Performance Solutions by the Numbers
Chapter 8 of 10 · 10 min read
Evidence for the products and systems a solution relies on
Match every product and system in a Performance Solution to a form of evidence the NCC accepts, check that the evidence covers the installed configuration, and evidence the conditions too.
A Performance Solution that depends on a detection system is only as good as the evidence that the system will do what the analysis assumed. Chapter 7 showed that the running case’s solution stands or falls on detection in scenario A. This chapter asks the next question: what proves that the detection, the alarm and the signs will perform, in this building, as installed?
Part A5 of NCC 2022 Volume One answers in general terms. Applying it to a real system means following an evidence chain from product to design to installation to commissioning, and checking at each link that the evidence covers the configuration actually built. It also means finding a way to evidence things that are not products at all, such as a restriction on what may be stored on a mezzanine.
What Part A5 requires
A5G1 requires materials, products, forms of construction and designs to be fit for their intended purpose, which means supported by evidence under A5G2 and A5G3 and properly constructed or installed. A5G2 adds two general rules: the form of evidence must be appropriate to the use of the thing it relates to, and any copy of documentary evidence must be a complete copy of the original.
A5G3(1) lists the forms of evidence that can show a Performance Requirement or a DTS Provision is met: a current CodeMark certificate, a current Certificate of Accreditation, a current certificate from a certification body, a report from an Accredited Testing Laboratory, a certificate or report from a professional engineer or other appropriately qualified person, or another form of documentary evidence, such as a Product Technical Statement. Each must set out what it covers and the basis for its conclusion.
Two definitions matter here. An Accredited Testing Laboratory is, broadly, one accredited by NATA for the tests concerned, or a recognised equivalent. A professional engineer is, broadly, a registered engineer in the relevant discipline where legislation provides for registration, or otherwise a person registered on Engineers Australia’s National Engineering Register or eligible for it, with experience and competence in the field. A report from someone outside those definitions may still be useful information, but it is not the kind of evidence the definitions describe.
The evidence chain for a detection system
A detection and alarm system is not a single product. It is a set of products, designed into a layout, installed, connected and commissioned. Evidence must cover each link.
The running case’s design called for smoke detection on the mezzanine and in the space beneath it, with an alarm sounder on the mezzanine that operates when any of those detectors does. The fire engineer specified the outcome; the fire services designer specified the system to AS 1670.1, the installation standard for fire detection and alarm systems as called up by the NCC. This book names that standard but does not quote it.
Matching evidence to configuration
Evidence belongs to a configuration. A detector tested and certified for flat ceilings tells you about flat ceilings. The space under a mezzanine has beams that can trap smoke in pockets, and the roof above the mezzanine is pitched and up to 7.5 m high. Whether detectors in those positions will respond as the analysis assumes is a design question that the standard and the designer must answer, not something a product certificate settles.
The fire services designer therefore recorded, for each location, the standard’s provisions relied on and the reason the layout met them: detector spacing under the mezzanine deck allowing for its beams, and the position of detectors relative to the roof pitch above the mezzanine. That record is design evidence. It is the link most often missing in small projects, where a contractor installs “some detectors” and provides a certificate that the installation was to the standard, without a design that shows how.
The error: a wireless kit and the wrong panel
The electrical contractor proposed connecting the new mezzanine detectors to the building’s existing fire indicator panel with a wireless detector kit, to avoid running cable across the workshop. The kit’s supplier provided a certificate of conformity and a compatibility list.
The fire services designer checked the list against the panel’s model and firmware. The panel installed in the building was an older model that was not on the list. The kit’s evidence therefore did not cover the configuration proposed: it covered the kit with other panels. The contractor’s view was that “it uses the same protocol”, which is a reasonable engineering hypothesis and not evidence.
The resolution was to use a wired loop extension that was listed as compatible with the existing panel. The cost was a day of cabling. Had the wireless kit been installed, the mezzanine detectors might have worked, but nobody could have shown that they would, and the Performance Solution’s central assumption would have rested on a guess. The error was detected because the designer compared the compatibility evidence with the actual equipment, model by model.
The evidence register for the solution
This is an illustration. The fire engineer kept a register for every item the Performance Solution relied on, separate from the building’s general evidence register, so that the certifier could see the solution’s dependencies in one place.
Reading a test report
Test reports are the most common evidence in industrial fit-outs, and they are often read only for their conclusion. A short routine catches most problems:
- The laboratory. Is it accredited for this test? NATA accreditation has a defined scope, and a laboratory accredited for some tests may not be accredited for the one in the report.
- The standard and its edition. Is the test method the one the NCC calls up, at the edition it calls up?
- The specimen. What exactly was tested: sizes, materials, fixings, orientation, supporting construction? This is where most mismatches hide.
- The results and their limits. What did the specimen achieve, and does the report state any limits on how the result may be applied?
- The date and completeness. Is it the whole report, as A5G2 requires, or an extract or summary? Is it still current?
