but should be used only with caution, and with implementation of strategies or design elements which eliminate the design and life safety problems noted above.

Refuge floors may be an appropriate solution for special or unique situations, and should not be a strategy of first choice without careful consideration. The CTBUH Fire Safety Working Group continues to explore the subject with the intent of delivering a full and complete treatise on this subject at some stage.

The original edition of this article appeared in the Proceedings of the CTBUH 2012 9th World Congress, Shanghai.

CTBUH Research Paper : “Considerations and Challenges for Refuge Areas in Tall Buildings”

Authors : Daniel J O’Connor, Chief Technical officer, Aon Fire Protection Engineering Corporation

Ervin Cui, Leader, Aon Fire Protection Engineering Corporation

Kim Clawson, Project Manager, Aon Fire Protection Engineering Corporation