In our modern society, we expect safety in the things we use, the activities we do, the places we go, and the buildings we occupy.

This is perfectly reasonable; in fact, we should not expect anything less than safety.

But what level of safety should we expect? No deaths, no injuries, no harm – 100 per cent safe?

Most people do not often take time to probe deep into these philosophical questions. But when we set policies and develop building regulations, these are questions that we must ask and answer.

In Australia, as in other countries, we are required by our regulations to design buildings to be fire safe. The objective is stated in our Building Code: “to safeguard people from illness and injury." But what does it mean? Must we design buildings such that absolutely no harm is to come to any occupant in building fires?

Irrespective of how you might wish to answer the question, from a design viewpoint, the answer must sadly be no. This is not because we wish not do so, but because it is impossible to do so. Absolute safety is an unreachable target.

To ensure as much would require limitless knowledge and infinite foresight of what might go wrong. This includes the performance of the buildings, their systems, the behaviours and activities of their occupants...the list goes on. The most significant factor governing fire safety is occupant behaviour which, ironically, is beyond the control of building design; and sometimes makes safety design somewhat oxymoronic.

This does not mean that we cannot design safe buildings; just that safety is relative rather than absolute. We cannot predict all possible events that could happen. But we can design a building to have greater propensity to deal with fire events to offer a safer environment. Unfortunately, accidents - along with their unpredictable consequences - can still happen. That is Murphy’s Law.

You might argue that we just have to crank the design lever harder to control everything and all activities. After all, various safety professions have proclaimed that “all accidents can be prevented.” If that were the case, it follows that all injuries can likewise be avoided, and abracadabra, safety is absolute.

This sounds all well and good. Unfortunately, the proclamation mirrors more of an aspiration rather than a practical possibility. It serves as a useful and fruitful safety campaign slogan. It spurs us on to take greater care in things we do to prevent careless mistakes. But on the flip side, it lulls us into a false belief that we can reach an impossible goal. It allows the delusion to seep in our psyche that we can fully control everything around us, that we are the masters of the universe, we can reach the stars, nothing is impossible, including absolute safety.

But when we return to earth from the aspirational heights, we have no choice but to accept the reality that such a goal is impossible. This is not a cop-out nor a concession to defeat, but rather a pragmatic outlook; we must strive for a practical and achievable safety level that is acceptable by us, as a society.

The next logical question is: as a society, how safe should we expect our buildings to be? What level of risk should we accept for our buildings?

Our regulators have been struggling and shying away from setting an acceptable level. They are probably daunted by the unpleasant reality that acceptance of a safety level would necessitate the acceptance of the corresponding level of possible harm, which counters the former like a balance scale – and one can’t exist without the other. This, though logical, is politically unpalatable; and the reality is harsh and unwelcome.

But we must be reasonable. We can’t design for absolute safety. Neither can we, as a society, afford to pursue unrealistically high targets. It is an ethical, emotional and economic decision to set an acceptable level. Design that ensures buildings are safe is an investment, and an initial small investment in design will provide a high return. But the law of diminishing returns would dictate that increasing investment would eventually yield smaller and smaller returns, and ultimately tends toward the unreachable asymptotic limit of absolute safety.

A logical acceptable level of safety would be one that is no worse than the high level we enjoy now, and that provides the best bang for our societal investment in terms of making buildings safe.

Many studies have been conducted in the area of fire safety engineering to investigate cost-effectiveness of safe design, performance-based design, quantification of a safety level, societal acceptance levels and tolerable risk levels. We now have the tools and the knowledge. What we need is to explicitly set a quantifiable safety level in our regulations so that we may design our buildings to attain a uniform, transparent, high level of safety. This is what we should absolutely expect – not absolute safety.

  • Surely occupants and owners of units in the Lacrosse Tower and other like buildings including Melbourne's women hospital have a rightful expectation that the cladding used on the exterior would not be highly flammable to the extent the CSIRO testing of that cladding had to be suspended because of danger.. The Victorian Fire Authority had declared that type of fire as one they had not witnessed before in its speed of spread and intensity.

  • Good case as regards new buildings Dr Weng , we do need to be reasonable or costs can blow out (except where specifications are inadequately checked and grossly inferior alternatives are permitted as per Lacrosse Tower and 500? similar buildings recently).

    But have we ever been reasonable as regards per-existing buildings as regards smoke alarms (mandatory since 1999 in new buildings only) and safety switches (1991)?

    Surely saving the lives of (say) a few back-packers in someone else's building, or two tenants in someone else's house go a long way to making it "reasonable (to require these measures)" in buildings built prior to these mandatory requirements for new buildings? I think you'll agree that older buildings are grossly under-regulated compared with current safety measures for new buildings.

    Another sadly neglected area is glazing, where in 1988, new homes were made safer by regulation, but older homes were neglected and are still quite dangerous. What is it with governments and building authorities? Do they really care for Australians?

  • The development of the Building Code of Australia in the late 80s incorporated the "deemed to satisfy DtS) concept of building fire safety design with the criteria drawn from existing AUBRCC and Warren Centre work with long held design features (of this length, that thickness etc) automatically adopted into the DtS regime.
    How they were "born" has been lost in the sands of time but clearly design criteria from the early 20th century should have had no place in modern Codes.
    Any fire engineering assessment (and I have done many) of a DtS design will certainly show that it will not meet the required performance criteria. As an example, the DtS separation distance between buildings is laughable, daily news footage of Class 1 residential building fires shows flame emergence from windows well in excess of the 900mm to the boundary, whilst the 3m separation between openings in external walls (requiring an FRL) of Classes 2 to 9 buildings to boundaries (ergo 6m to the adjoining lot building opening) is well short of fire propagation from modern thermoplastic furnishings with resultant fire spread.
    Similarly, DtS egress travel distances and assumed occupant friendly fire isolated stairs cannot always be used. Disabilities such as wearing of multi-focal spectacles which impair downward vision, and therefore stair use, are ignored.
    Over the years, the BCA has increased reliance on Australian Standards design criteria such that few building services designers understand compliance; e.g. AS2293.1 requires emergency lighting design to take into account the presence of smoke in the building, and yet every design has the luminaries at ceiling height – so much for safe egress!

  • The impossibility of absolute safety is a good point. I explain it as "all accidents may be preventable, but not all accidents can be prevented". Given this, we as engineers, designers, operators etc must do everything *reasonable* to prevent accidents, knowing that we can never stop them all.

    Quantified risk tolerability criteria is one approach used in an attempt to demonstrate that all reasonable steps have been taken with regard to safety. Another approach is to focus on the available precautions, and in particular recognised good practice.

    May I suggest you read the Sourceable article linked below? (I am a co-author.) In it we discuss the history of theses two approaches, and their strengths and weaknesses in meeting our legal duties to ensure safety as designers, engineers, operators etc. We find that a precaution-based, rather than risk-based, approach provides for more rigour and clarity in safety-related decision-making, and that this view is supported by both OHS/WHS legislation and the courts.