Why Temporary Working Platforms Need Real Engineering

A temporary working platform holds up some of the heaviest, most safety-critical plant on a construction site, such as cranes, piling rigs and large excavators. When a platform fails, the rig or plant can topple, putting the operator and everyone nearby at serious risk and causing significant property damage. Yet of all the engineered elements on a site, working platforms often get the least engineering attention. It's a global problem, serious enough that it's the subject of the EFFC/DFI Guide to Working Platforms (Figure 1), the international best-practice document Cook Costello contributed to.

Figure 1: The EFFC DFI Guide to Working Platforms: 2^nd Edition (May 2025). 

Many failures come down to the same issue: a platform that doesn't have the strength and stiffness it needs, built and signed off on assumptions rather than measurements. Industry practice commonly relies on assumed stiffness (E) and friction (φ) parameters rather than measured ones, and the verification methods in common use don't always suit the material being tested. The result is platforms that are sometimes over-conservative, sometimes inconsistent, and sometimes unsafe.

Limitations of the Current Approach

Part of the problem is that the wrong test is often used for the material. The common in-situ tests each have a maximum particle size they suit, and coarse granular platform fills frequently fall outside that range:

• Dynamic Cone Penetrometer (DCP): suitable up to a maximum particle size of about 6 mm.

• Nuclear Density Meter (NDM): suitable up to about 40 mm.

• Plate Load Test (PLT): suitable for 75 to 200 mm material, using a 300 to 762 mm plate.

For many of the coarse gravels used in New Zealand platforms, the DCP and the Scala penetrometer simply aren't appropriate, yet they're still used. Testing a gravel platform with a tool meant for fine material gives a number, but not a reliable one.

Figure 2: The Scala penetrometer should

not be used for testing gravel platforms.

Measuring instead of assuming

The Plate Load Test offers a direct, practical way to measure how a platform actually performs in place, bridging the gap between laboratory parameters and real-world behaviour. Interpreted to the DIN 18134 standard, it gives a first-load modulus (Ev1) and a reload modulus (Ev2), and the ratio between them (Ev2/Ev1), which together characterise both the stiffness and the state of compaction of granular fill. These are the parameters that provide a direct indication of whether the platform is likely to perform as intended.

Figure 3: Example PLT Test results showing load, unload, reload cycle. 

To understand how this works for local materials, Cook Costello analysed PLT data from gravel platforms across New Zealand: 41 tests at seven sites, each platform built under controlled conditions and compacted to a density ratio (Dpr) of 95%, with test pressures up to 1,400 kPa applied using a modified DIN 18134 procedure. We derived Ev1 and Ev2 to assess stiffness and compaction, and estimated the mobilised and ultimate friction angle (φ) using the Corke et al. method.

The most important finding was that there was no clear correlation between stiffness (Ev1) and friction angle (φ). In other words, a platform's strength and its stiffness have to be measured independently, you can't reliably infer one from the other, which is exactly the assumption a lot of current practice quietly makes.

Figure 4: Results from 41 Plate Load Test results, showing the lack of correlation between strength and stiffness of the gravel platforms. 

Acceptance criteria for New Zealand platforms

From this data, we proposed a practical New Zealand acceptance threshold for well-compacted gravel platforms (Dpr ≥ 95%, the traditional NZ target):

• A reload modulus Ev2 ≥ 80 MPa, and

• A modular ratio Ev2/Ev1 ≤ 2.8

The data confirms these as broadly suitable for New Zealand gravel platforms, while reinforcing the central point: both the modulus (E) and the friction angle (φ) should be verified by measurement, not assumed. A platform can look adequate on one parameter and fail on the other.

Figure 5: Two PLT results showing the estimated friction angle for a very loose and a very dense material. The strength does not capture the variable stiffness and likely failure mechanism of the platform.

Why it matters

Moving from assumed to measured parameters changes the reliability of the whole exercise. Plate Load Testing provides a defensible and repeatable way to verify gravel platform performance. It supports safer crane and piling operations because the ground response is measured rather than assumed, and it reduces the risk of failure, delay and insurance exposure.

Temporary platforms may be temporary, but the plant they carry, and the people working around them, are not something to verify on assumptions.

Cook Costello has specialist national expertise in stiffness-based assessment of compacted fill, temporary working platforms and earthworks quality assurance. Our team contributed to the EFFC/DFI Guide to Working Platforms, 2nd Edition, and we use Plate Load Testing and related methods to help designers, contractors and asset owners verify platform performance using measured evidence.

If you are designing or verifying a working platform for crane, piling or heavy plant operations, we can help you assess it on evidence rather than assumption. The full technical method is set out in our NZGS Symposium paper, which you can download below, and you can read the EFFC/DFI guide here

Using the Plate Load Test to Estimate Soil Modulus and Friction Angle for Temporary Gravel.

Download • 819KB

Read the EFFC/DFI Guide to Working Platforms