As in many parts of the world, Australia has a two-prong approach to the fire safety design of buildings. The first is a traditional prescriptive approach and the other is an engineered performance based approach. In both cases the National Construction Code’s Building Code of Australia (BCA) governs building design. Therefore the BCA specifies the rules of engagement in our performance-based design environment.
The performance-based environment is based on Objectives and Functional Statements leading to Performance Requirements. These requirements are specified by the Australian Building Codes Board (ABCB) and generally adopted by each state and territory. These are not just for fire but also for all aspects of building design including, for example, structural elements. A typical Objective is “safeguard people from illness or injury due to a fire in a building” and the associated Functional Statement is “allow occupants time to evacuate safely”.
There are two ways of achieving this, one of which is the traditional prescriptive approach called “Deemed-to-Satisfy” (DtS). By definition, meeting the requirements of this approach meets the Objective. On a side note, there is no guarantee that the DtS approach will actually meet the objective, just that it is assumed to do so and so codified by legislation. The DtS approach is often based on a historic solution, which over the years appears to have been a successful methodology for achieving society’s expectations. However, it may fall apart as new materials of construction are introduced along with new design approaches and applications.
The other way of meeting the Performance Requirements is through the use of an Alternative Solution (AS). The AS may be shown to be adequate simply through an equivalency evaluation to the DtS; in which case it has the same safety limitations, advantages and shortfalls of the DtS approach. Alternatively the AS may be shown to meet the Performance Requirements through engineering analysis, as would be found in a Verification Method, Expert Judgement or using appropriate Documentary evidence.
Usually, the design process follows the International Fire Engineering Guidelines that culminates in the production of a Fire Engineering Report (FER) that justifies and documents the AS. Once the building is built the systems are usually checked through a commissioning process.
This paper discusses a case study where a building was built incorporating a complex AS into the egress design, yet the system was never properly implemented. Evaluating the existing design and programming the building control systems based on computer simulation was the highlight of this project.
Case study – Improving egress from an existing high rise building
A thirteen-storey high-rise building in Melbourne was designed in the early 2000’s using an Alternative Solution. The BCA’s DtS provisions specified that two pressurized, fire isolated stairs were required. The FER provided justification for the use of a single non-pressurised stair to serve the upper levels, with the lifts being available for evacuation from the ‘fire floor’ (the floor of fire origin, which in practice meant the floor where the fire was first detected by one of the building’s fire detection systems). The FER required additional active fire safety systems beyond that required by DtS. These included two separate sprinkler zones on each floor (to distinguish between fire in an apartment versus one in the common corridor), addressable smoke detectors throughout the building including apartments, and corridor exhaust and pressurization systems.
The FER outlined a strategy whereby on fire detection (unless sprinklers in a common area activate), both lifts go to the ground level to let passengers out. After that, one of the lifts (selected at random) is available for calls to the floor of fire origin, if it’s on levels nine to twelve only. The other lift was to remain at the ground level for use by the fire brigade. Levels ground to eight are provided with two stairs hence the lifts were not required to serve those levels.
The AS required the lifts to be used for evacuation (see Figure 1) from levels nine to twelve only, and only for the level of fire origin. For example, a fire on level nine would require the occupants of levels ten, eleven and twelve to use the stairs to evacuate.
Whilst carrying out some upgrade works the building’s owners found that the AS had not been implemented correctly. The main issue was that the lifts had never been programmed to provide egress during fire. This needed to be urgently rectified to provide a second means of escape from the high-rise parts of the building (the high-rise parts of the building started at level nine where the height exceeded twenty five metres).
Due to the complexity of the system the residents did not understand what evacuation options were available to them during a fire. The AS did not define any methodology of notifying occupants which level the fire had actually been detected on, or which system to use (stairs, lift or both).
Therefore, residents might wait in the lift lobby for a long time before realizing that the lifts were not available for egress and that they should either retreat back to their apartment or use the single fire isolated stair. In practice, the confusion and lack of training and information meant that all able-bodied residents would try to evacuate via the single stair.
Because the lifts were not used during evacuation this placed more reliance on the single non-pressurised stair. This increased the time for evacuation, and ultimately the risk to occupants.
The owners’ main objective was to make use of the existing systems as far as possible, whilst simplifying the evacuation scheme and meeting the performance requirements of the BCA.
We evaluated the situation and proposed to alter the lift operation to allow occupants on any of the high-rise levels (levels nine to twelve) to use the lifts during fire regardless of the floor of fire origin.
To achieve these goals we carried out evacuation modeling using the software package Pathfinder (see Figure 2) to benchmark how long occupants would take to evacuate if all requirements of the original AS had been implemented correctly. We then modelled the evacuation time for the new evacuation scheme. The new scheme had the following main features:
- Lifts are available to all floors served by the single stair during any fire scenario, except for a confirmed fire in a lift lobby. This is instead of the lift only serving the level of fire origin. This reduces confusion to occupants yet had minimal impact on the evacuation time from the fire floor and actually decreased the overall evacuation time for the building.
- Evacuation signals cascade to two levels above the floor of fire origin and one below, at thirty-second intervals. This is in line with industry practice that had been implemented in the original construction. The original FER required a single 120-second delay between the first floor being notified and occupants on all other floors being alerted.
The results of the egress modelling is summarized below.
Interestingly, although any occupant on the upper four levels could now use the evacuation lift, the pedestrian modelling showed that the evacuation time was not increased from that which was predicted based on the original scheme. This was because the cascading alarm system effectively provides occupants of the floor of fire origin the ability to wake up first, start moving and get ‘first dibs’ on using the lift to evacuate.
The end result is a building that meets the objective of the BCA with minimal disruption to its occupants.
In Australia the use of a performance-based option often allows innovative and flexible design. Modern analytic technics using computer simulation of fire and egress often play a key part in the performance-based design world. This paper discussed how the use of a modern egress modeling tool justified deviations from the original design allowing continued occupation of an existing building with no disruption and little expense.
For further information, go to www.redfireengineers.com.au