Journal Design Engineering Masthead
African Civil Engineering Journal | 24 July 2014

Methodology for Geotechnical Foundation Design on Expansive Soils in the Angolan Context

I, s, a, b, e, l, a, d, o, s, S, a, n, t, o, s, ,, M, a, r, i, a, C, h, i, v, a, l, e, ,, C, a, r, l, o, s, K, i, a, l, a
Expansive soilsFoundation designAngolan geologyReliability-based design
Presents a three-phase methodology integrating site characterisation, swell prediction, and risk-based design.
Calibrated heave model reduces prediction coefficient of variation to 0.22 using local Angolan data.
Framework explicitly accounts for parameter uncertainty to balance safety and cost-effectiveness.
Case study in Huíla Province demonstrates practical utility for raft and pier foundations.

Abstract

{ "background": "Expansive soils present a significant geotechnical challenge for foundation design in many regions, including Angola, where seasonal moisture variations cause substantial volume change. Current design codes often lack specific guidance for the unique soil mineralogy and climatic conditions found in the country, leading to either overly conservative or inadequate designs.", "purpose and objectives": "This article presents a novel, context-specific methodology for the geotechnical design of shallow foundations on expansive soils in Angola. The primary objective is to provide a systematic procedure that integrates site characterisation, swell prediction, and risk-based design to improve reliability and cost-effectiveness.", "methodology": "The proposed methodology is structured in three phases. First, a site investigation protocol is defined to determine the soil's swell potential and active zone depth using a combination of oedometer tests and soil suction measurements. Second, a predictive model for free field heave, $\\hat{\\deltah} = \\sum{i=1}^{n} \\left( \\frac{C{s(i)}}{1+e{0(i)}} \\cdot H{(i)} \\cdot \\log \\frac{\\sigma'{f(i)}}{\\sigma'_{0(i)}} \\right)$, is calibrated with local data. Third, a reliability-based design framework is implemented, incorporating the predicted heave and its uncertainty into a limit state design approach for raft and pier foundations.", "findings": "The application of the methodology to a case study in Huíla Province demonstrated its practical utility. The calibrated heave model reduced the coefficient of variation for predicted displacement to 0.22, indicating improved precision. A key finding was that the design approach, which explicitly accounts for parameter uncertainty, yielded foundation solutions with an estimated 15-25% reduction in material costs compared to conventional empirical methods while maintaining a target reliability index (β) of 3.0.", "conclusion": "The developed methodology offers a rational and systematic framework for foundation design on expansive soils, directly addressing the geotechnical conditions prevalent in Angola