A warehouse project south of Durban presented a range of geotechnical challenges due to the site's complex and highly variable ground conditions. The subsoil beneath the area consists of a mix of sands, silty sands, intercalated black soft clays (also known as "hippo muds"). Hippo muds are notorious for causing settlement related problems, and as such a robust foundation solution to ensure the long-term performance of this project.
Geotechnical challenge
The geological profile of the site features a surface layer of granular fill, underlain by stiff silty clay extending to depths of few metres below ground level. Beneath this layer lies a subsoil consisting of estuarine silty sands and clays, interspersed with pockets of highly compressible, very loose to very soft materials, which are randomly distributed throughout the profile.
This significant variation of the soil profile, both in plan and elevation, combined with the presence of highly compressible silty clay, presents significant challenge for the foundation design of the proposed structure. Significant settlement risk is expected if no improvement was carried out below the structure. This risk is exacerbated by the variability of the ground profile and the time dependency of the randomly located, compressible silty clay pockets/layers.
To better characterise the soil profile and its variability, ten cone penetration tests (CPTu) were conducted, in addition to the ten borehole results available from the original ground investigation.
Proposed engineering solution
In response to these challenges, rigid inclusions (RIs) were selected as the most suitable solution. This method, although still relatively novel in South Africa, has been used around the world as a cost-effective ground improvement method for large footprint loaded structures such as warehouses, embankments, storage tanks and reservoirs.
Rigid inclusions ground improvement involves installation of concrete columns (typically diameter 300 – 600 mm) in a grid format (spacing typically 1 – 3 m). These concrete columns are installed through the soft/compressible soil layers to found on rock or competent layer.
The system also requires a load transfer platform (LTP) constructed on top of the RIs to transfer and distribute the load to the RIs, similar to the function of a pile cap. The load transfer platform is typically a compacted granular layer which can be replaced or minimized by the addition of stone columns above the RI.
For this project, rigid inclusions are installed using a displacement technique where a steel tube is vibrated into the ground using a ring vibrator, bypassing weak or soft soil layers that may not be immediately apparent during geotechnical investigations. Once the tube reaches refusal, concrete is pumped into the tube, and the tube is then extracted, leaving behind a concrete column. To complete the RI, a short stone column is constructed in the wet concrete column.
The use of displacement method allows RIs depths to be adapted to the ground variability, ensuring all columns are founded on competent soil layers. The columns varied in length between 16m to 30m, and these excessive depths could only be achieved using purpose-built equipment.
Both RIs and piles were technically viable solutions to satisfy the project requirements. It should, however, be borne in mind that factors such as program, cost, constructability and geotechnical risk be evaluated. The design and construction of floors and foundations are generally simplified with ground improvement solutions and therefore leads to overall cost and program savings. RIs are installed from the working platform level and the completed platform is flat and clean. Unlike piled solutions, there is no reinforcement protruding above platform level, nor is trimming of piles required. In addition, the only excavation required after completion of the filling works is excavation for footings. This simplifies the floor construction significantly, which results in optimised program and costs for this project.
As the construction industry continues to address its role in global carbon emissions, sustainability remains a central focus for Franki Africa. In alignment with this commitment, the carbon footprint of this project was calculated using the EFFC/DFI carbon calculator, which estimates CO2 emissions per square metres of ground improvement. Total CO2 emission for conventional piling solution (including ground beam and slab) was calculated at 668 CO2e/m², while RI ground improvement solution (including LTP and floor) resulted in 541 CO2e/m². This translates to a 20% reduction in the carbon footprint for the foundation scope of the project.
Conclusion
Although rigid inclusions remain a relatively new ground improvement solution in South Africa, Franki Africa’s successful application of this technique on this project demonstrates the growing acceptance of the solution in the local construction and geotechnical industry. This project highlights the potential for wider adoption of rigid inclusions as a reliable, cost-effective and sustainable ground improvement solution.