A two-page summary from a supplier’s website is not a complete copy of a test report, even if it carries the laboratory’s logo. Ask for the full report, and file it with the register so that a later reviewer can read what you read.
Product Technical Statements and marketing material
A5G3(1)(f) allows other documentary evidence, such as a Product Technical Statement, provided it demonstrates compliance and sets out the basis for it. The phrase “sets out the basis” is what separates evidence from marketing. A statement that a product “complies with the NCC” without saying which provision, by what means and under what conditions does not set out a basis. A statement that names the provision, the test or assessment relied on, the configuration covered and any limitations does.
In the running case, a supplier of photoluminescent direction signs provided a brochure stating that its signs were “NCC compliant”. The designer asked for the documents showing compliance with Specification 25, which governs photoluminescent exit signs, and selected the product only once those were provided. The brochure went in the file, but not in the evidence register.
Evidence for the analysis itself
The products are not the only things that need evidence. The comparative analysis, the time-line calculation and the Expert Judgement are themselves claims that the Performance Requirements are met. Under A5G3(1)(e), a certificate or report from a professional engineer or other appropriately qualified person, certifying compliance and setting out its basis, is one form of evidence. The final report in Chapter 10 performs that role for the running case. That is another reason the report must set out its basis fully: the report is not only a description of the solution, it is evidence of it.
Evidencing conditions that are not products
Some parts of the solution are not products. The 3 m clear zone at the stair foot, the restriction to non-combustible storage on the mezzanine, the occupant limit and the impairment procedure are all conditions on how the building is used. Part A5’s evidence forms do not fit them, because there is no test report for “keeping an area clear”.
The running case handled them with three kinds of record:
- Physical evidence. Floor markings for the clear zone, signs at the stair top stating the storage restriction and the occupant limit, and a sign position for the impairment notice.
- Management evidence. A written procedure for the impairment of the detection system, a storage rule in the stores procedure, and training records for stores staff.
- Inspection evidence. A monthly check by the operations manager, recorded on a one-page form, covering the clear zone, the storage on the mezzanine and the condition of the signs.
The certifier’s role with these is limited. They can check that the conditions are written clearly in the final report and that the physical measures are in place at completion. They cannot supervise the business afterwards. That is why Chapter 10 treats the conditions as obligations the owner must carry, and why the engineer should write only conditions the owner can realistically keep.
Evidence over time
Evidence is dated, and some of it expires. A5G3(1) refers to “current” certificates in several of its items. A certificate that was current when the system was installed may lapse later, and products may be replaced with different models during maintenance.
For the running case, the evidence register includes the expiry or review date of each certificate and a rule for replacements: any replacement detector, sounder or panel component must appear on the same compatibility evidence, or the fire services designer must review the change. That rule is in the maintenance contract. Without it, a routine replacement could break the evidence chain years after approval, with nobody noticing until an incident.
How much evidence is proportionate?
It is possible to demand more evidence than a small solution needs. The running case did not require full-scale fire tests of the mezzanine or smoke modelling of the detector response. It relied on certified products, a standard design, a competent installer, commissioning tests and clearly written conditions. The test of proportion is whether each assumption in the analysis has evidence of a kind the NCC accepts, at a depth matched to how much the conclusion depends on it. Detection carries the whole of scenario A, so its evidence chain was checked link by link. The direction signs carry a smaller part of the argument and were checked more lightly.
The opposite failure, too little evidence, is more common on small projects. Owners and contractors sometimes treat evidence as paperwork to be gathered at the end, after the products are installed. By then, a mismatch such as the wireless kit can only be fixed by removing work. Assembling the evidence register at design stage, as the running case did, turns each mismatch into a purchasing decision rather than a defect.
Exercise: a substitution during maintenance
Two years after completion, the maintenance contractor replaces a failed smoke detector under the mezzanine with a different brand that was in stock. The replacement is certified as a smoke detector. Is the evidence chain intact?
Worked answer. Not necessarily. The replacement’s certification shows that the device is a suitable smoke detector, which covers the first link. It does not show that the device is compatible with the building’s panel, which is the second link, unless it appears on the panel’s compatibility evidence or the replacement detector’s own documented compatibility includes that panel. The layout and commissioning links also need checking: if the replacement has a different sensitivity or mounting requirement, the design assumption and the commissioning test may no longer hold.
The owner should require the contractor to show compatibility evidence for the replacement, test it in place as part of the system, and record the change in the evidence register. If compatibility cannot be shown, the device should be replaced with a compatible one, and while that is arranged the detection should be treated as impaired, which under the impairment procedure in Chapter 7 means closing the mezzanine. A routine maintenance call is exactly where evidence chains quietly break. The cure is not more paperwork but a habit: every change to a system that a Performance Solution relies on is treated as a change to the solution until someone competent has shown otherwise, and the result of that check is written into the register with the date and the name of the person who made it.
